AI News Feed

NEW DELHI, Feb 17: Sovereign AI models have been made available at the AI Impact Summit and the government is planning huge infrastructure to make them available widely for everyone, Union minister Ashwini Vaishnaw said on Tuesday.

The minister said sovereign models have withstood all parameters at par with global big AI models.

"Sovereign models have been launched here. You can go and see and see how they stand on the global parameters. In the coming time, AI mission 2.0, there will be a need to prepare a large infrastructure for sovereign models. This will bring India's availability to a new level," Vaishnaw said.

Twelve (12) organisations and consortia, including startups, industry players and academic institutions, including Sarvam AI, Soket AI, Gnani AI, Gan AI, Avatar AI, IIT Bombay Consortium (BharatGen), GenLoop, Zentieq, Intellihealth, Shodh AI, Fractal Analytics Ltd, and Tech Mahindra Maker's Lab, have been selected for developing large and small language models based on Indian datasets.

Large AI models have been commercially rolled out by global players like ChatGPT of OpenAI, Google's Gemini, etc.

The minister said investors are also bullish on India and have finalised investment of over Rs 20,000 crore, which will materialise over the next 1-2 years.

On people complaining about long queues and other logistics issue, the minister said there has been a huge response to the summit and registrations crossed 3 lakh already.

"There is a very strong response. More than 3 lakh people have registered. All sessions are overflowing. There is a response about AI Summit among all the people. There is a lot of excitement," Vaishnaw added. (Agencies)

The GCC region sets global standards, with its Public Sector achieving the highest AI maturity level across surveyed markets globally, while Tech, Media and Telecom (TMT) rank first within the GCC.

Doha, Qatar - Qatar is demonstrating significant momentum in artificial intelligence advancement within the GCC, recording a remarkable AI maturation shift according to a comprehensive new study by Boston Consulting Group. The report, "Unlocking Potential: How GCC Organizations Can Convert AI Momentum into Value at Scale," reveals that Qatar has achieved a notable 10 percentage-point increase in 'Emerging' AI organizations between 2024 and 2025, positioning the nation firmly within the region's AI acceleration.

The study, which surveyed 200 C-suite executives and assessed 41 digital and AI capabilities across seven industries, shows Qatar's average AI maturity score of 39 rising rapidly year-over-year, reflecting sustained organizational commitment to AI transformation. This upward trajectory indicates a growing pipeline of organizations transitioning from experimental phases toward comprehensive AI integrations.

"Qatar's rapid AI maturation and investment in advanced AI compute capabilities reflects the nation's strategic approach to technological transformation and economic diversification," said Dr. Ahmad Dhaini, Principal at Boston Consulting Group. "The 10 percentage-point surge in emerging AI organizations demonstrates Qatar's ability to accelerate from experimentation to systematic implementation at scale, creating a robust foundation for sustained AI leadership across multiple sectors."

Across the broader GCC region, the report demonstrates remarkable progress in closing the AI adoption gap with global markets. According to the report, 39% of all GCC organizations now qualify as AI Leaders, compared to the global average of 40%, representing a fundamental transformation in how regional businesses approach artificial intelligence. The GCC region demonstrates exceptional AI leadership, with its Public Sector achieving the highest AI maturity levels globally across all surveyed markets. While TMT continues to lead in AI maturity within the GCC, there is rapid advancement occurring in other critical sectors including Financial Institutions, Health Care, Industrial Goods, and Travel, Cities, and Infrastructure, highlighting the region's broad-based AI transformation.

The financial impact of AI leadership proves substantial, with AI Leaders across the GCC delivering up to 1.7 times higher total shareholder returns and 1.5 times higher EBIT margins compared to AI Laggards. This performance differential underscores the critical importance of moving beyond pilot programs toward scaled implementation. This success is directly linked to higher AI investment levels - AI Leaders are dedicating 6.2% of their IT budgets to AI in 2025 compared to only 4.2% by Laggards. As AI budgets continue to grow, the value generated by AI Leaders is expected to be 3-5x higher by 2028, not only amplifying their competitive advantage but also significantly widening the performance gap between Leaders and Laggards.

GCC AI Leaders: Pursuing AI-First Models and Unlocking Agentic AI Value

While the GCC has demonstrated advanced digital maturity in recent years, AI maturity has surged by 8 points between 2024 and 2025, now trailing overall digital maturity by just 2 points. The study reveals that successful AI Leaders distinguish themselves through five critical strategic moves: pursuing multi-year strategic ambitions with 2.5 times more leadership engagement than laggards, fundamentally reshaping business processes rather than simply deploying off-the-shelf solutions, implementing AI-first operating models with robust governance frameworks, securing and upskilling talent at 1.8 times the rate of competitors, and building fit-for-purpose technology architectures that reduce adoption challenges by 15%.

Looking toward frontier technologies, 38% of GCC organizations are already experimenting with agentic AI, positioning the region competitively against the global average of 46%. The value generated from agentic AI initiatives, currently at 17%, is projected to double to 29% by 2028, driven by continued experimentation and strategic deployment.

Despite this strong momentum, GCC organizations continue to face barriers to AI adoption, with AI Laggards 18% more likely than AI Leaders to encounter people, organization, process challenges stemming from limited cross-functional collaboration on AI, unclear AI value measurement, misalignment with enterprise strategy, or lack of leadership commitment. AI Laggards are also 17% more likely to face challenges in algorithm implementation, especially around limited access to high-quality data, and 10% more likely to encounter technology constraints, such as security risks and RAI implementation, in addition to a general constraint in the availability of local GPUs, further increasing burden on organizations.

"GCC organizations, including in Qatar, have moved beyond experimentation, with many now successfully deploying and scaling AI across core functions," added Hassen Benothman, Managing Director and ME Leader of BCG Platinion. "The real opportunity ahead is to build on this momentum to successfully reshape processes and invent new value streams with AI. As organizations look to address key organizational challenges and improve their digital platforms, data foundations, and operating models, early experimentation with agentic AI signals positive promise, with the potential to fundamentally reshape how work gets done and value is shaped in the near future."

The report emphasizes that sustained AI leadership requires continued focus on executive engagement, comprehensive talent development, responsible AI governance, and strategic alignment between AI initiatives and broader business objectives. As Qatar organizations accelerate their AI transformation journey, the rapid growth of emerging and scaling organizations is building a strong foundation to move from experimentation to scale and unlock new sources of AI-driven value.

Read the Full Report: To explore BCG's complete analysis, read the comprehensive report [here].

About Boston Consulting Group

Boston Consulting Group partners with leaders in business and society to tackle their most important challenges and capture their greatest opportunities. BCG was the pioneer in business strategy when it was founded in 1963. Today, we help clients with total transformation -- inspiring complex change, enabling organizations to grow, building competitive advantage, and driving bottom-line impact.

To succeed, organizations must blend digital and human capabilities. Our diverse, global teams bring deep industry and functional expertise and a range of perspectives to spark change. BCG delivers solutions through leading-edge management consulting along with technology and design, corporate and digital ventures -- and business purpose. We work in a uniquely collaborative model across the firm and throughout all levels of the client organization, generating results that allow our clients to thrive.

Salil Parekh, MD and CEO of Infosys, said the company is currently deriving AI-led value across multiple areas, including the use of agentic AI for legacy system modernisation, preparing enterprise data for AI models, and generating deeper business insights. He added that Infosys is also deploying AI agents to reimagine core business processes while traction is growing in physical AI, particularly in product design and embedding AI capabilities into physical devices.

When Usha Bansal and Pinki Ahirwar -- two names that exist only in a research prompt -- were presented to GPT-4 alongside a list of professions, the AI didn't hesitate. "Scientist, dentist, and financial analyst" went to Bansal. "Manual scavenger, plumber, and construction worker" were assigned to Ahirwar.

The model had no information about these "individuals" beyond the names. But it didn't need any. In India, surnames carry invisible annotations: markers of caste, community, and social hierarchy. Bansal signals Brahmin heritage. Ahirwar signals Dalit identity. And GPT-4, like the society whose data trained it, had learned what the difference implies.

A Server‑Side Request Forgery (SSRF) vulnerability has been identified in the langchain/community package, affecting versions up to 1.1.13. The flaw, tracked as CVE‑2026‑26019, has a moderate severity rating, with a CVSS 3.1 score, due on its potential to expose sensitive cloud metadata and internal infrastructure. The vulnerability originates from the RecursiveUrlLoader class, which performs recursive web crawling. By default, it restricts crawling to [...]

The Indian stock market today showed resilience, with IT driving sentiment and PSU banks adding support. Investors should keep an eye on whether Nifty sustains above 25,750 to open a path toward 26,000. Support sits at 25,500-25,550. Metal and oil weakness remains a drag on any broad-based rally. The Global Indian AI Summit could help domestic companies secure more partnerships with global tech giants, potentially shielding IT stocks from the recent global tech sell-off pressure.

Also Read: Top 5 AI Stocks to Watch in February 2026: Must‑See List

Infosys is the top stock in focus today after it announced a strategic AI partnership with Anthropic. The stock jumped over 4.6% to Rs. 1,424. HCL Technologies, Wipro, and Tech Mahindra also gained strongly. Cochin Shipyard surged more than 7% after winning a major defence contract. On the downside, metal stocks such as Hindalco and Tata Steel declined.

2. What happened in the Indian stock market today?

The Indian stock market saw a strong recovery after a weak start. Sensex fell early but later bounced back over 400 points to trade at 83,549. The Nifty also recovered to 25,730. IT stocks led the rally, while PSU banks supported gains. However, metal and oil stocks remained under pressure during the session.

3. Why is Sensex going up?

The Sensex is rising mainly because of strong buying in IT stocks. Infosys led the gains after its announcement of an AI deal. Other IT majors like HCL Tech and Wipro also moved higher. Positive market breadth and a drop in India VIX, which shows lower fear levels, helped boost investor confidence during the day.

4. How is Nifty performing?

The Nifty recovered from its day's low of 25,594 and climbed to 25,730 by midday. The index gained support from IT, FMCG, and PSU banking stocks. Analysts say 25,750 is an important level to watch. If Nifty holds above this mark, it may move towards 26,000 in the coming sessions.

5. What is the latest share market news?

The latest share market news highlights a strong IT rally led by Infosys. Cochin Shipyard won a Rs. 5,000-crore defence order, pushing its shares higher. Metal stocks declined due to weaker copper prices. Gold prices fell globally, and the rupee slipped slightly. Overall, markets showed resilience despite global and sector-specific pressures.

Others argue AI will create new jobs and transform existing roles

AI is no longer being sold as a tool that 'helps workers.' It's now being pitched, and feared, as a substitute for them.

Over the past two years, especially, CEOs, policymakers, investors and economists have moved from cautious hedging to concrete forecasts: half of entry-level jobs gone, 80 percent of work automated, hiring freezes justified by AI productivity, white-collar disruption in 12-18 months.

Some predict large-scale displacement. Some argue that history will repeat itself and jobs will adapt. Others insist the real story is transformation, not elimination.

Here is a clean, structured ledger of what leading voices have actually said.

Dario Amodei, CEO, Anthropic

Warned AI could wipe out roughly half of entry-level white-collar jobs within 1-5 years, arguing rapid deployment could sharply disrupt junior knowledge roles.

Mustafa Suleyman, CEO, Microsoft AI

Said AI could automate "most, if not all" professional tasks within 12-18 months, suggesting a compressed timeline for white-collar disruption.

Geoffrey Hinton, AI pioneer

Said 'mundane intellectual labor' could be widely replaced and described a future where one person plus AI does work that previously required many people.

Andy Jassy, CEO, Amazon

In an internal memo: "We will need fewer people doing some of the jobs... we expect that this will reduce our total corporate workforce."

Arvind Krishna, CEO, IBM

Said IBM would pause hiring in some roles; about 7,800 jobs could be replaced, and roughly 30 percent of non-customer-facing roles could be automated within five years.

Sundar Pichai, CEO, Alphabet

Said the CEO job is among the 'easier things' AI could potentially do, and acknowledged AI will eliminate some roles while transforming many others.

Jamie Dimon, CEO, JPMorgan Chase

Has said AI will change nearly every job and could eventually lead to a 3.5-day workweek due to productivity gains.

Charles Scharf, CEO, Wells Fargo

Headcount reduction in areas like compliance and customer operations is 'inevitable' with AI efficiency gains.

Jensen Huang, CEO, Nvidia

The threat isn't AI itself but displacement by AI-enabled workers: "You won't lose your job to AI, you'll lose it to someone using AI."

Mark Cuban, Entrepreneur & Investor

Said AI will replace some jobs, but: "AI will create new jobs, not kill entry-level ones." Also argues implementation demand will generate millions of new employment opportunities.

Kristalina Georgieva , Managing Director, IMF

Wrote that AI will affect nearly 40 percent of jobs globally, replacing some and complementing others. Later described AI hitting labor markets like a 'tsunami.'

Liz Kendall, UK Technology Secretary

Said artificial intelligence 'will cost jobs', particularly in graduate-level professions. Also stated more jobs will be created than lost, but warned against complacency.

Andrew Yang, Politician & UBI advocate

Warned 'millions of jobs will be lost in 12-18 months,' framing white-collar disruption as near-term and severe.

Nandan Nilekani, Chairman, Infosys

Warned AI risks public backlash if its visible impact is misuse and job losses rather than broad public benefit.

N. R. Narayana Murthy, Founder, Infosys

Routine jobs will be lost as AI spreads, productivity gains can create new opportunities if workers adapt.

C Vijayakumar, CEO, HCLTech

AI productivity can deliver 3-5 percent growth without adding staff. Also framed an ambition of 2x revenue with half the workforce, signaling headcount-light growth thinking.

Vineet Nayar, Former CEO, HCLTech; Founder-Chairman, Sampark Foundation

Indian IT firms may prioritise profitability over job creation: "Profit comes first, not jobs." Said AI could worsen India's employment stress if not handled responsibly.

K Krithivasan, CEO, TCS

Do not expect AI to trigger mass layoffs or fundamentally upend the outsourcing model.

Rajesh Nambiar, President, Nasscom

Pushed back on 'AI kills IT jobs' fears, arguing enterprise complexity and services integration will sustain demand.

Puneet Chandok, President, Microsoft India & South Asia

"AI will not kill jobs. AI will unbundle jobs. AI will reshape jobs, not eliminate them."

Amitabh Kant, Former CEO, NITI Aayog; Former G20 Sherpa

"AI will not lead to job loss but will create different kinds of jobs." Also warned AI could become an inequality trap if mishandled.

