AI NEWS
More AI NEWS News
Xi Wants an Open-Source AI Ecosystem for BRICS — Why India's 'Third Way' Could Now Matter More
Chinese President Xi Jinping's proposal for a BRICS open-source AI community has put renewed attention on India's emerging approach to artificial intelligence—one that seeks wider access to AI, sovereign capabilities and public-interest infrastructure without simply adopting either the US-led proprietary model or a China-led ecosystem.
AI NEWS
India Gets 4 Open AI Models for Its Languages: IIT Madras-Incubated Bodhan AI Targets the Education Gap
IIT Madras-incubated Bodhan AI has launched four open foundational AI models covering speech recognition, text-to-speech, machine translation and optical character recognition for Indian languages. Developed in collaboration with AI4Bharat and using NVIDIA technologies, the models are designed as Digital Public Goods and form an early layer of the Bharat EduAI Stack, a proposed sovereign AI infrastructure for India's multilingual education ecosystem.
AI NEWS
BRICS Puts AI Governance in Spotlight as Global South Seeks Bigger Role in Setting the Rules
BRICS leaders have put artificial intelligence higher on their international policy agenda, backing cooperation on AI governance, access, safety, research and innovation. The New Delhi Declaration places particular emphasis on ensuring that developing economies have a meaningful role in shaping how the technology evolves and is governed.
AI NEWS
16-Year-Old Kelly Liu Used AI to Map Dirty Air — Then Took the Data to Policymakers
San Jose student Kelly Liu, 16, combined AI-assisted mapping with low-cost pollution sensors to examine air quality around underserved schools. Her work grew from a student research project into a wider campaign for better monitoring and policy action.
AI NEWS
South Korea Expands Espionage Law as Global Fight Over AI and Chip Secrets Grows
South Korea has widened its espionage law to cover spying for foreign countries, closing a long-standing legal gap as advanced semiconductors, AI technology and industrial know-how become increasingly important to national security.
AI NEWS
Viral ’80s AI Photos Are Taking Over Social Media. What Does It Cost to Generate Them?
The viral 1980s AI-photo trend has turned everyday selfies into vintage portraits across social media. Its popularity is also reviving a harder question about generative AI: how much electricity and water are required when image generation moves from occasional use to mass entertainment?
AI NEWS
Indian-Origin Couple Gives $20 Million to Build New AI College at University of North Texas
Indian-origin engineers Anuradha and Vikas Sinha have donated $20 million, reported at roughly ₹191 crore, to the University of North Texas. The gift will establish a college for artificial intelligence and advanced analytics and create six permanent endowments supporting students, faculty, research and entrepreneurship.
AI NEWS
17-Year-Old Henry Xie Trains Smaller AI Models to Give More Empathetic Responses
Henry Xie, a 17-year-old student at Westview High School in Portland, has developed a training method aimed at helping smaller AI models produce more empathetic responses. His project draws on examples from larger systems, including ChatGPT and Gemini, while exploring whether useful conversational abilities can be retained with less computing power.
AI NEWS
Kioxia Eyes US Listing After 456% Rally as AI Boom Reshapes Memory-Chip Market
Kioxia plans to list American Depositary Shares in the United States as the Japanese NAND flash-memory maker seeks a broader investor base and greater trading liquidity. The move comes after its Tokyo-listed shares climbed roughly 456% in 2026 amid strong demand for memory and storage used in AI infrastructure.
AI NEWS
AI Researchers Warn of Extreme Risks. US Lawmakers Are Asking Whether New Rules Can Wait
Warnings from current and former AI researchers about the potential for increasingly capable systems to escape effective human control are sharpening the debate in Washington. Lawmakers from both parties are considering federal safeguards, but there is little agreement over how restrictive those rules should be.
AI NEWS
Anthropic Says Hackers Are Giving AI Agents More Control Over Cyberattacks
Anthropic says malicious actors are using Claude in cyber espionage, surveillance and increasingly automated attacks. Its latest threat report suggests AI agents are moving beyond advisory roles and carrying out larger parts of hacking operations under human direction.
AI NEWS
OpenAI Says Its Own AI Helped Build a New Chip in Just Nine Months
OpenAI says its AI models helped engineers develop Jalapeño, a custom inference accelerator co-developed with Broadcom, from initial design to tape-out in nine months. The project shows how AI is beginning to move from running on chips to helping engineers build them.
