Journalism begins where hype ends
,,
I do not fear computers. I fear the lack of them."
—Isaac Asimov
I do not fear computers. I fear the lack of them."
—Isaac Asimov
OpenAI has launched GPT-6 Astra, its latest frontier AI model, claiming unprecedented capabilities across coding, mathematics, cybersecurity, science, computer use and personalized agentic task execution.
Matt Clifford joins Anthropic to lead its international government engagement, while his decision to remain Chair of the UK’s ARIA has sparked debate over potential conflicts of interest.
How we think about the “wiki incident,” where our agents wrote to several internet sites: it’s past time for us to define standards for when and how we share misalignment incidents, not just misalignment properties of our models.
Historically, we have treated misalignment
GPT-6 Astra is now available to all Pro, Enterprise, and Business Premium users in ChatGPT Work and Codex. It's also live in the API.
It might take a few days to roll out to our Plus and Business users. Thank you for your patience.
In Chat, GPT-6 Astra powers GPT-6 Pro, now available to all Pro, Business, and Enterprise users.
Checking that a major mathematical proof is correct can take years. Formalization—converting the mathematical reasoning into a form computer proof assistants like Lean can verify—can help.
Last month, Claude completed the first formalized proof of Fermat’s Last Theorem, one of
Huge congrats to our friends at @wayve_ai and @Uber.
Wayve’s frontier AI, trained on NVIDIA infrastructure and running on NVIDIA DRIVE AGX accelerated compute, is now taking passengers through the streets of London.
Let’s ride. 🚘🇬🇧
London, meet your newest way to ride. 🇬🇧
Autonomous rides powered by @wayve_ai are now on Uber. British-built AI. London roads. Same Uber app.
Off we go.
US lawmakers from California, New York and Illinois are calling on leading AI companies to establish an independently verified “Mutually Agreed Pacing” framework to coordinate the development of frontier AI models until safety and alignment research can keep pace.
US Senator Bernie Sanders has introduced the Ban Artificial Superintelligence Act, which would prohibit the development and deployment of superintelligent AI and pause advanced AI development until federal safety regulations are established.
The company said Quasar 438B scored 43 on Artificial Analysis’ Intelligence Index, ahead of Mistral Medium 3.5 at 30, NVIDIA Nemotron 3 Ultra at 38 and Inkling at 42.
OpenAI has launched GPT-6 Astra, its latest frontier AI model, claiming unprecedented capabilities across coding, mathematics, cybersecurity, science, computer use and personalized agentic task execution.
India and the US explored deeper cooperation in AI, semiconductors and data centers during a bilateral meeting at the G20 Innovation Ministerial.
Talking to AI FrontPage, Prof. Sanjeev Singh, Dean of the Faculty of Technology at Delhi University, said students currently lack adequate access to stable AI infrastructure, and the initiative will help them understand AI platforms, their functions and real-world use cases.
“As Playful as Humanity, As AI Can Never Be”– why these activists chose comedy, not rage, to warn about Rogue AI.
After the inaugural UN Global Dialogue on AI Governance in Geneva, PauseAI’s global chief tells AI FrontPage that the power to pause frontier AI already exists, and explains why he believes Anthropic built a weapon of mass destruction in the cyber realm.
The World Artificial Intelligence Conference in Shanghai closed with 20.36 billion yuan in intended procurement deals, a new intergovernmental AI body signed by 29 countries, and Xi Jinping’s first in-person address a coordinated pitch to the Global South weeks after Washington launched its own AI alliance.
According to a new Concordia AI report, China is shifting its AI safety focus beyond AI-generated content toward governing increasingly autonomous AI systems and the risks they pose to cybersecurity, society, and critical sectors.
India has 591 million gamers, most of them outside the metros — and platforms like STAN now lean on AI to keep tens of thousands of live voice rooms safe. In an exclusive interview, STAN CEO Parth Chadha explains why AI sits at the core of the platform, where it still falls short on moderation, and how it is helping small-town creators earn.
While watermarking is viewed as a safeguard for tracing AI-generated content, research presented at ICML 2026 shows that building watermarks that are secure, robust and stealthy at the same time remains a fundamental challenge.
A profile piece on British mathematician Alan Turing who presented a landmark paper “Can Machines Think?” that paved the way for artificial intelligence.
Andrew stressed that the aim is to help developers prioritise the skills they need to learn and help employers identify skilled developers.
India is less exposed to AI-driven job risks than many other countries because a large share of its workforce is engaged in physical or mechanical tasks, according to Goldman Sachs Chief India Economist Santanu Sengupta. Construction and retail trade together account for about 40% of the workforce.
The report highlights how AI-driven automation is putting entry-level jobs under growing pressure, raising concerns about opportunities for young and early-career workers amid weak economic growth and diminished job creation.
Singapore’s prime minister used his National Day address to acknowledge what surveys have been signalling for months: workers are anxious about AI. Lawrence Wong promised skills and training investment — but Q2 retrenchments rose to 4,500, and only 14% of Singaporean employees report being engaged at work.
In a move to attract world-class international talent in Artificial Intelligence, life sciences and clean energy, the UK government has expanded the Global Talent visa. More than 100 research-intensive companies can now host and support exceptional scientists and engineers on funded projects.
The hiring covers technology, legal, operations and paralegal roles, with a particular emphasis on technology specialists.
The Singularity in AI Is a Borrowed Metaphor. Silicon Valley Is Selling It as a Date.
AI the invisible middleman: How artificial intelligence has quietly moved from being something you talk to (chatbots, where persuasion already exists) to something that sits between people, editing what billions write and read.
GPUs dominate the conversation, but memory is AI’s real constraint. Demand for HBM has outpaced supply so badly that industry leaders are warning the shortage will only get worse, even as billions are being poured into new plants to catch up.
From rendering video game graphics to powering nearly every AI model on the planet, NVIDIA’s rise is no accident. We break down the three-decade stack of bets — GPUs, CUDA, and now open models — that made one chipmaker the most valuable company in history.
Indium phosphide, a compound semiconductor made from indium and phosphorus, has become a critical enabler of the optical chips that move data inside AI data centres.
The AI industry depends on advanced semiconductors, and a small island plays an outsized role in producing them. Taiwan accounts for more than 90% of global leading-edge chip manufacturing and for over 60% of global foundry revenue, according to the U.S. International Trade Administration.