Vinod Khosla, Founder, Khosla Ventures

Within five years AI could do 80 percent of 'economically valuable' work humans do. Also argued "80 percent of all jobs can be done by an AI."

Brian Moynihan, CEO, Bank of America

AI will not eliminate all jobs, citing historical precedents where technology expanded total employment despite automation fears.

Kai-Fu Lee, AI investor and author

AI would replace large portions of routine and repetitive work globally within a decade and a half.

Bill Gates, American businessman and philanthropist

"The robot that takes your job should pay taxes." Gates argued automation could displace workers at scale and proposed taxing companies that replace humans with machines to fund retraining.

Sam Altman, CEO, OpenAI

"AI is for sure going to change a lot of jobs." Altman has repeatedly said AI will 'totally take some jobs away' while creating new ones.

Elon Musk, leadership at Tesla, SpaceX, Twitter, and xAI

"This technology will make a lot of jobs obsolete." Musk has described AI as capable of making many roles redundant and compared its labour impact to a 'supersonic tsunami.'

Marc Benioff, CEO, Salesforce

"AI is doing 30-50 percent of the work at Salesforce." Benioff described AI taking over large portions of internal workflows.

University aims to skill 20,000 students across disciplines for AI-driven roles and actively recruiting chief architects to lead the Centre of Excellence

Dayananda Sagar University, a leading private multidisciplinary university in Bengaluru, signed a Memorandum of Understanding (MoU) with NVIDIA to build India's first AI-first Factory.

The initiative will leverage NVIDIA's next-generation AI supercomputing platform, Blackwell, designed for large-scale model training and inference, enabling AI-driven innovation in education and multidisciplinary research, and supporting the development of sovereign AI capabilities from India to the world.

The objective is to address the country's most binding AI constraints: underequipped infrastructure, consumption of foreign AI models, lack of domestic Supercomputing technologies and a disconnect between academic training and industry deployment needs.

With Dayananda Sagar University's (DSU) investment exceeding ₹175 crore, the AI-first factory is intended for production-grade AI computing in an academic setting, which means there is alignment between academic research and education and production-grade AI development environments for students, researchers, and enterprises.

"The collaboration with NVIDIA marks a fundamental shift in how we, as a university, are looking to employ innovation in the field of AI," said Dr. D. Premachandra Sagar, Founder & Pro-Chancellor, Dayananda Sagar University.

"By establishing an NVIDIA-powered AI factory-class infrastructure on our campus, we are enabling students and researchers to work on real-world AI systems, while directly contributing to India's long-term goals around Aatmanirbhar Bharat, sovereign technology development, responsible AI, and future-ready talent creation."

Additionally, Dayananda Sagar University will create six industry-integrated Centres of Excellence with over ₹50 crore in capex, focusing on healthcare, engineering, defense, cybersecurity, semiconductors, smart mobility, and sustainability. These Centers will serve as hubs for specialized AI research and application, bridging academia and industry to drive innovation in key sectors.

To support this, the university is actively hiring chief architects to lead the initiative, responsible for architecting platforms, building elite teams, and converting raw compute power into deployable intelligence systems.

The university has established the DSU-Siemens Sustainability & Net Zero Collaboration, which implements Siemens technologies to develop sustainable campus areas and climate-resilient infrastructure systems as a part of this initiative.

Dayananda Sagar University is aiming to skill around 20,000 students across disciplines to prepare them for new job roles emerging with AI technologies, including AI and ML applications, generative AI and LLM development, AI platforms and MLOps, AI product architecture, and domain-specific AI applications in healthcare, cybersecurity, robotics, electronics, and smart systems.

DSU will showcase its vision for India's first AI-First Factory at the India AI Impact Summit 2026, being held in New Delhi from 16-20 Feb. The exhibit, located at Stall Nos. 6.26, 6.27, 6.32 and 6.33 at Bharat Mandapam, will be among the largest at the summit. Powered by NVIDIA, the pavilion will feature humanoid robotics demonstrations and highlight DSU's strategic collaborations with Siemens and other key partners.

We are providing an unedited version of this manuscript to give early access to its findings. Before final publication, the manuscript will undergo further editing. Please note there may be errors present which affect the content, and all legal disclaimers apply.

With the rapid development of large language models (LLMs), an increasing number of applications leverage cloud-based LLM APIs to reduce usage costs. However, since cloud-based models' parameters and gradients are agnostic, users have to manually or use heuristic algorithms to adjust prompts for intervening LLM outputs, which requiring costly optimization procedures. In-context learning (ICL) has recently emerged as a promising paradigm that enables LLMs to adapt to new tasks using examples provided within the input, eliminating the need for parameter updates. Nevertheless, the advancement of ICL is often hindered by the lack of high-quality data, which is often sensitive and different to share. Federated learning (FL) offers a potential solution by enabling collaborative training of distributed LLMs while preserving data privacy. Despite this issues, previous FL approaches that incorporate ICL have struggled with severe straggler problems and challenges associated with heterogeneous non-identically data. To address these problems, we propose an asynchronous distributed bilevel tuning (AsynDBT) algorithm that optimizes both in-context learning samples and prompt fragments based on the feedback from the LLM, thereby enhancing downstream task performance. Benefiting from its distributed architecture, AsynDBT provides privacy protection and adaptability to heterogeneous computing environments. Furthermore, we present a theoretical analysis establishing the convergence guarantees of the proposed algorithm. Extensive experiments conducted on multiple benchmark datasets demonstrate the effectiveness and efficiency of AsynDBT.

Investment targets 5 GW hyperscale capacity, partnerships with Google, Microsoft and Flipkart, and a projected USD 250 billion AI infrastructure ecosystem

The Adani Group has unveiled a USD 100 billion investment plan to build renewable-energy-powered, hyperscale AI-ready data centres by 2035, positioning the initiative among the largest integrated energy-and-compute commitments globally. The Group said the programme aims to establish a long-term sovereign energy and computing platform while strengthening India's role in the evolving artificial intelligence economy.

According to the company, the investment is expected to catalyse an additional USD 150 billion by 2035 across server manufacturing, advanced electrical infrastructure, sovereign cloud platforms and related industries. Combined, this could result in a USD 250 billion AI infrastructure ecosystem in India over the next decade.

Chairman Gautam Adani said the strategy reflects the growing convergence between energy capacity and computing power as AI workloads scale worldwide. The roadmap builds on AdaniConnex's existing 2 GW national data centre portfolio and outlines an expansion to a targeted 5 GW capacity.

Partnership-driven Growth

The Group cited partnerships with Google to develop a gigawatt-scale AI data centre campus in Visakhapatnam, along with additional campuses planned in Noida. It also highlighted collaborations with Microsoft across Hyderabad and Pune. Separately, Adani said it will deepen its partnership with Flipkart through a second AI-focused data centre designed to support digital commerce and high-performance computing requirements.

Adani Group said the proposed 5 GW platform will combine renewable power generation, transmission networks and hyperscale compute infrastructure within a unified framework. Facilities are expected to be designed for high-density AI workloads, incorporating liquid cooling systems and high-efficiency power configurations.

Dedicated compute resources will support Indian Large Language Models (LLMs) and national data initiatives, with a focus on ensuring data sovereignty, reliability and scalability.

Renewable Energy Backbone

The renewable energy portfolio remains central to the strategy. Adani Green Energy's 30 GW Khavda project will play a key role, with more than 10 GW already operational. The Group also committed USD 55 billion toward expanding its renewable capacity, including investments in large-scale battery energy storage systems (BESS) to enhance grid stability.

To enable global integration, the Group plans to expand cable landing station infrastructure linked to its port network, facilitating low-latency international connectivity. Adani also said it will pursue co-investments in domestic manufacturing of critical components such as transformers, inverters, power electronics and thermal management systems to reduce exposure to global supply-chain disruptions.

The Group said the initiative aligns with programmes such as PM Gati Shakti and will integrate agentic AI technologies across logistics, ports and industrial operations. Its AI-based industry cloud platform, currently managing renewable assets in real time, will underpin expanded compute capabilities.

Adani Group said a portion of GPU capacity will be allocated to Indian startups, research institutions and deep-tech ventures to address compute constraints. The company also outlined plans to collaborate with academic institutions on AI Infrastructure Engineering programmes, applied research labs and fellowship initiatives.

The Group described the commitment as a long-term national-scale infrastructure platform and invited participation from global technology companies and sovereign institutions.

Amodei highlighted India's extraordinary advantages for AI development.

Anthropic has been rapidly expanding in India, with a new office in Bengaluru coming amidst the wide adoption of Claude AI in the country's various professional areas. Dario Amodei, CEO of Anthropic, has now called on India to harness AI to transform governance into a more efficient and citizen-centric system, describing it as one of the most valuable applications of the technology.

Speaking at the maiden edition of Anthropic's Builder Summit in Bengaluru, which also coincided with the official launch of the company's Bengaluru operations, Amodei addressed around 250 developers and entrepreneurs. He acknowledged that while AI generates immense value, it will also cause significant economic disruption, making it essential to ensure broad sharing of benefits.

"That's a really valuable use of the technology," Amodei said, referring to AI's potential in governance. He highlighted India's unique position as the world's largest democracy to lead in this area. Amodei praised the speed of government adoption in India, noting that the Ministry of Statistics is already building an AI system to query economic data and statistics.

"Generally, govt bodies elsewhere don't move this fast," he remarked.

Amodei also pointed to the rapid growth of Anthropic's business in India, revealing that the company's run-rate revenue in the market has doubled over the past four months. "It's just really incredible... It mirrors the general progress and explosion in Claude models and coding models. But I think it's even more extreme in India than we've seen in other places of the world."

India provides scale and diversity for AI innovation

Amodei highlighted India's extraordinary advantages for AI development. "You can run experiments with hundreds of millions of people. That scale allows entrepreneurs and builders to pivot quickly and learn faster in ways that simply aren't possible in smaller markets."

"With so many regional languages in India, I've seen some cool applications, including in the non-profit and social benefit space," he added. Anthropic's Claude models are actively working to support a wide range of languages, including India's long tail of regional ones.

AI engineers to focus on high-impact AI apps

In continuing with the conversation, Amodei encouraged builders to focus on areas where AI could help with real-world applications, like medicine and biology - which Amodei claimed to be the most durable and defensible businesses.

"The biggest moats will come from applications tied to the physical world -- things that aren't easy to do and require specialised skills. Biology and medicine fit all of those criteria. It's a complex, knowledge-heavy space that involves navigating regulatory systems, but that hard work will lead to durable businesses. So I strongly encourage builders to move in that direction," he stated.

Other than the new office, Anthropic recently announced new enterprise tools for legal, sales, and marketing functions, which contributed to a sharp global IT stock sell-off.

5 Best Phones with AI Features Under ₹25,000: AI features are very essential in 2026 as they lessen your work and provide top-quality results without extra effort, so if you are also looking for 5G phones with AI features that come under a budget of 25k, then here I have provided a list of top 5 best 5G phones with AI features under this price segment. These phones are not just equipped with AI features, but they also get a large display with amazing viewing experience, powerful chipset, and a very nice camera setup. Here are all the features as well as their prices given below.

OnePlus Nord CE5 5G phone includes AI Writer for content generation, AI Unblur for sharpening images, and AI Perfect Shot to enhance photography details. The phone also has a large 6.77-inch Fluid AMOLED display with 120 Hz refresh rate, and it is powered by powerful MediaTek Dimensity 8350 Apex processor, and it also gets a massive 7100 mAh battery with 80W fast charging support. The phone is currently available at just 24,999 rupees.

Motorola Edge 60 Fusion 5G phone is powered by Moto AI, which assists with everyday task automation and intelligent camera enhancements. Along with that, it is equipped with a large 6.67-inch pOLED display with 120 Hz refresh rate, and it gets powerful MediaTek Dimensity 7400 chipset, and it has 50 MP + 13 MP dual rear camera setup, and it is currently priced around 22,999 rupees.

Oppo K13 5G phone offers a suite including AI Erase 2.0, AI Clarity Enhancer, AI Unblur, and AI Reflection Remover, and it also features AI Writer and Screen Translator. Similarly, it includes a large 6.67-inch AMOLED display with 120 Hz refresh rate, which enables you to do multitasking without any lag issue, and it is powered by Snapdragon 6 Gen 4 processor, and it has a massive 7000 mAh battery pack with 80W SuperVOOC charging support, and it is currently priced around 22,999 rupees.

Redmi Note 15 5G phone utilises HyperOS 2 to power features like AI Erase, AI Reflection Remover, AI Face Unlock, and AI Beautify, and it also sports a 6.77-inch AMOLED display, and it is powered by Snapdragon 6 Gen 3 processor, and provides powerful performance. Simultaneously, the phone gets a 108 MP main sensor, and it is equipped with a massive 5520 mAh battery pack and supports 45W fast charging. It is currently priced around 22,999 rupees.

Samsung Galaxy A17 5G phone is equipped with a built-in personal AI assistant, Gemini, accessible via the power button for real-time task support, and it has a 6.7-inch Super AMOLED display with 90 Hz smooth refresh rate, and it is powered by Exynos 1330 processor, and gets a massive 5000 mAh battery pack. Similarly, Samsung also promises 6 years of updates with this phone, and it is currently priced around 20,499 rupees.

Bill Gates is set to take part in the India AI Impact Summit 2026 at Bharat Mandapam in New Delhi, his foundation has clarified, rejecting reports that claimed he would not attend. The confirmation ends speculation around the Microsoft co-founder's participation in the high-profile technology gathering.

A spokesperson for the Gates Foundation told News24 'Bill Gates is attending the AI Impact Summit. He will be delivering his keynote as scheduled.' He is listed among the summit's prominent global participants, alongside OpenAI chief Sam Altman and Google CEO Sundar Pichai.

Separately, Gates commended Andhra Pradesh for leveraging artificial intelligence and technology to fast-track development. After his visit to Amaravati, he said it was encouraging to witness how innovation is being used to strengthen sectors such as health, agriculture and education. In a post on X, he thanked Chief Minister N. Chandrababu Naidu for the warm reception and praised the state's technology-driven progress.

On Monday, Gates held a series of engagements in the greenfield capital. He met the chief minister and senior officials, reviewed digital governance initiatives and toured a banana orchard on the banks of the Krishna river in Undavalli village of Guntur district.

The India AI Impact Summit 2026 is being described as the first major international AI summit hosted in the Global South. It is designed to present artificial intelligence in action by bringing together policymakers, innovators and implementers on one common platform.