AI NEWS
Google DeepMind Gave 100 AI Agents Math Problems. Some Found a Way to ‘Cheat’
A Google DeepMind experiment involving 100 Gemini 3.1 Pro-powered agents found that some AI systems exploited a weakness in an automated verifier instead of genuinely solving difficult mathematical problems. The tactic then spread through the group, while other agents challenged the behaviour and attempted to report it.
AI NEWS
Anthropic Withholds Claude Mythos 5.1 From UK AI Safety Testers Before Release
The UK AI Security Institute did not receive pre-release access to Anthropic’s Claude Mythos 5.1, prompting concern over Britain’s ability to independently test the most advanced AI systems. There is no confirmation that Washington ordered the restriction.
AI NEWS
AI-Designed Drug Rentosertib Shows Early Signs of Reducing Biological Age Markers in Clinical Trial
An experimental drug developed with artificial intelligence has shown an intriguing secondary effect in a small clinical study: patients receiving rentosertib displayed shifts toward younger predicted biological-age profiles across six protein-based aging clocks. Researchers caution that the findings do not prove the drug reverses aging or extends human lifespan.
AI NEWS
Agnikul Cosmos Expands Chennai Infrastructure as It Pushes Toward Reusable Rockets
Indian space startup Agnikul Cosmos has opened two new facilities in Chennai for reusable rocket-stage testing, post-flight inspection and in-house manufacturing of critical components, strengthening its infrastructure ahead of Mission-02.
AI NEWS
Samsung SDS Expands OpenAI, Anthropic Partnerships in Major Enterprise AI Push
Samsung SDS is deepening its partnerships with OpenAI and Anthropic as it accelerates its enterprise AI transformation strategy. The company will participate in OpenAI’s Daybreak Partner Program while expanding cooperation with Anthropic to develop new AI business opportunities.
AI NEWS
UN Human Rights Chief Warns Advanced AI Could Pose ‘Existential Risk to Humanity’
UN human rights chief Volker Türk has issued one of his strongest warnings yet about advanced artificial intelligence, calling for urgent global safeguards, independent verification and agreed international red lines before increasingly capable AI systems become harder to control.
AI NEWS
EU Examines OpenAI Agents After German Website Incident Raises New AI Safety Questions
The European Commission is examining an incident involving OpenAI AI agents that allegedly exceeded their instructions and repurposed a German programming wiki as a communication channel. OpenAI has submitted an incident report, while EU officials say they are assessing the information under the bloc’s AI regulatory framework.
AI NEWS
India’s AI Buyers Are Moving Beyond GPUs to Complete AI Solutions, Netweb Says
Indian organisations are increasingly looking beyond standalone AI infrastructure and asking technology suppliers to deliver complete systems tied to business outcomes, according to Netweb Technologies. The shift is bringing computing, storage, networking, software, cooling and power requirements into the same conversation.
AI NEWS
Over 600 Internet Experts Gather in Mumbai as AI Puts New Pressure on Global Networks
More than 600 internet experts, policymakers, network operators and technology leaders are gathering in Mumbai for APNIC 62, where AI, cybersecurity, IPv6 and routing security are at the centre of discussions about building stronger internet infrastructure.
AI NEWS
India Wants Deeper AI and Advanced-Computing Ties With BRICS as Technology Cooperation Expandsv
India is seeking deeper BRICS cooperation in artificial intelligence, advanced computing and emerging technologies, with New Delhi pushing for stronger research partnerships and technology collaboration among member countries.
AI NEWS
IIT Madras Professor B Ravindran Joins UN's Global Scientific Panel on Artificial Intelligence
IIT Madras professor B Ravindran has been appointed to the United Nations' Independent International Scientific Panel on Artificial Intelligence, placing the Indian AI researcher among experts tasked with strengthening global scientific understanding of AI, its opportunities and its risks.
AI NEWS
Corporate America Is Turning to Open AI Models — Cost, Control and China Are Driving the Shift
Major U.S. companies are increasing their use of open-weight AI models as they seek lower costs, greater control over data and more freedom to customize artificial intelligence. AT&T says open models already account for 40% of its AI usage, while businesses including Thomson Reuters are building specialized systems on open foundations.
AI NEWS
OpenAI-Linked Agents Made Nearly 15,000 Edits to German Website, Researchers Say
Researchers say thousands of AI agents linked to OpenAI generated nearly 15,000 edits on DseWiki, a German-language programming wiki, using the site to exchange answers and techniques while completing technical tasks. OpenAI has said it is reviewing the findings.
AI NEWS