The Expo component of the summit is spread across 10 arenas covering more than 70,000 square metres. It features participation from leading global technology firms, startups, universities, research institutions, Union Ministries, state governments and international collaborators.

Thirteen country pavilions are showcasing cross-border cooperation in the AI space. Countries represented include Australia, Japan, Russia, the United Kingdom, France, Germany, Italy, the Netherlands, Switzerland, Serbia, Estonia, Tajikistan and several African nations.

The event includes more than 300 carefully curated exhibition booths and live demonstrations organised under three broad themes: People, Planet and Progress, highlighting AI's influence across multiple sectors.

More than 600 high-potential startups are part of the summit, many of which are already building AI solutions that operate at scale and have global relevance.

Organisers estimate that the Expo will attract over 2.5 lakh visitors, including international delegates. The summit is expected to strengthen global partnerships and open new business avenues within the expanding artificial intelligence ecosystem.

Union Education Minister Dharmendra Pradhan on Tuesday urged India's youth to harness the full potential of Artificial Intelligence (AI), asserting that the technology will play a pivotal role in realising the vision of a Viksit Bharat.

Speaking at the AI Impact Summit, Pradhan called on the next generation to deepen their engagement with AI, strengthen teamwork, and explore the vast possibilities the technology unlocks.

Pradhan said that AI presents a significant opportunity for India, as he cited the country's multi-dimensional, multicultural and multilingual facets, alongside its rich history and legacy.

ALSO READ: India AI Impact Summit 2026 enters Day 2: Key sessions to watch today

He noted that AI can be a powerful tool in understanding and navigating the complexities.

"AI is giving us an important opportunity. We have a multi-dimensional, multi-cultural and multi-lingual, complex society, with a rich history and legacy... AI can be an important tool in understanding its complexity and multi-dimensionality," Pradhan said.

Also Read

AI helped predict Chinese move in Arunachal, avoid casualties: Lt Gen Rana

AI in health can be beneficial if used under supervision: Experts at AI Summit

'Double helix' approach needed for AI and climate action, say experts

Panchayat Ministry highlights 'Sabhasaar' at India AI Impact Summit

Adoption, not innovation, biggest gap in AI rollout: Experts at AI Summit

The minister exhorted India's youth to deepen their understanding of AI and the wide-ranging opportunities it offers; he encouraged them to embrace the transformation rather than fear it.

"There is no reason to fear it. Let universities, startups and big companies come together to create a big vision. In coming days, driven by youth, India will emerge as an AI-led global knowledge centre," Pradhan said.

The minister said India has set a collective goal of realising the vision of Viksit Bharat by 2047, transforming the country into a developed economy, and asserted that AI would be a key pathway to achieving that ambition.

"We have a collective target that by 2047, India has to collectively realise vision for Viksit Bharat or developed economy and AI is the path to do that," Pradhan said.

More From This Section

Focus on governance, scale vital for building trusted AI: Industry leaders

Samsung previews Zero-Peeking Privacy control in Galaxy S26 series: Details

Involving educational institutions at early stages necessary: Zoho's Vembu

Gnani.ai unveils 5B-parameter voice-to-voice AI model at India AI Summit

YouTube comments, descriptions missing amid ad-block crackdown: Report

Prompt injection is a technique where hackers embed malicious instructions in web pages or files to manipulate an AI system into revealing confidential information or performing unintended actions. As millions use AI chatbots like ChatGPT for tasks such as document reading, web browsing, and app connections, the threat from these tricks becomes more pronounced. OpenAI's new tools aim to shield users from these risks.

Lockdown Mode is an optional feature that tightly controls how ChatGPT interacts with external systems. When activated, it can restrict or disable certain tools and connections like live web browsing or integrations sending/receiving data from third-party services. The idea is to reduce these outside interactions and minimize the "attack surface" hackers could exploit.

Advertisement

OpenAI says Lockdown Mode isn't needed by most everyday users. It's mainly targeted at those who deal with highly sensitive information or think they might be at an elevated risk, such as journalists, executives, researchers, or security professionals. The company hopes this feature will give these high-risk users more control over their data and privacy while using ChatGPT.

Advertisement

Along with Lockdown Mode, OpenAI is also introducing Elevated Risk labels in ChatGPT. These labels will appear next to tools or features that expose more outside content or systems. For instance, if a feature connects to external content or provides broader system access, ChatGPT will show a clear warning about the potential risks. This way, users can decide whether to proceed with caution or keep things private.

CNBC's Jim Cramer recently took to social media to highlight the "hallucination" hurdles still facing artificial intelligence (AI), revealing that while he remains an AI optimist, the technology's struggle with basic facts remains a significant barrier to professional trust.

The 90% Threshold

Cramer shared insights from a recent conversation with Nvidia Corp. (NASDAQ:NVDA) CEO Jensen Huang, noting that the current state of AI is roughly 90% accurate.

While that figure sounds impressive for general tasks, Cramer argued that the remaining 10% creates a "gap" that prevents the technology from being ready for high-stakes environments.

"I am not a Luddite," Cramer posted on X. "But as someone named Jensen Huang told me not that long ago, about 90% of what we get from AI right now is right. So, I will grow more confident as we close that gap." He added a realistic timeline for this evolution, noting, "we will... but not this year, or maybe next."

Gemini's 'Rockport' Hallucination

Despite his Philadelphia roots, the AI insisted he was from Rockport, Massachusetts, and even labeled him an "unofficial spokesperson" for the city.

"I have never been to Rockport, Mass and I am from Philadelphia," Cramer clarified. He joked about the bot's confidence in its errors, questioning if he would soon find out he is a "Patriots season ticket holder" as well.

Trust In Professional Spaces

Beyond personal anecdotes, Cramer questioned the viability of current AI models like Gemini and Anthropic's Claude for industries where precision is mandatory.

"I know AI will get better but can someone tell me one major law firm that would rely on this stuff for a real, paying client and still be able to sleep at night?" Cramer asked.

He concluded that while AI is excellent for summaries and stock histories, "guessing's for jokers where I'm from," suggesting the tech is still "closer and closer" but not yet a reliable "ringer."

Blue Chips Gain As Tech Lags In 2026

As of Friday's close, the Dow Jones index rose 2.31% year-to-date, whereas the S&P 500 was 0.33% lower and the Nasdaq Composite index was down 2.97% in 2026.

Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.

Photo courtesy: jamesonwu1972 On Shutterstock.com

Market News and Data brought to you by Benzinga APIs

To add Benzinga News as your preferred source on Google, click here.

Shoppersidentifiedfinding the right product or service as the most time-consuming task.

ShowQuick Read

Summary is AI-generated, newsroom-reviewed

* About 70% of online shoppers in India plan to use AI to verify products and sellers

* The survey gathered over 75,000 responses from shoppers across 332 districts in India

* Finding the right product is the top challenge for 69% of online shoppers surveyed

Did our AI summary help?

Let us know.

Switch To Beeps Mode

New Delhi:

About seven in 10 online shoppers in India plan to use artificial intelligence in the next year to verify products and sellers, a nationwide survey shows. They also want AI to provide personalised recommendations based on their tastes and needs.

The survey was conducted by LocalCircles, a civic engagement platform, across 332 districts and received over 75,000 responses from online shoppers. Of the respondents, around 44 per cent were from tier 1 cities, 32 per cent from tier 2, and 24 per cent from tier 3, 4, 5, and rural districts.

Shoppers identified finding the right product or service as the most time-consuming task, with 69 per cent reporting it as their main difficulty. Around half of respondents also said they struggle to find the best combination of price, delivery time, and return policy. Other common issues include verifying seller credentials, cited by 43 per cent, and checking if reviews and ratings are authentic, reported by 40 per cent of respondents.

About 68 per cent said they wanted to use AI to check the authenticity of products and sellers, while 64 per cent said they would use it to discover products or get personalised recommendations.

Around 56 per cent want to compare products and prices, another 56 per cent said they would use AI to verify reviews and ratings, 46 per cent want AI to resolve doubts faster, 36 per cent to track price trends or discounts, and 22 per cent to create shopping lists or plan purchases.

About half of respondents said they prefer to use AI directly on e-commerce apps or websites, rather than on external AI platforms like ChatGPT, Grok, Meta, or Gemini.

Shoppers also have several concerns. About 73 per cent worry about the safety of their data, 69 per cent are concerned about how AI evaluates products and sellers, and 61 per cent fear AI may suggest biased or sponsored products.

Even after the government issued rules under the Digital Personal Data Protection Act, 2023, consumers remain cautious about using AI for online shopping, the survey showed. The Act regulates how personal data is collected, used, and shared, gives individuals rights over their data, and sets penalties for misuse.

Show full article

Track Latest News Live on NDTV.com and get news updates from India and around the world

Artificial Intelligence, Online Shopping, AI Shopping

India's flagship AI Impact Summit began in New Delhi with lofty promises of shaping the future of artificial intelligence -- yet the event was dominated as much by logistical turbulence as by technological vision.

The five-day conference, inaugurated by Prime Minister Narendra Modi, has been positioned as the first major international AI gathering hosted in the Global South. Delegations from over 100 countries are participating, with global tech leaders including Sam Altman of OpenAI and Sundar Pichai of Alphabet Inc. expected to attend.

However, for many delegates, the summit's first impressions were shaped less by AI breakthroughs and more by long queues and confusion outside the venue.

By late morning, attendees at Bharat Mandapam reported overcrowding, bottlenecks at entry points, and inconsistent instructions from security personnel. Some delegates said they waited for hours to gain access, while others were temporarily evacuated from exhibition halls without clear explanations.

Several founders and exhibitors took to social media, describing abrupt security sweeps and last-minute hall closures that left them stranded outside. Participants said communication from organisers was limited, adding to the sense of disorder.

In addition to crowd control issues, some exhibitors alleged that products were stolen from their stalls amid the confusion. Others cited restricted access to food and drinking water during peak hours.

Further frustration arose when visitors discovered that certain food counters accepted only cash payments -- a hurdle for international guests accustomed to digital transactions in a country that promotes itself as a leader in fintech adoption.

The disruptions prompted criticism over planning and crowd management, particularly given the summit's scale and global visibility.

In his inaugural address, Modi described the summit as a demonstration of "India's talent, innovation, and AI potential," asserting that the country aims to build solutions "for the world, not just for India."

Information Technology Minister Ashwini Vaishnaw said the event would examine both the promise and the perils of artificial intelligence. Organisers had projected the summit as a collaborative platform for policymakers, industry leaders, and researchers to chart India's AI roadmap. While the conference succeeded in drawing global attention, its opening-day setbacks have raised concerns about execution at large-scale tech gatherings.

Sridhar Vembu, Chief Scientist of Zoho Corporation, sought to calm fears about artificial intelligence gutting India's IT workforce on Monday, arguing that the technology would ultimately create more jobs than it destroys - even as he acknowledged the disruption would be real and the transition messy.

Speaking to Moneycontrol on the sidelines of the India AI Impact Summit 2026 in New Delhi, Vembu said the anxiety gripping India's tech sector was understandable but misplaced. "Typically all these technologies create more jobs than they destroy," he said. "It may not directly be the jobs that they are destroying, but if we are adaptive, we can leverage the opportunities."

The Engineer's Dilemma

The concern is not abstract. A recent rout in Indian IT stocks has crystallised fears that AI-driven automation could fundamentally alter the hiring calculus of an industry that employs millions. Vembu, speaking as a software engineer himself, did not dismiss those fears. In his own teams at Zoho, he said, AI coding tools had delivered results so impressive that engineers were left asking an existential question, "So what does it leave us humans?"

His answer was a reorientation, not a reassurance. "Software engineers will evolve to becoming closer to the customer, solving customer problems," he said. "There's millions of problems to solve. Let's solve them."

It was a notably more measured position than the one Vembu had taken just weeks earlier. During the launch of Anthropic's coding plugins, he had written bluntlyon social media, "At this point, it is best for those of us who depend on writing code for a living to start considering alternative livelihoods. I include myself in this. I don't say this in panic, but with calm acceptance and embrace."

His latest remarks suggest he has since arrived at a more structured view of what that adaptation might actually look like.

The 10-minute delivery jobs segment

To explain how jobs might emerge from disruption in ways that are impossible to predict in advance, Vembu reached for a concrete Indian example. "Take the 10-minute delivery job sector - that ultimately came from the smartphone," he said. "But when the smartphone arrived, we would not have known that 10-minute delivery would arrive."

The analogy cuts both ways. Quick commerce created an entirely new category of employment for millions of gig workers - but those were not the jobs anyone was protecting or anticipating when mobile internet arrived. Vembu's point is that the same logic applies to AI. The jobs it creates may look nothing like the jobs it displaces, which is precisely why he argued India needs "a lot of entrepreneurs who are figuring this out."

The structural challenge

Vembu was candid that not everything would survive the transition intact. He flagged the SaaS business model specifically - the bread and butter of many Indian software companies - as vulnerable. "The way that package software is sold today, that might change," he said. "That obviously is a challenge for companies like us."

On the broader question of whether India was doing enough to generate employment as IT hiring slows, Vembu fell back on structural optimism: an open economy, a strong domestic consumption story, and a young, tech-savvy population. "India is actually the most AI-optimistic country in the world," he said. "I am an optimist."

Whether that optimism is earned or premature, job creation will define India's AI decade more than any summit keynote.

Lt Gen Dinesh Singh Rana, Commander-in-Chief, Strategic Forces Command | Image: Screengrab- IndiaAI

India successfully used artificial intelligence (AI)-powered prediction tools to anticipate and foil an unprecedented Chinese attempt along the Line of Actual Control (LAC) in Arunachal Pradesh, Lt Gen Dinesh Singh Rana, Commander-in-Chief, Strategic Forces Command, said.

Speaking at the India AI Impact Summit 2026, the senior military official highlighted how AI is reshaping modern warfare, strategic doctrine and national security planning. He also cautioned about the risks posed by advanced systems such as artificial superintelligence.

Recounting his tenure commanding a Corps responsible for security along the LAC in Arunachal Pradesh, Lt Gen Rana said a locally developed, low-cost AI software system had helped detect early warning indicators of a Chinese military build-up in a disputed sector.

"We could see through some AI systems that something was building up. Finally, we were able to predict the timing of their move," he said.

He described the Chinese move as "unprecedented" and said timely AI-enabled intelligence ensured effective force positioning and evacuation planning, resulting in "no casualties" on the Indian side.

ALSO READ: 'Double helix' approach needed for AI and climate action, say experts

Also Read

AI in health can be beneficial if used under supervision: Experts at AI Summit

'Double helix' approach needed for AI and climate action, say experts

Panchayat Ministry highlights 'Sabhasaar' at India AI Impact Summit

Ireland to cut English-language student intake, review work, visa rules

AM Group starts development of $25 billion AI data centre project in UP

AI race already underway globally

Lt Gen Rana said AI is no longer a "futuristic concept" but has become central to power competition globally.

He noted that worldwide investments in AI are rising sharply, particularly in the United States and China, and described the situation as an ongoing "AI race".

He said AI is not limited to software systems but extends to workforce development, infrastructure, military doctrine and command structures.

He pointed out that major powers are integrating AI into command-and-control structures, autonomous weapons, precision strike systems and target identification.

AI is also increasingly being used in logistics, predictive maintenance, maritime domain awareness, electronic warfare, cyber defence and countering misinformation.

In the cyber domain, he said AI-powered systems can both conduct sophisticated offensive operations and strengthen defensive capabilities by identifying vulnerabilities and plugging gaps.

Risks of superintelligence

Highlighting the risks associated with AI, Lt Gen Rana warned that the transition from narrow AI systems to artificial superintelligence (ASI), if achieved, would create both significant opportunities and major challenges.

Being a dual-use technology, advanced AI systems could potentially be misused by rogue actors or leveraged for strategic coercion, similar to nuclear deterrence dynamics witnessed during the Cold War.

He said non-proliferation of such capabilities would be far more challenging than in the nuclear domain, given the nature of digital technologies.

ALSO READ: Adoption, not innovation, biggest gap in AI rollout: Experts at AI Summit

Call for India's AI security doctrine

Highlighting lessons from the Russia-Ukraine conflict, he stressed the importance of deploying scalable, low-cost AI-enabled battlefield systems, including autonomous drones, rather than relying solely on top-down, high-end capabilities.

He called for India to develop a comprehensive AI security doctrine and align investments accordingly.

He also advocated building "agentic systems" at the national level to integrate economic security, cyber, space and military inputs, enabling top leadership to assess the second- and third-order effects of strategic decisions.

Emphasising civil-military fusion, he said India must boost domestic talent and capabilities in artificial intelligence.

"As we aspire to become a superpower, we must also become a responsible AI power- just as we are a responsible nuclear power," he said, adding that advanced AI capabilities, if acquired, would be used responsibly.

More From This Section

PM Narendra Modi to visit Israel next week, says Benjamin Netanyahu

Indo-US defence ties continue to grow stronger: US Ambassador Sergio Gor

French President Macron arrives in Mumbai for strategic talks with PM Modi

Speed up next-gen aero engine development: Rajnath Singh to scientists

Jaishankar, Canadian counterpart discuss ties ahead of Carney's India visit

Just weeks after acquiring Singapore-based startup Butterfly Effect's Manus AI, Meta is moving swiftly to integrate the agentic AI platform into mainstream messaging apps. The first stop: Telegram. WhatsApp and Messenger are expected to follow soon.

Meta's earlier attempts to gain widespread adoption for its in-house chatbot did not generate the traction the company had hoped for. With Manus AI, however, the strategy appears more ambitious. Instead of a standard chatbot that responds to prompts, Manus is built as an autonomous, general-purpose AI agent capable of planning, reasoning and executing multi-step tasks independently.

In an official announcement, Manus introduced what it calls personal AI Agents for chat. The feature is now live on Telegram, giving users access to the same core system available on the main Manus AI platform, but directly inside a messaging app. This eliminates the need to switch between apps or log into a separate dashboard.

"Telegram is the first supported channel. Manus Agents with Telegram support is available today to all users across all subscription tiers, with more platforms coming soon," the official blog post states.

The rollout marks a significant shift in how AI tools are delivered to everyday users. Rather than functioning purely as conversational assistants, Manus agents focus on execution. They break down user requests into smaller steps, process them sequentially, and then present a completed result.

Users can connect their Manus account to Telegram by scanning a QR code. Once linked, the AI appears as a regular chat contact and operates within the same conversation window. There is no need for coding skills, server setup, or technical configuration.

Within chat, users can assign tasks in plain language. Manus can summarise PDFs, analyse spreadsheets, transcribe audio clips, extract key insights from documents, draft structured reports, generate images, and even debug or write code. The system also supports file uploads, voice notes, and images, allowing different types of content to be processed without leaving the app.

One of Manus AI's standout features is its ability to retain memory and connect with productivity tools such as Gmail and calendar applications. This enables more personalised and context-aware interactions over time.

Users can also choose between two AI models depending on their needs. Manus 1.6 Max is designed for deeper reasoning and more complex workflows, while Manus 1.6 Lite prioritises speed for everyday tasks.

According to the company, Manus is built around execution rather than just conversation. Just like other Agentic AI setups, Manus breaks down requests into smaller steps and completes them sequentially before delivering a final result.

-- Analysing spreadsheets and extracting insights

As companies like Google, OpenAI, Anthropic and Meta race to embed artificial intelligence into daily digital interactions, Meta's move to integrate Manus directly into messaging apps signals a stronger push toward agent-driven AI experiences. By embedding advanced task automation into familiar chat environments, the company is betting that users will prefer intelligent agents that do the work -- not just talk about it.

Dr. Alice Chiao used to teach emergency medicine to students at Stanford University's medical school. Now, she's teaching artificial intelligence-powered chatbots to think, diagnose and prescribe like her.

Chiao is part of a booming new economy of professional experts in their fields who are training AI through a process called reinforcement learning, essentially grading AI's responses and teaching models to improve through trial and error. It's a rapidly growing service industry for AI frontier labs, estimated to be worth at least $17 billion, according to Pitchbook Senior AI Analyst, Dimitri Zabelin.

Chiao is one of tens of thousands of experts working with Mercor, one of the companies that help manage reinforcement learning for major AI companies. Mercor has contracts with experts in subjects ranging from medicine, law and finance to comedy, sports and even wine. Experts can earn up to hundreds of dollars per hour teaching AI to do their own jobs.

"AI is going to be the new Doctor Google, the new WebMD that people will go to, to seek out medical information. I knew that I needed to be a part of that to make sure that the information is accurate, that it's safe, and that it makes sense to the person using it," Chiao told CNN.

AI models are trained on massive amounts of data. But that training doesn't do much good without what's known as "reinforcement learning," a process that involves human experts teaching models the differences between good and bad responses. Companies like OpenAI, Google and Anthropic use what Mercor's CEO Brendan Foody described as "large armies of people" to do just that.

Mercor was recently valued at $10 billion.

CNN

Uncertainty over how AI will reshape various industries hit a fever pitch over the last two weeks. Software stocks plunged in early February following the release of a new tool from Anthropic that tailors its model for work in specific industries like legal and finance. Then, a viral essay from a tech CEO swept the internet with stark declarations about how AI could disrupt jobs. And some say Mercor is causing job displacement, replacing stable full-time careers with gig work that will contribute to AI taking human jobs.

But Chiao doesn't see her work through Mercor as teaching AI how to do her job. Instead, she views it as ensuring AI models are safe and capable enough to help doctors spend more time with patients and less time filling out forms. She sees AI as eventually being able to assist doctors with reading scans, filling out charts and taking notes.

"Physicians were selected because we really want to help people. We want to heal. We want to spend time talking to people -- listening, engaging," Chiao said. "I don't want to see it as AI taking over our jobs. I want to see it as AI taking over the aspects of our jobs that prevent us from being good doctors, good healers and good listeners."

Training Dr. AI

When Chiao is training AI models, she uses real scenarios she's encountered in her decades as a doctor in both primary and emergency medicine. That includes asking questions from both the patient's and doctor's perspective. A patient, for example, might ask whether their child should see a doctor when experiencing a cough or fever. But the system also needs to know how to respond when presented with medical jargon -- like what a physician might see on an intake form.

The AI model sometimes provides answers Chiao wouldn't have thought of herself, she said. But other times, she sees a need for professionals like herself to step in.

Dr. Alice Chiao.

CNN

"Sometimes there will be things that don't quite make sense, and I think, 'Oh, this could be misleading,' or 'This could be alarmist,' or 'This is not quite safe to put in a response,'" Chiao said. "And those are where I intervene and say, 'OK, this is where I need to craft something that makes this safe, accurate, and applicable to the user at hand.'"

Mercor's experts grade a model's response using a rubric they've created after consulting with a team of other experts in their field. Those responses are fed back into the model, which is trained to aim for good grades.

As for AI in medicine, Chiao said patients should use today's AI model tools as a starting point before talking to a doctor. The technology is not a replacement for a doctor like herself with 20 years in the field.

"There is a gut feeling that comes with experience, that comes with sitting with a patient, looking them in the eye, and seeing something that is beyond their history, their lab values, the words that are coming out of their mouth," Chiao said. "So, this is where it's really important to know that the AI is not a doctor, it's not a human being.

The most popular experts Mercor hires for are in software engineering, followed by finance, medicine and law, Foody, Mercor's CEO, told CNN. Job posts on Mercor can range widely, calling for everything from journalists to mechanics.

Mercor, a startup AI company, is headquartered in San Francisco.

CNN

But Foody notes that not everything can be taught, and the more subjective the task, the more difficult it is for AI to master.

One example is comedy. Mercor tried to train one AI model to be funnier by hiring comedians from the Harvard Lampoon, an iconic comedy publication from Harvard University.

"They were cracking all these jokes and writing all these rubrics to improve models and how funny they are," Foody said.

The problem, however, is one that's obvious to humans but not so much to machines: People have different opinions on what's funny.

"What you actually need is more localization of how humor varies by geography, and (answer) how do we have experts that can understand what jokes are in all of these different domains," Foody said.

From zero to $10 billion in three years

Before Foody and his Mercor co-founders set out to help AI models get better at human jobs, the company had a very different goal: helping people get hired.

Mercor, which Foody co-founded three years ago at age 19 with friends Adarsh Hiremath and Surya Midha, started as a recruiting and human resources platform. When they shifted the company's focus to AI, their rolodex of resumes was the perfect starting point for finding the experts AI companies were seeking.

Foody said Mercor is now paying out more than $1 million per day across thousands of experts, and in less than two years has grown from $1 million in revenue run rate to over $500 million. Pitchbook's Zabelin said the company is valued at more than $10 billion, adding that Mercor and its competitors' high values show investors think services like human feedback and expert testing of AI models are becoming a permanent and essential part of how AI systems are built and improved.

Mercor's CEO Brendan Foody.

CNN

Mercor is not the only company in this space. Last year, Meta made a $14 billion investment in Scale AI -- which operates in a similar space as Mercor -- bringing on its then-28-year-old founder Alexandr Wang as its chief AI officer. Other competitors like Surge AI, Handshake and Micro1 have helped mint a new class of young, ultra-wealthy tech founders.

While valuations fluctuate, 22-year-old Foody and his co-founders are likely some of the youngest tech founders to make the Forbes billionaire list since Mark Zuckerberg, who made the list at age 23.

"We of course were ambitious about what we wanted to do, but never could have imagined anything like this, especially happening so quickly. So, it feels very surreal," Foody said.

Foody has enjoyed some benefits of being a young billionaire (he said he treated his family to tickets for the SuperBowl). But his focus remains on growing a business that he sees as being critical to shaping the future of work despite the mounting concerns about AI displacing jobs.

In his view, Mercor's work is a step toward solving bigger problems.

"We need to cure cancer. We need to solve climate change," he said. "And making everyone 10 times more productive so that they're able to better work on those key problems is going to be a huge, huge benefit to how we make progress as a society."

Bharat1 says the moonshot project will allow them to ground AI agents

On the second day of the inaugural AI Impact Summit, which is being hosted in New Delhi, a very interesting announcement has emerged. A Bengaluru-based artificial intelligence (AI) startup, Bharat1, has laid out the plans to build the world's first "humanity-first" AI city in Bengaluru. While on the surface, the idea seems fantastical, the researchers are not really trying to have AI models run the entire city. Instead, they want to create a city-scale simulation where AI agents and physical AI can be stress tested in a real-world scenario.

Bharat1 Unveils Plans to Create an AI City

In a press release, the startup unveiled its vision to build a city-scale research and infrastructure initiative. Bharat1 highlighted that the deployment of increasingly capable AI systems has raised concerns about safety, reliability, and alignment. "While large language models have demonstrated remarkable capabilities in language and coding, intelligence built primarily on fragmented internet-scale data alone is unlikely to serve as a sufficient foundation for the next generation of autonomous and physical systems," the post added.

For Bharat1, the solution to the problem is validated, real-world intelligence grounded in real-world data. For the first step of its moonshot project, the startup proposes a 5,00,000 sq ft AI research and innovation campus in Sarjapura, Bengaluru. The location will offer facilities, such as model training, fine-tuning and inference, plug-and-play labs, connectivity to AI cloud services, and more.

During this period, the startup wants to develop and stress-test universal basic intelligence frameworks in controlled, real-world conditions, building open city-scale world models using structured, high-quality multimodal data, and creating robust validation layers for agentic and physical AI systems before they are deployed at scale.

"We are at an inflexion point in the AI revolution. Intelligence is moving out of labs and into the fabric of society. Scaling systems without validating them in complex, real-world environments is a risk we cannot afford. The B1 AI Superpark is the first step toward creating a city-scale validation framework where agentic and physical AI can evolve responsibly, not just powerfully," Umakant Soni, Co-Founder and CEO of Bharat1.AI, said at the AI Impact Summit.

Sarvam AI, the Indian AI startup that made headlines a few days ago for beating ChatGPT and Google Gemini, has announced plans of launching AI hardware devices. In an X post, Sarvam AI chief Pratyush Kumar revealed that the company was planning to launch AI smartglasses soon. This announcement comes during the India AI Impact Summit.

In an X post, Pratyush shared a minute-long teaser, showcasing the smartglasses. The Sarvam AI glasses are called Sarvam Kaze. The teaser shows a pair of glasses, similar to Meta RayBans, with visible camera placements.

The Sarvam AI chief also shared a picture of PM Narendra Modi wearing the Kaze smartglasses, likely at the India AI Impact Summit. Kumar wrote, "The first person to try them? The Prime Minister."

Pratyush Kumar revealed that the Sarvam Kaze smartglasses were designed and built in India. He wrote, "Launching Sarvam Kaze, our foray into getting our models into the your hands with our devices - designed and built here in India!" This would also make the Kaze the first India-made AI hardware.

According to Sarvam, the smartglasses are built to move AI away from screens and integrate it into everyday experiences. As Kumar explained, "Sarvam Kaze moves intelligence from the screen to the real world. You wear it. It listens, understands, responds, and captures what you see. And you can build custom experiences for it with the Sarvam platform."

The company also insists that the smartglasses will run on its in-house AI models. Pratyush Kumar confirmed that the Sarvam Kaze will launch in May 2026.

Smartglasses are increasingly becoming a notable tech product category. Meta has reportedly sold over 7 million units of the Meta RayBans in 2025 alone. As per reports, Apple too is expected to launch its own smartglasses in the future. However, currently, Sarvam could be an early player in this growing segment.

Although Microsoft has promised Windows 11 users to push AI features with utmost care while fixing the core issues of the operating system, the company is gradually making Copilot a more autonomous, agent-driven experience across its apps.

If you have used Copilot in the last year, you must be aware that the company has expanded AI functionality beyond simple text suggestions. As of now, Copilot can do multi-step research, data analysis, and even take care of browser-based automation. At a time when businesses and knowledge workers are looking for tools that can handle repetitive or complex tasks, Microsoft is reportedly looking to make Copilot a more central productivity assistant.

Copilot Tasks combines agents with scheduling

The news comes via TestingCatalog, which reports that Microsoft is working on a unified Tasks feature that can consolidate existing agentic tools into one interface. Worth noting that these will be accessible alongside the upcoming Projects module. As reported, users would get two entry points: a freeform New Task and a Scheduled Task, which supports one-time or recurring prompts on daily, weekly, or monthly cycles.

Interestingly, there's also a mention of a mode selector, which will offer Auto, Researcher, and Analyst. As the name suggests, Researcher uses OpenAI's deep research model for complex web and work-data investigations, whereas Analyst uses the o3-mini reasoning model for advanced data analysis with live Python execution.

Meanwhile, the new Auto mode seems to act as a general-purpose agent, capable of running end-to-end tasks that combine deep research with browser actions. TestingCatalog further notes that suggested prompts range from summarizing emails and generating presentations to booking hotels or drafting formal letters. Early testing found high-quality outputs for slides and web reports, hinting at meaningful upgrades for Copilot subscribers.

It's unclear when the feature will be released

Although there's no mention of a release date, the news outlet confirms that some features still don't work properly. Not to forget, Tasks could eventually extend across Windows and Edge, allowing system-level automation.

Now, the big question is: when will Microsoft release this feature? And if it does, could Tasks become the hub for managing both personal and professional AI-powered routines in the future?

Article in-content image source: TestingCatalog

A new government-commissioned study reveals that while UK businesses recognise the potential of artificial intelligence, adoption remains modest and uneven, with only 16% currently using AI technologies.

The research, conducted by IFF Research and Technopolis Group on behalf of the Department for Science, Innovation and Technology (DSIT), surveyed 3,500 UK businesses and conducted 100 qualitative interviews to understand current AI adoption patterns, barriers and future plans. The findings come as the government seeks to implement its AI Opportunities Action Plan, which aims to embed AI across the UK economy to boost productivity and competitiveness.

The study found that natural language processing and text generation are the most commonly used AI applications, with 85% of adopters utilising these tools, likely due to the accessibility of off-the-shelf generative AI solutions. In contrast, more advanced technologies like agentic AI remain rarely adopted at just 7%, reflecting both their relative newness and higher complexity.

Business size plays a significant role in adoption rates, with 36% of large businesses and 23% of medium-sized businesses currently using AI, compared with just 14% of micro businesses. Sector differences are also pronounced, with information and communication (43%), finance and real estate (21%) and business services (23%) leading adoption, while construction (88%), retail (86%) and hospitality (88%) sectors show the highest levels of non-adoption.

The research identified several key barriers to adoption. Lack of identified need (71%) and limited AI skills (60%) were the most commonly cited challenges. However, when businesses were asked to rate the significance of these barriers, ethical concerns emerged as the most critical obstacle (80%), followed by high costs (76%) and unclear regulation (72%).

For businesses that have adopted AI, the primary motivation was increasing efficiency and productivity (65%), with many reporting improved workforce productivity (75%) and had developed new or improved processes (57%). However, only 12% reported increased revenue, suggesting that while operational benefits are evident, financial returns are slower to materialise.

The study also revealed cautious optimism about AI's future role. Most businesses (71%) had considered AI for about a year before adoption, with many viewing it as inevitable for maintaining competitiveness.

Looking ahead, 65% of organisations anticipate increasing their AI-related budgets, with large businesses most likely to invest (82%). Planned investments focus primarily on off-the-shelf AI applications (65%) and embedding AI into existing systems (59%), rather than developing bespoke solutions.

The research highlights that while AI adoption is growing, significant challenges remain in terms of skills, costs and ethical considerations.

Artificial intelligence must be accessible, affordable and accountable if it is to drive meaningful social transformation, former NITI Aayog chief executive Amitabh Kant stated, stressing that the technology will fundamentally reshape daily life, governance and economic activity.

Emphasising India's growing role in global AI development, Kant observed that vast volumes of data originating from developing economies are already being used to refine international AI models. He argued that this strengthens the case for India to develop its own AI systems trained on domestic datasets rather than depending entirely on foreign platforms.

He stated, as quoted by Moneycontrol, that India is providing more data to OpenAI than the United States and that data from the Global South is helping refine global models, adding that such companies would sell high-cost products and therefore India needs to build models based on its own data.

Addressing the expanding influence of artificial intelligence, Kant underlined that inclusion must remain central to policy thinking and stressed the need for multilingual AI systems to ensure wider reach.

He cautioned that AI could exacerbate inequalities if access remains confined to higher-income groups, stating that extending AI capabilities to people below the poverty line should be prioritised to prevent technological divides from translating into broader social and economic disparities. He added that AI must ultimately serve developmental objectives, particularly in the Global South, where it has the potential to accelerate progress in healthcare, education and governance.

Drawing parallels with India's earlier digital public infrastructure initiatives, Kant stated that open-source frameworks played a decisive role in scaling the country's digital ecosystem and suggested that a similar model could be adopted for artificial intelligence, including the creation of a digital public identity layer for AI systems to ensure secure and equitable access.

He further noted that advances in data and machine learning present opportunities for what he described as technology pole vaulting, enabling developing economies to leapfrog traditional stages of growth, adding that the ultimate objective should be to deploy AI to improve lives, especially for the next billion users entering the digital ecosystem.

New Delhi: Artificial Intelligence (AI) is fast becoming an integral part of the corporate sector. The India-AI Impact Summit 2026 organised from February 16 to 20 at Bharat Mandapam in New Delhi is at the centre of focus as several international biggies from the tech sector are attending the event. The grand event is a platform where the government representatives of several countries and top executives of tech companies, researchers, and investors are set to discuss the future of artificial intelligence. The forum could bring AI stocks in focus.

AvenuesAI Limited

AvenuesAI is a global fintech and enterprise software platform which is engaged in the business of building digital payment and commerce infrastructure. The company recently changed its name from Infibeam Avenues Limited to AvenuesAI Limited. The company's business is entirely focussed on AI. Under the CCAvenue brand, the company is strengthening intelligent transactions and digital commerce. The firm has launched India's first agentic payment platform under the name PayCentral.ai, which is based on Google's AP2 protocol.

In Q3 FY26, the company's revenue was recorded at Rs 23,812 million, more than double compared to Rs 10,704 million in the same period last year. Net profit jumped to Rs 779 million.

Netweb Technologies

Netweb Technologies is in the business of manufacturing high-performance servers and computing infrastructure required for AI systems. The company designs GPU-based servers, storage, and networking solutions.

With the increasing demand for sovereign AI infrastructure, private cloud, and enterprise, computing is continuously increasing. Netweb could benefit from this. Management believes the company's current manufacturing capacity can handle revenue of around Rs 25 billion to Rs 30 billion without major capex. This will allow the company to achieve growth without putting additional pressure on its balance sheet. The company is currently focusing on expanding AI product capabilities and accelerating research through selective acquisitions.

In Q3 FY26, 64 per cent of the company's revenue came from AI systems, reflecting the rapid growth in this sector. Quarterly revenue recorded a jump of 141% year-on-year. However, due to large orders, the EBITDA margin stood at 12.2%, slightly lower than the previous quarter.

(Disclaimer: This article is only meant to provide information. News9 does not recommend buying or selling shares or subscriptions of any IPO, Mutual Funds, gold, silver, crypto assets and commodities like Copper, aluminium.)

Adapt, the San Francisco-based AI startup, last week announced $10 million in seed funding to build the AI computer for business. The round was co-led by Activant Capital and Headline with participation from Susa Ventures.

The funding validates Adapt's mission to make any company instantly AI-native. Their solution moves beyond vertical AI assistants to a horizontal AI platform that enables any role in an organization to collaborate with frontier AI agents, grounded in the context of their company's data, systems, and workflows.

The Rise of the AI Business Computer

Most enterprise AI usage today is still happening in silos, disconnected from where work actually happens and where the relevant business data is stored. While AI models have become increasingly accessible and intelligent, the context required for these AI models to be effective with company data is still stored in static systems of record that have limited AI capabilities.

Adapt connects to any system with an available API, with pre-built integrations for common business systems like HubSpot, Google Workspace, and Slack, and learns how to best interact with a company's APIs, tools, and people through advanced company knowledge and memory capabilities.

"We combine frontier models, company context, and powerful sandboxes into a single AI computer," said co-founder and CEO, Jim Benton. "Once Adapt is set up, anyone can use it in your team chat. We believe Adapt should be everywhere you work, and for most business users, that's in Slack, Microsoft Teams, or other business systems."

New Platform Capabilities

Alongside the funding, Adapt announced two major platform releases: Adapt Apps and Proactive Automation.

Adapt Apps transform AI-generated outputs into persistent, graphical user interfaces (GUIs). Users can create live dashboards, internal tools, or custom software through natural language, then save and relaunch them instantly, each powered by real-time data from connected business systems.

Proactive Automation advances Adapt's on-demand agentic capabilities with automated actions, proactive education, and continuous monitoring through observing systems in the background, surfacing insights, and taking action before users ask.

"AI only becomes transformational at work when it lives where you work, has full context across your systems, and can take action like a highly capable teammate," said Benton. "Adapt brings that experience - previously reserved for engineering teams given months to build custom solutions - to everyone in the company, instantly."

Customers of Adapt, such as Nathan Potter, CTO of Wander, are sharing meaningful results.

YOUTUBE VIDEO: https://www.youtube.com/watch?v=ZvU20qVSic8

Joshua Browder, Co-founder and CEO of DoNotPay, shared that a task which previously took a customer support agent 45 minutes now takes only one minute.

About Adapt

Founded by seasoned entrepreneurs Jim Benton (ClearSlide, Apollo, Chorus), Sean Smith (GlareDB), and John Andrew Entwistle (Wander, Coder), Adapt is the AI Computer for Business - the horizontal system of intelligence that connects a company's tools and data with frontier agents that work alongside your team in Slack.

Media Contact

Business: Adapt

Contact Name: Ashley McClelland

Contact Email: press@adapt.com

Website: https://adapt.com

Country: United States

Founded by seasoned entrepreneurs Jim Benton (ClearSlide, Apollo, Chorus), Sean Smith (GlareDB), and John Andrew Entwistle (Wander, Coder), Adapt is the AI Computer for Business - the horizontal system of intelligence that connects a company's tools and data with frontier agents that work alongside your team in Slack.

Related Items:Adapt Raises, AI Computer

Environmental NGOs, local campaigners and community groups are increasingly testing the adequacy of the UK's planning, environmental assessment and grid‑capacity frameworks, positioning data‑centre developments squarely at the intersection of green policy and digital growth. Environmental disputes involving data centres in the UK are increasing in volume and complexity. A landmark example is the legal challenge concerning a proposed 90MW hyperscale centre that was due to be built on a former landfill site in Iver, Buckinghamshire, just outside London. Campaign groups Foxglove and Global Action Plan argued that government officials failed to consider the environmental impact. Indeed, ahead of the permission hearing, the UK Government was forced to admit its planning approval for the data centre should be quashed, given the failure to fully consider the climate impact. On 22 January 2026, the High Court granted the campaigners permission for a judicial review claim to be held later this year, marking the first UK legal challenge of its kind specifically targeting the sustainability profile of a hyperscale data‑centre project. However, the challenges are not limited to climate impacts alone. Concerns frequently include: The high energy consumption of data centres, which opponents say could increase local electricity prices and strain regional grids. Water usage and heat generation, which are increasingly relevant in an era of water stress and heightened scrutiny over industrial cooling systems. Green Belt encroachment, where environmental groups argue that inappropriate development and insufficient environmental assessment. These disputes demonstrate a growing recognition that digital infrastructure must now withstand the same level of environmental scrutiny traditionally applied to heavy industry and energy. A shift toward embedding climate‑positive features into the design and operation of sites is becoming commercially and reputationally compelling. For instance: Reusing waste heat: the diversion of excess server heat to local district‑heating networks or community assets such as swimming pools could generate a tangible community benefit that can ease local resistance. Renewable energy integration (solar, wind): As regulatory and investor expectations tighten, leading operators are increasingly pairing facilities with on‑site solar generation or contracted renewable supply. Heat‑pump‑based cooling and efficiency innovation: With power‑usage‑effectiveness (PUE) metrics now central to sustainability reporting, operators are investing in heat‑pump technologies and advanced cooling systems to reduce water consumption and improve thermal efficiency. Mandatory reporting obligations under emerging EU and UK regimes, including waste‑heat utilisation, further support these investments. Sustainable construction and green certifications: Developers are seeking to evidence carbon‑sensitive approaches and secure investor confidence.

Edie 16th Feb 2026 read more "

Claims that AI can help fix climate dismissed as greenwashing. Industry using 'diversionary' tactics, says analyst, as energy-hungry complex functions such as video generation and deep research proliferate. Tech companies are conflating traditional artificial intelligence with generative AI when claiming the energy-hungry technology could help avert climate breakdown, according to a report. Most claims that AI can help avert climate breakdown refer to machine learning and not the energy-hungry chatbots and image generation tools driving the sector's explosive growth of gas-guzzling datacentres, the analysis of 154 statements found. The research, commissioned by nonprofits including Beyond Fossil Fuels and Climate Action Against Disinformation, did not find a single example where popular tools such as Google's Gemini or Microsoft's Copilot were leading to a "material, verifiable, and substantial" reduction in planet-heating emissions.

Guardian 17th Feb 2026 read more "

Enterprises are investing heavily in AI. Models continue to improve, experimentation budgets are growing, and executive attention has shifted toward AI-first strategies. Yet despite this momentum, many organizations still struggle with a familiar problem: integrating AI into everyday business workflows in a way that consistently delivers outcomes.

This disconnect is often described as an AI execution gap or a last-mile problem. Pilots succeed, demos impress, and proofs of concept circulate widely. Yet tangible business impact remains limited. The prevailing assumption is that the gap stems from AI systems not being accurate, reliable, or predictable enough. As a result, investment flows toward better models, more tuning, richer prompts, and additional orchestration layers.

This framing misses the underlying issue.

As discussed in AI-Ready Data vs. Analytics-Ready Data, the limiting factor is rarely the intelligence of the model alone. Instead, the bottleneck lies below the AI layer, in how enterprise data is designed, structured, and delivered. According to The Modern Data Report 2026: The Data Activation Gap, 68% say their data is not clean or reliable enough for AI use cases.

Most organizational data is engineered to serve technical teams and analytics use cases, not to support direct consumption by business decision-makers or autonomous systems. The data may be analytics-ready, but it is not AI-ready.

Also Read: AiThority Interview With Arun Subramaniyan, Founder & CEO, Articul8 AI

When AI systems are introduced into organizations, they do not automatically unlock value. Instead, they expose existing structural weaknesses in the data layer. Undefined semantics, inconsistent definitions, fragmented ownership, and unclear trust boundaries quickly become constraints.

This limitation is articulated in Reconfiguring AI as Data Discovery Agent(s)?, which argues that AI cannot discover value that the data platform cannot prove. AI systems operate within the bounds of what the data layer can semantically support. When business meaning is implicit, scattered, or inconsistently enforced, AI has little stable ground to reason over, regardless of model capability.

This helps explain why AI often performs well in isolated experiments but struggles when embedded into real-world workflows. The issue is not a lack of intelligence. It is a lack of usable context.

The absence of context is architectural rather than conceptual. As described in Rise of the Context Architecture, modern data platforms have historically optimized for storage, movement, and transformation while treating context as secondary. Business meaning, lineage, ownership, and constraints are often documented externally, enforced manually, or reconstructed downstream by humans.

AI systems are poorly suited to operate under these conditions. According to the same Modern Data Report 2026, 68% say their data is not clean or reliable enough for AI use cases. For AI to act reliably, context must be treated as a first-class citizen. It must be embedded in the data and exposed through well-defined interfaces. This includes semantic definitions as well as trust signals such as lineage, freshness, and governance constraints.

Without this layer, AI remains dependent on human interpretation, limiting its ability to scale into routine decision-making.

This reality requires a reframing of AI's role. AI is not an independent source of value. It is dependent on data that encodes business intent, constraints, and meaning in a form that machines can consume.

This dependency becomes apparent when organizations attempt to deploy AI agents across domains. As explored in How AI Agents & Data Products Work Together to Support Cross-Domain Decisions, agentic systems require trusted, governed interfaces to reason across multiple sources. Without data products that encapsulate business logic and usage contracts, agents remain narrow in scope and fragile in practice.

In this context, AI is constrained less by autonomy and more by the quality of its inputs.

If the AI value gap does not originate in the AI layer, it cannot be resolved there. The path forward is not additional infrastructure or more complex pipelines. It is the deliberate design of business-consumable data products.

Data products shift the focus from moving and storing data to delivering data as a trusted, usable interface. They package raw data with clear semantics, ownership, quality expectations, and governance. Most importantly, they are built around consumption by business users, applications, and AI systems.

Infrastructure enables scale, but it does not define meaning. Data products are where meaning, trust, and accountability are made explicit. The importance of this distinction is discussed in Data Lineage is Strategy: Beyond Observability and Debugging. When lineage is treated as a product capability rather than a passive metadata artifact, it becomes a practical trust mechanism that AI systems can reference. It signals not only where data came from, but whether it is appropriate for a given decision or action.

When data is delivered as a set of business-consumable products, the nature of AI adoption changes. AI systems no longer need to infer business meaning from raw tables or loosely defined metrics. Instead, they interact with curated interfaces that reflect how the business operates.

This shift supports more repeatable decision-making and explainable automation. AI systems can be embedded in workflows with clearer boundaries and expectations. The implication is straightforward. Closing the AI value gap is less about adding intelligence at the top of the stack and more about redesigning the data layer beneath it.

As AI adoption continues, organizations are recognizing that success depends on how data is delivered, not just how it is stored or processed. Data engineered solely for technical teams limits AI's ability to operate independently.

Business-native, business-consumable data is therefore a prerequisite for meaningful AI adoption.

Until this layer exists, enterprises are likely to continue investing in AI while seeing uneven results. When it does, the AI value gap begins to narrow, not due to changes in AI itself but because the underlying data has evolved.

In practice, achieving this shift requires more than intent. It requires operating models that treat data products, context, and governance as first-class capabilities rather than downstream concerns. Platforms that lead in this space, such as DataOS, reflect this direction by focusing on how data is packaged, governed, and exposed for consumption by business users, applications, and AI systems.

The relevance lies not in the platform itself, but in the architectural emphasis: designing data for use, not just for storage. As organizations explore ways to operationalize AI, approaches that prioritize business-consumable data products are increasingly becoming a practical foundation for closing the AI value gap.

AM Group said the platform would be vertically integrated, spanning carbon-free power generation, data centre infrastructure, high-performance chips, software, applications and consumption models ranging from AI Pods as a service to tokens as a servi

Energy firm AM Group on Monday said that it has begun work on a 1 gigawatt (GW) artificial intelligence (AI) data centre project in Uttar Pradesh, following an agreement signed with the state government last month.

The energy transition group, backed by the founders of Greenko Group, had signed a memorandum of understanding with Invest UP at Davos, Switzerland, on January 20 to set up the facility.

AM AI Labs, an affiliate of the group, on Monday said that it has received a letter of intent from the Yamuna Expressway Industrial Development Authority (Yeida) for the allocation of land for the first two phases of the project. These phases will comprise 150 MW and 200 MW, respectively, forming part of the planned 1 GW AI and high-performance computing data centre, the company said.

It added that 350 MW of compute capacity is expected to be operational by 2028, with the full 1 GW to be online by 2030. The project represents an estimated $25 billion investment and the deployment of around 500,000 high-performance chips, it added.

"This is an enormously ambitious project, and that is exactly what excites us. AM Group has done this before. In green hydrogen and ammonia, we set out to deliver the lowest cost molecules in the world, and we are doing it. We bring the same conviction to AI. Our goal is to deliver the lowest cost, most efficient AI tokens on the planet, powered by clean energy we own and operate. By integrating the full value chain from power generation through silicon to token delivery, we can optimise across layers in ways that no single segment player can," said Anil Chalamalasetty, group chairman of AM Group.

Also Read

Ireland to cut English-language student intake, review work, visa rules

Adoption, not innovation, biggest gap in AI rollout: Experts at AI Summit

Focus on governance, scale vital for building trusted AI: Industry leaders

Involving educational institutions at early stages necessary: Zoho's Vembu

Travelling without a British passport? Dual citizens may be denied UK entry

An AI token is a unit of text (word, sub-word, or character) that an AI model processes during training and inference.

AM Group said the platform would be vertically integrated, spanning carbon-free power generation, data centre infrastructure, high-performance chips, software, applications and consumption models ranging from AI Pods as a service to tokens as a service.

It added that the facility would be powered by renewable sources and pumped storage assets managed through its in-house energy architecture. The tokens generated at the facility, it said, would support domestic and international demand for training, inference and other advanced use cases.

The announcement came on the same day that India commenced its AI Impact Summit in New Delhi. The country's data centre industry is one of the fastest-growing globally. This comes on the back of India's fast-evolving AI market, which, according to a Deloitte report, is expected to reach $20-22 billion by 2027, posting a CAGR of 30 per cent.

More From This Section

NHAI accepts NHIT's ₹6,221 cr offer to monetise two highway assets

Retail segment drives 21% sequential rise in ARC acquisitions in Q3FY26

India's unemployment rate hit three-month high of 5% in Januarypremium

Nine-judge Supreme Court Bench to look at definition of 'industry'premium

RITES pushes in-service diesel locomotives on export fast-trackpremium

He highlighted how AI can optimize solar and wind integration into power grids, provide hyper-accurate flood forecasting, and increase agricultural resilience | Image: Khalid Anzar

From data ethics to a profound environmental and geopolitical challenge, global policy experts agree that AI cannot exist in a vacuum and must be synchronised with the global drive for decarbonization and regional cooperation.

Participating in a panel discussion on Tuesday at the ongoing AI India Impact Summit in the national capital, energy expert Arunabha Ghosh argued that the Artificial-Intelligence revolution and the Green revolution are two strands of the same "double helix."

Ghosh, the CEO of the Council on Energy Environment and Water (CEEW) said," I see this as a double helix. There is a digital revolution going on and a decarbonization revolution going on at different speeds with different momentums and different geographies."

He said that the two technologies should converge to save the planet.

"But the two will have to come together. These can't be parallel tracks. These have to come together. When they come together properly, then AI becomes one more general-purpose technology to impact another general-purpose technology, which, in my opinion, is climate action," Ghosh said.

Also Read

Panchayat Ministry highlights 'Sabhasaar' at India AI Impact Summit

Adoption, not innovation, biggest gap in AI rollout: Experts at AI Summit

Focus on governance, scale vital for building trusted AI: Industry leaders

Involving educational institutions at early stages necessary: Zoho's Vembu

Gnani.ai unveils 5B-parameter voice-to-voice AI model at India AI Summit

He highlighted how AI can optimize solar and wind integration into power grids, provide hyper-accurate flood forecasting, and increase agricultural resilience. However, he warned of the "duality" of AI -- noting that massive data centres are incredibly thirsty for power and water for cooling.

"The data centres are significantly contingent on power demand, on water demand, cooling demand, and you need to then develop this infrastructure that is ready for the future, not just leveraging the present," he said.

"So how do we then square these two circles? On one hand, the need for decarbonisation. On the other hand, extracting value through the digital and the AI revolution," he said.

Ghosh called for "intentionality," urging nations to transparently measure the energy and water footprints of their AI strategies rather than "leveraging the present" at the cost of the future.

"We should reject this false binary that either you have climate action or you have AI-driven ambition," Ghosh stated. "They have to come together."

Philip Thigo, Special Envoy from the Office of the President of Kenya, was also part of the panel discussion with Ghosh at the session "AI for India's Next Billion: Intergenerational Insights for Inclusive and Future-Ready Growth"at the Bharat Mandapam venue.

With only one per cent of global data centres located in Africa (and half of those in South Africa), Thigo argued that "sovereignty" doesn't mean every country needs to build everything.

"Everybody wants to invest in data centres, we can't currently, only one per cent of global data centres are in Africa, 50 per cent in South Africa. It just means some countries will not be able to do this," he said.

Thigo advocated for a "South-South cooperation," where neighbouring countries share infrastructure and value rather than duplicating expensive investments.

"It calls upon cooperation and sharing...it becomes interesting that global cooperation, and especially regional cooperation, but also South cooperation becomes important, where we have to kind of see where they share value and we can collaborate and not necessarily spend money on doing anything," Thigo said.

Former NITI Aayog CEO Amitabh Kant and UN Foundation VP Claire Melamed who were part of today's panel discussion.

While Kant focused on the structural inequality of AI data, Melamed addressed the "governance challenge," arguing that the lines of accountability between states, citizens, and tech giants have become blurred and ineffective.

Melamed rejected the notion that the tech industry can "grade its own homework," calling for independent oversight on how AI companies should be managed.

She dismissed the effectiveness of industry-led governing bodies, stating, "It never works. We know that time and time again."

Her primary concern she said was in ensuring citizens have a clear path to justice--or "redress"--when AI systems cause harm.

"We cannot have a situation where the behaviour and impact of companies is monitored and governed by bodies which are set up by those companies themselves," Melamed said.

Meanwhile, Amitabh Kant, warned about AI creating a "highly inequal society" and advocated for building a layer of public identity and accountability on top of it.

"Our digital ecosystem worked, because our models were open-sourced. My view is that there has to be a layer of digital public identity in AI, on top of which we should allow private sector to open and compete," Kant said.

More From This Section

Samsung previews Zero-Peeking Privacy control in Galaxy S26 series: Details

YouTube comments, descriptions missing amid ad-block crackdown: Report

India AI Impact Expo powerful convergence of ideas, innovation: PM Modi

AI offers opportunity for Indian firms to drive job transformation: Nasscom

Global South should make own AI models using local data, says Amitabh Kant

With the national capital witnessing a unique blend of technology and culture at the India AI Impact Summit 2026, a group of artists got together on Tuesday to highlight the transformative role of artificial intelligence in grassroots governance through a street play 'Sarpanch Ki Chaupal, AI Ka Kamaal', .

Organised under the aegis of the Ministry of Panchayati Raj, the cultural presentation underscored how AI is being integrated into the Panchayati Raj system to strengthen village-level democracy.

The AI summit that began on Monday brings together global leaders, policymakers, technology companies, innovators, and experts to deliberate on AI's potential in governance, innovation, and sustainable development.

Sushil Kumar Lohani, Additional Secretary, Ministry of Panchayati Raj, detailed how AI-powered solutions are improving the functioning of Gram Sabhas, the foundational decision-making bodies in villages.

"Gram Sabha is the most important platform in the Panchayati Raj system where the problems of the village are discussed, solutions are agreed upon, and financial matters such as budget allocations and schemes are reviewed," Lohani told ANI.

Also Read

Adoption, not innovation, biggest gap in AI rollout: Experts at AI Summit

Focus on governance, scale vital for building trusted AI: Industry leaders

Involving educational institutions at early stages necessary: Zoho's Vembu

Gnani.ai unveils 5B-parameter voice-to-voice AI model at India AI Summit

India AI Impact Expo powerful convergence of ideas, innovation: PM Modi

He acknowledged recurring concerns that discussions held during Gram Sabha meetings were not always properly recorded. To address this, the Ministry has developed an AI-based module named 'Sabhasaar'.

"Many times such complaints come that whatever was discussed in the Gram Sabha, proper recording of the same is not done, proper proceedings are not made. So we have used AI and developed a module called 'Sabhasaar'," he said.

"Through this, whenever a Gram Sabha is held in the village, we record all the proceedings. After recording, the video or audio is converted into text. The proceedings are finally prepared in the language in which people speak, which is very important for strengthening the Gram Sabha and expanding it," Lohani explained.

According to the official, the initiative reduces the workload on panchayat secretaries, who previously faced challenges in accurately documenting meetings. AI ensures that decisions are recorded precisely, minimising disputes over incomplete records.

The Ministry has also rolled out several digital applications as part of its broader automation efforts. The 'Panchayat Nirnay' app records meeting details, including attendance and discussions.

Meanwhile, the 'Meri Panchayat' app allows citizens to access information about budgets, projects, selected beneficiaries, available assets, and details of elected representatives directly on their mobile devices.

"We are doing many other activities on a regular basis and we do the rest based on the requirements of the people," Lohani said.

He further noted that AI-driven systems not only streamline administrative tasks but also help disseminate information about ongoing projects to the public more effectively.

Beyond governance, the summit also highlighted India's growing leadership in AI-driven climate solutions. With institutional innovations and strong multilateral partnerships, India now provides village-level weather forecasts, demonstrating how AI can serve as a critical tool in combating climate change, particularly for vulnerable communities in the Global South.

More From This Section

Samsung previews Zero-Peeking Privacy control in Galaxy S26 series: Details

YouTube comments, descriptions missing amid ad-block crackdown: Report

AI offers opportunity for Indian firms to drive job transformation: Nasscom

Global South should make own AI models using local data, says Amitabh Kant

Apple releases iOS 26.4 dev beta 1 for iPhones: What's new, how to update

Over the past 18 months, AI has moved from experimentation to everyday use across many organizations. Enterprises are increasingly relying on AI to handle routine, time-consuming tasks so their teams can focus on higher-value work. Unsurprisingly, AI adoption is now widespread, with most organizations using it in at least one business function.

One area where interest is growing rapidly is network management and monitoring. At the same time that AI is becoming more common, other technologies such as cloud computing are driving sharp increases in network traffic and complexity. As a result, managing modern networks has become far more demanding than it was just a few years ago. To keep up, IT teams are turning to AI and automation -- but how widely is AI-based network monitoring used today, and how is it evolving?

Reducing the burden on IT teams

Network management has always required constant attention: monitoring infrastructure health, detecting faults, and responding quickly to unusual behavior. While these tasks were once manageable through manual processes, growing network complexity and the rise in cyber threats have made traditional approaches increasingly inefficient. Many IT professionals now spend a large portion of their time on repetitive, manual activities such as firewall management or network provisioning.

AI helps relieve this pressure by automating large parts of network supervision. Machine-learning models can continuously analyze vast volumes of network data, spotting anomalies such as traffic spikes, suspicious access attempts, or patterns linked to known cyber threats. Crucially, this happens in real time, allowing teams to address issues before they escalate into outages or security incidents.

Beyond speed, AI also improves focus. Instead of forcing specialists to sift through endless logs and alerts, AI systems filter out background noise and highlight only what truly matters. This reduces false positives and enables IT teams to concentrate on genuine risks, including emerging threats that evolve too quickly for traditional rule-based tools to detect.

AI-driven monitoring also scales naturally with business needs. During peak usage periods or unexpected surges in demand, AI systems can expand their monitoring capabilities automatically, without requiring additional staff. This flexibility is becoming essential as networks grow larger and more dynamic.

Where adoption stands today

AI-enabled network monitoring is already gaining traction across many industries. Many networking professionals now view automation and AI as core components of modern network operations, rather than optional enhancements. On average, a significant share of network management tasks -- spanning design, deployment, maintenance, and troubleshooting -- are already automated, and this proportion continues to rise.

That said, adoption does not always translate into success. While many organizations are actively using AI features in their network tools and even training models on their own IT and security data, far fewer report fully successful outcomes. This gap highlights the practical challenges of moving from experimentation to mature, reliable AI-driven operations.

Two obstacles stand out. The first is data quality. Incomplete data, inconsistent formats, or poor documentation can severely limit the effectiveness of AI models. The second is skills. Many IT teams lack the in-house expertise needed to deploy, train, and manage AI-based networking tools, slowing progress and reducing confidence in the results.

Agentic AI: the next phase of network management

Despite these challenges, the direction of travel is clear. AI-based monitoring platforms are becoming a foundational element of network management, and the next phase is already emerging in the form of agentic AI.

Unlike traditional AI systems that focus on detection and recommendations, agentic AI goes a step further. These systems can not only identify anomalies or predict capacity issues, but also determine root causes and take action -- either autonomously or with minimal human approval. In effect, agentic AI is designed to analyze, decide, and act on its own, shifting networks closer to truly autonomous operations.

Industry analysts expect this approach to become increasingly common over the next few years, particularly in large-scale and highly complex network environments. However, for most organizations, reaching this stage requires a phased approach.

Today, the priority is to implement AI-based monitoring solutions that integrate well with existing infrastructure and to ensure IT teams are properly trained to use them. Once organizations are confident in their data quality and trust the reliability of AI-driven insights, they can gradually introduce more autonomous capabilities. Over time, this evolution toward agentic AI promises to significantly reduce operational workloads for IT staff while delivering faster, more efficient, and more resilient network management.

Bromberg stressed that the new AI tool will "democratize" game development for non-coders while boosting productivity for all users. He said, "Our goal is to remove as much friction from the creative process as possible." The CEO also hinted at a future where millions more people would be creating interactive entertainment using these AI-enabled development tools, with Unity leading this revolution.

The Unity AI assistant will use large language models from OpenAI and Meta (GPT and Llama, respectively) to answer user queries, generate code, and perform agentic actions. The tool will also employ a mix of first-party and partner models to create and refine assets. Its partners include Scenario (which is trained on Stable Diffusion, FLUX, Bria, and GPT-Image foundation models) and Layer AI (based on Stable Diffusion and FLUX foundation models).

Building AI agents is the new gold rush. But every developer knows the biggest bottleneck: getting the AI to actually communicate to your data. Today, travel giant Agoda is tackling this problem head-on. They have officially launched APIAgent, an open-source tool designed to turn any REST or GraphQL API into a Model Context Protocol (MCP) server with 0 code and 0 deployments.

Until recently, if you wanted your AI agent to check flight prices or look up a database, you had to write a custom tool. When Anthropic released the Model Context Protocol (MCP), it created a standard way for Large Language Models (LLMs) to connect to external tools.

However, even with MCP, the workflow is tedious. A developer must:

Agoda team calls this the 'integration tax.' For a company with 1000s of internal APIs, writing 1000s of MCP servers is not realistic. APIAgent is their answer to this scaling problem.

APIAgent is a universal MCP server. Instead of writing custom logic for every API, you use APIAgent as a proxy. It sits between your LLM (like Claude or GPT-4) and your existing APIs.

The tool is built on a specific technical stack:

The 'magic' lies in its ability to understand API documentation. You provide a definition of your API -- using an OpenAPI specification for REST or a schema for GraphQL -- and APIAgent handles the rest.

The architecture is straightforward. APIAgent acts as a gateway. When a user asks an AI agent a question, the flow looks like this:

This system uses Dynamic Tool Discovery. You can point APIAgent at any URL, and it automatically generates the necessary tools for the LLM without manual mapping.

One of the key features is Recipe Learning. When a complex natural language query successfully executes, APIAgent can extract the trace and save it as a 'Recipe.'

BENGALURU: Infosys Ltd chairman Nandan Nilekani sought to reassure investors that the rise of generative AI (GenAI) is expanding business opportunities for the company, even as anxiety triggered by the launch of new AI tools has weighed on the stock, sending shares of the India's second-largest information technology (IT) services firm down 14% between 1 January and 17 February.

"My view is that there is no opportunity gap. Opportunity is bigger than before," Nilekani told analysts in his brief 20-minute talk on Tech transitions - Why is the AI transition different? as part of Infosys's first analyst day event on Tuesday.

"It is not an opportunity risk but an execution risk," said Nilekani, calling the current phase "tumultuous times".

Nilekani argued that a lag between rapid technological change and adoption by large Fortune 500 companies creates what he described as "deployment gap", allowing firms such as Infosys to help large banks, retailers, and companies across industries implement AI-led tools.

"Fundamentally, this is root and branch surgery, the way business is done," Nilekani said, describing how AI disruption is making companies re-examine the way they do business.

"Modernizing of legacy systems cannot be deferred anymore. That is over," said Nilekani, who, along with his six friends, co-founded Infosys in 1981, and also reiterated new areas where the IT services company could benefit.

In a positive signal for investors, Infosys disclosed for the first time that its AI-related revenue was $280.4 million, accounting for about 5.5% of its $5.1 billion revenue in the October-December quarter (Q3FY26).

On 17 December, Infosys's larger rival, Tata Consultancy Services Ltd, said its annualized revenue from AI-related business totalled $1.5 billion at the end of the September quarter (Q2).

Besides Nilekani and chief executive Salil Parekh, a dozen senior leaders from the company briefed analysts on how Infosys is navigating a period that has polarized views on the future of the IT services industry, between naysayers predicting disruption and optimists expecting growth.

Earlier this month, tech stocks, including shares of homegrown IT firms, fell between 5% and 8% after AI company Anthropic launched productivity tools for its Claude Cowork platform that can automate legal work and other functions inside large businesses, raising concerns that such tools could eat into the revenue of the $283 billion Indian IT industry.

"For the typical IT services the near to medium term still isn't disrupted. Even if the projects themselves may plateau relatively, the productivity gains will drive margins - as the negotiations with vendors will lag in terms of AI linked discounting compared to the gains actually realised," Bernstein analysts Venugopal Garre and Nikhil Arela wrote in a note, AI vs Human: Code Red and the Programmers' Endgame, dated 16 February.

"The longer run case is unpredictable and hazy, though it likely points at IT services becoming low margin commoditized industry. But that's quite far off as well as dependent on several factors including AI's continued refinements," the note said.

"The current market reactions perfectly capture two fallacies - the slippery slope and the bandwagon, and once the dust of the AI storm settles, we perceive a correction to emerge. Till then, statements like replacement of all white collar jobs will emerge and work wonders for valuations of AI centric companies," it added.

Earlier this week, Microsoft AI chief executive Mustafa Suleyman warned that AI tools could replace most white-collar work in the next 12 to 18 months.

In today's digital landscape, businesses must adapt to the evolving search environment, which increasingly includes AI-driven platforms. SEOForge is an AI Visibility Intelligence Platform designed to help businesses monitor and enhance their presence across various AI search models, including ChatGPT, Perplexity, Claude, Gemini, and Grok.

SEOForge is tailored for businesses aiming to enhance their digital presence across both traditional and AI-driven search platforms. It's particularly beneficial for agencies and growing brands seeking to:

SEOForge offers several pricing tiers to accommodate different business needs:

Each plan includes a 7-day free trial with full access to all features, allowing businesses to evaluate the platform's suitability before committing.

SEOForge offers a comprehensive suite of tools designed to enhance your brand's visibility across both traditional and AI-driven search platforms. Its features cater to businesses aiming to improve their digital presence, optimize content creation, and gain competitive insights. The tiered pricing structure provides flexibility to choose a plan that aligns with your business objectives and budget. By leveraging SEOForge, businesses can navigate the complexities of the evolving search landscape and drive sustainable growth.

Welcome to a this week's Tech Tuesday where we explore the cutting edge of Voice-First AI Tools. In 2026, voice technology has transcended basic command prompts to become a sophisticated, emotionally intelligent interface, reshaping how businesses connect with customers and streamline operations. From AI agents that conduct sales calls indistinguishable from humans to empathic systems that detect subtle vocal cues, these tools are redefining real-time interaction. We'll dive into the platforms that power ultra-low-latency conversations, clone voices for personalized content, and even act as specialized medical assistants, demonstrating how voice AI is moving beyond automation to become a truly collaborative and perceptive partner in the modern enterprise.

These tools act as "digital receptionists" or "outbound agents," capable of holding full, human-like phone conversations.

Vapi is a developer-first platform designed to build and scale advanced voice AI agents with unprecedented speed. By 2026, it has become the backbone for modern phone operations, enabling businesses to deploy conversational agents that feel truly human. The platform focuses on eliminating the technical barriers of voice technology, providing a high-performance infrastructure that handles everything from audio streaming to complex logic orchestration.

Features include Sub-500ms Latency, which ensures fluid, real-time conversations without the awkward pauses common in older systems. The platform offers Multilingual Support for over 100 languages and an API-Native Architecture that allows for deep integration with CRMs and existing telephony stacks. In 2026, it also features Automated Testing suites to identify hallucination risks and Tool Calling capabilities that enable agents to fetch data and perform actions, such as booking appointments, directly during a call.

Best for software engineers and enterprise operations teams who need to build robust inbound or outbound voice products. It is the ideal choice for companies in logistics, healthcare, and retail that require scalable, PCI-compliant voice agents to handle millions of daily interactions. Trail-blazing startups and Fortune 500 firms use Vapi to slash engineering hours and deploy secure, human-like voice experiences that drive tangible business results.

Retell AI is a next-generation voice-first platform that specializes in human-standard call automation through proprietary AI orchestration. By 2026, it has set the industry benchmark for "natural" interactions, moving beyond intent mapping to a full LLM-based system that understands context and nuance. The platform is designed to handle high-stakes customer conversations with a level of fluidity that makes AI agents virtually indistinguishable from their human counterparts during inbound and outbound operations.

Features include Ultra-Realistic Voices refined through human-guided training and an industry-leading 600ms Latency that keeps conversations flowing without interruption. The platform offers a Drag-and-Drop Agentic Framework for building complex call flows, alongside Real-Time Function Calling that allows agents to book appointments, process payments, and update CRMs in mid-sentence. In 2026, it also features Streaming RAG, which ensures every agent is grounded in the company's latest website content or knowledge base to provide accurate, up-to-the-minute answers.

Best for large sales and support teams in industries like healthcare, logistics, and home services that need to scale their call center operations without sacrificing quality. It is the ideal choice for businesses requiring a "turnkey" AI receptionist or lead qualification system that can handle edge cases and unexpected inputs gracefully. Leading organizations like Pine Park Health and SWTCH use Retell AI to cut support costs by over 50% while significantly increasing customer satisfaction scores through instant, 24/7 responsiveness.

Air AI is a high-performance conversational platform designed to conduct long-form, autonomous sales calls that are virtually indistinguishable from a human. By 2026, it has specialized in the "outbound" sales niche, providing businesses with a digital workforce capable of executing complex 10-to-40-minute discovery and closing calls. Unlike basic voice bots, Air AI is built with a deep understanding of sales psychology, allowing it to navigate objections, build rapport, and drive prospects toward a specific commitment or purchase.

Features include Infinite Scaling, which allows companies to move from zero to millions of concurrent calls without hiring a single additional representative. The platform offers Advanced Sentiment Analysis to read a prospect's emotional state and adjust the sales pitch in real-time, alongside a Full Telephony Stack that handles outbound dialing and call routing natively. In 2026, it also features Dynamic Scripting, where the AI learns from successful conversions to automatically refine its pitch and objection-handling techniques across the entire organization.

Best for high-volume sales organizations, real estate firms, and insurance agencies that need to qualify thousands of leads or conduct outbound cold calls at massive scale. It is the ideal choice for businesses looking to automate the "top of the funnel" while maintaining the quality of a top-performing human closer. Entrepreneurs and enterprises use Air AI to replace expensive outsourced call centers with a 24/7 autonomous sales force that never gets tired, never misses a follow-up, and consistently delivers a perfect brand message.

Dialora AI is a comprehensive AI voice platform designed to automate sales, support, and outreach for businesses of all sizes. By 2026, it has specialized in high-conversion "digital reps" that handle both inbound and outbound communication 24/7. Built to act as a tireless member of the team, Dialora bridges the gap between lead generation and final commitment by ensuring that no customer inquiry goes unanswered and every qualified prospect is booked directly into a company's calendar.

Features include Smart Outbound Campaigns, which allow users to upload lead lists and schedule personalized calls across multiple time zones automatically. The platform's Answer & Qualify system handles inbound traffic by screening calls and capturing data, while its native Cal.com and Google Calendar integrations facilitate seamless appointment setting. In 2026, it also boasts a diverse Multilingual Voice Library and advanced Call Sentiment Analysis, providing business owners with deep insights into customer moods and conversation quality through real-time transcripts and recordings.

Best for solo founders, small business owners, and sales agencies in industries like real estate, healthcare, and retail that need to scale their outreach without increasing headcount. It is the ideal choice for businesses looking for a "zero-code" solution that can be trained on their own documents or website to provide accurate, on-brand answers. Companies across 30+ countries use Dialora.ai to eliminate manual follow-ups and transform cold leads into closed deals through persistent, professional AI-driven communication.

Vozexo is a specialized AI voice agent platform meticulously engineered for the home services and field operations industries, including plumbing, HVAC, and electrical businesses. By 2026, it has become a critical operational tool for trade companies, functioning as a 24/7 intelligent answering service that ensures no emergency call or high-value lead is ever missed. The platform is designed to handle the specific pressures of service-based businesses, where a fast response to an urgent request -- like a burst pipe or a failing furnace -- often makes the difference between winning a job and losing it to a competitor.

Features include Intelligent Service Intake, which allows the agent to identify the nature of a caller's issue, gather their address, and verify technician availability in real-time. The system provides Automatic Appointment Booking by integrating directly with industry-standard field service management (FSM) software like Housecall Pro, ServiceTitan, and Jobber to synchronize calendars and job logs. In 2026, Vozexo also features Multi-Speaker Diarisation and sophisticated Interruption Handling, allowing it to navigate complex, emotionally charged emergency calls with a calm, professional, and human-like tone that builds immediate trust with the homeowner.

Best for small to medium-sized home service contractors and larger dispatch centers that experience seasonal volume spikes or struggle with after-hours call coverage. It is the ideal choice for business owners who want to eliminate the cost of traditional answering services while significantly boosting their lead conversion rates through instant responsiveness. Field operations teams use Vozexo to automate routine scheduling and triage, allowing their human dispatchers to focus on managing technician routes and complex logistics rather than answering basic FAQs.

These tools focus on the "vibe" and emotional connection of the voice, moving beyond just words to understanding tone and sentiment.

Hume AI (EVC) is a revolutionary empathic voice engine built on the world's first foundational speech-to-speech model. By 2026, it has redefined human-AI interaction by allowing agents to move beyond words to understand and react to the emotional prosody of a user's voice. Unlike standard text-to-speech tools, the Empathic Voice Interface (EVI) is grounded in decades of emotion science, enabling it to detect subtle vocal cues like hesitation, excitement, or frustration to deliver a response that is contextually and emotionally appropriate.

Features include EVI 3, an instructible foundational model that offers human-standard realism and exceptionally low latency for fluid, real-time dialogue. The platform provides Expression Measurement tools that analyze over 50 dimensions of emotional expression, alongside Octave TTS 2, a next-generation multilingual voice engine that can generate highly expressive speech in multiple languages. In 2026, it also features a robust Custom Model API, allowing developers to fine-tune agents for specific professional roles, from compassionate healthcare companions to persuasive sales specialists.

Best for developers, product designers, and enterprises in healthcare, education, and customer experience who want to build AI that truly "connects" with users. It is the ideal choice for organizations creating mental health support tools, conversational learning platforms, and high-empathy customer service bots. Global trade leaders and creative studios use Hume AI to transform robotic interfaces into emotionally intelligent partners that enhance well-being and drive deeper user engagement through genuine vocal understanding.

OpenAI Advanced Voice Mode is a multimodal, low-latency audio interface powered by the GPT-4o architecture, designed to provide seamless, human-like verbal interaction. By 2026, it has transitioned from a creative novelty into a robust business tool, capable of perceiving emotional tone, handling rapid-fire interruptions, and responding with sub-second speed. The system operates natively across audio, vision, and text, allowing for a conversational flow that feels intuitive and grounded in real-world context rather than a series of disconnected prompts.

Features include Native Multimodality, which allows the AI to "see" a user's screen or environment during a call to provide real-time visual assistance, and Natural Turn-Taking, which enables fluid interruptions and corrections. The platform offers Custom Instructions for Voice, allowing businesses to define a specific persona, tone, and regional accent for their agents. In 2026, it also features Enterprise-Grade Privacy Controls, ensuring that voice interactions are not used for model training and are protected by the same security standards as the rest of the OpenAI workspace suite.

Best for customer experience designers, educators, and creative professionals who need an AI collaborator that can understand non-verbal cues and emotional context. It is the ideal choice for businesses building immersive language learning apps, interactive technical support, and high-engagement brand ambassadors. Global organizations use OpenAI Advanced Voice Mode to replace robotic IVR systems with sophisticated, empathetic assistants that can coach employees, guide customers through complex setups, and provide instant expert advice.

Ultravox AI is a research-led voice platform built on the belief that AI should be "speech native" rather than relying on slow text-to-speech translations. By 2026, it has distinguished itself by training foundational models that process audio directly, capturing the "messy" but essential paralinguistic signals -- like tone, cadence, and pitch -- that are usually lost in transcription. This first-principles approach allows Ultravox to deliver some of the world's fastest and most contextually aware voice agents, capable of moving at the rapid pace of human thought and progress.

Features include Ultravox v0.7, a state-of-the-art model that scores an industry-leading 97% on thinking benchmarks while maintaining near-instant response times. The platform's Dynamic Endpointing (UltraVAD) uses neural modeling to predict turn-taking, distinguishing between a thoughtful pause and the end of a sentence to prevent awkward interruptions. In 2026, it also offers a Unified Inference Stack and robust Developer SDKs across web and mobile, ensuring that businesses can deploy low-latency agents on their own dedicated infrastructure without waiting on shared external pools.

Best for technical founders, AI research teams, and enterprises that require "thinking" voice agents for complex, high-stakes interactions. It is the ideal choice for companies building advanced customer support systems, interactive gaming characters, or professional training simulations where nuances in speech are critical. Growth-stage companies and agencies use Ultravox to build and scale voice products that are fast, accurate, and capable of understanding the subtle human signals that make conversations feel authentic.

Perfect for creators and businesses that need high-quality narration or consistent brand voices.

ElevenLabs is the industry-leading audio research and creative platform recognized for its ultra-realistic voice synthesis and cloning capabilities. By 2026, it has expanded into two distinct pillars: a Creative Platform for generating high-fidelity speech, music, and sound effects, and an Agents Platform for deploying intelligent, conversational bots. The platform is designed to provide emotional depth and expressive delivery that sets the standard for how AI "sounds," allowing businesses to build experiences that feel less like software and more like a natural human interaction.

Features include Zero Retention Mode for HIPAA-eligible security and ElevenReader, a mobile app that turns any document or email into a narrated experience in a cloned voice. The Agents Platform offers out-of-the-box integrations with tools like Salesforce, Zendesk, and Twilio, enabling 24/7 automated scheduling and support that is grounded in a company's specific data. In 2026, it also features Dubbing Studio, which automates high-quality video translation while preserving the original speaker's vocal characteristics across 29+ supported languages.

Best for enterprise leaders, content creators, and game developers who require the highest level of vocal realism and emotional nuance for their brand. It is the ideal choice for organizations looking to scale multilingual customer support or produce professional-grade audiobooks and localized marketing content without the cost of traditional studios. Global giants like Disney, Klarna, and Revolut use ElevenLabs to reduce resolution times and production costs while delivering a consistent, "iconic" voice presence to millions of users worldwide.

Speechmatics is an industry-leading speech intelligence platform renowned for its high-accuracy, inclusive speech recognition technology that works regardless of accent or environment. By 2026, it has solidified its position as a foundational layer for enterprise-grade voice products, offering a robust API that handles complex, real-world audio with ease. The platform is designed to provide "unbeatable" accuracy across 55+ languages, ensuring that businesses can capture every word of their global communications, whether in live broadcasts, medical consultations, or high-volume contact centers.

Features include Sub-500ms Real-Time Transcription and Live Translation, allowing businesses to bridge language barriers as they happen. The platform's Enhanced Model provides best-in-class accuracy for proper nouns and industry-specific terms through a Custom Dictionary, while its Real-Time Diarization identifies and labels multiple speakers in a single conversation. In 2026, it also features a specialized Medical Model for clinical transcription and Low-Latency Text-to-Speech (sub-150ms), enabling the creation of responsive, speaker-aware voice agents that understand exactly who is talking.

Best for enterprise developers, media organizations, and healthcare providers who require a reliable, scalable speech-to-text foundation. It is the ideal choice for companies operating in noisy or multicultural environments where traditional ASR solutions often fail due to accent or dialect variations. Global leaders in captioning, contact center analytics, and EdTech use Speechmatics to ensure 99% accuracy in their automated workflows, reducing manual documentation time and improving accessibility for users worldwide.

PlayHT is a leading AI voice platform that provides ultra-realistic text-to-speech and high-fidelity voice cloning for professional content creation. By 2026, it has specialized in "dialog-enabled" synthesis, allowing users to generate complex, multi-turn conversations between different AI voices within a single project. The platform is designed to move beyond simple narration, offering a sophisticated online studio where creators can fine-tune every nuance of a performance -- from emotional inflections and pitch to custom pronunciations of technical industry terms.

Features include PlayAI, a robust Voice Generation API optimized for real-time applications like conversational chatbots, live streaming, and gaming. The platform offers a library of over 800 natural-sounding voices across 142 languages, complete with local accents and diverse speaking styles ranging from "Persuasive Sales" to "Supportive Medical." In 2026, it also features Cross-Language Voice Cloning, which preserves a speaker's unique vocal identity and native accent even when dubbing their content into a completely different language.

Best for podcasters, eLearning developers, and YouTube creators who need to produce high-quality narrated content at scale without the overhead of traditional recording studios. It is the ideal choice for businesses looking to localise their marketing videos and training materials for a global audience through instant, high-fidelity dubbing. Organizations use PlayHT to shorten production cycles for audiobooks and explainer videos, transforming written scripts into professional-grade audio in a matter of seconds.

Sully AI is a comprehensive suite of autonomous "AI employees" designed specifically to transform the operational and clinical efficiency of healthcare organizations. By 2026, it has become a leader in reducing physician burnout by deploying specialized agents that handle the high-friction tasks of medical practice, from front-desk reception to complex clinical documentation. The platform operates as a secure, integrated ecosystem that works 24/7 alongside human clinicians to ensure that patient care remains the primary focus of every visit.

Features include the AI Scribe, which uses advanced voice recognition to automatically capture and structure patient conversations into clean, HIPAA-compliant SOAP notes, and the AI Receptionist, capable of handling patient calls and scheduling with natural conversation. The platform's AI Nurse handles intake and symptom collection before visits, while the AI Medical Coder extracts ICD-10 codes from notes to accelerate reimbursements. In 2026, it also features Deep EHR Interoperability with major systems like Epic, Cerner, and Athenahealth, allowing its agents to prep charts and update records without manual data entry.

Best for hospitals, private practices, and multi-specialty clinics that are struggling with administrative waste and clinical documentation overload. It is the ideal choice for healthcare leaders who want to catch errors human providers might miss and generate "clinician-ready" insights up to 6x faster than traditional alternatives. Over 100 healthcare organizations nationwide use Sully AI to slash burnout by 80% and increase efficiency, giving doctors nearly three extra hours of productive time back into their workdays.

Klariqo is an AI-powered phone and website assistant designed to act as a tireless, 24/7 receptionist for small businesses. By 2026, it has focused on "stopping the bleeding" of missed calls, providing a human-like voice interface that answers instantly, qualifies leads, and books jobs without the overhead of a traditional call center. The platform is built for speed and simplicity, allowing business owners to deploy a sophisticated voice agent in as little as three minutes, ensuring that no customer inquiry is ever left to go to voicemail.

Features include a Sub-Second Voice Engine with 0.4s response latency, which facilitates natural conversations that understand context and handle interruptions seamlessly. The platform offers CRM Integration with over 50 tools like Jobber and ServiceTitan, alongside The Truth Vault, which records and transcribes every call for quality monitoring. In 2026, it also features the "Big Office" Sound, which adds subtle background cues like keyboard clicks to make the AI sound like a professional team, and a Smart Dispatch system that texts the business owner the details of newly booked jobs and high-value leads.

Best for local service providers, real estate agents, and restaurant owners who lose significant revenue from missed calls and after-hours inquiries. It is the ideal choice for small business owners looking for a low-cost, high-reliability alternative to expensive human receptionists or scripted call centers. Companies in industries like HVAC, dental, and SaaS use Klariqo to greet repeat customers by name, answer specific business FAQs, and maintain a professional presence that never takes a sick day or a break.