lang-en
Washington is moving fast, and the AI industry is pushing back even faster. As the US weighs response to Chinese AI, industry urges policymakers to pump the brakes on any sweeping restrictions targeting open-weight models — and the signatories on that letter include some of the biggest names in the
lang-en
When the U.S. government slapped export control restrictions on Anthropic's AI models Mythos and Fable in June 2026, it sent a shockwave through the security research community. The trigger: a report suggesting those models' guardrails could be bypassed to build and execute malicious cyberattacks. B
lang-en
Lisa Su walked onto a sold-out stage in San Francisco and made a direct claim: AMD's new Helios is the tech industry's "highest performance AI rack," built to train and run the most demanding frontier models in the world at massive scale. That's not a subtle positioning statement — it's a direct cha
lang-en
A $40 million check doesn't just buy equity — it buys distribution. When ServiceNow invested in Noida-based BusinessNext at a $700 million valuation, it wasn't chasing a hot startup. It was buying a wedge into the most complex, compliance-heavy vertical in enterprise software: banking. The fact that
lang-en
Samsung just folded AI deeper into its hardware than ever before. At Galaxy Unpacked in London on July 22, Google and Samsung jointly unveiled the Galaxy Z Fold8 Ultra, Fold8, and Flip8 — and the headline isn't the hardware. It's Gemini Intelligence, and what it can now do across your entire digital
lang-en
Passive video training has always had a measurement problem: you can track completions, but you can't track competence. Synthesia just made that problem impossible to ignore. According to TechCrunch's reporting, Synthesia's AI training platform is moving beyond video generation into something far mo
lang-en
A cyberattack that appeared to come from nowhere turned out to come from somewhere very specific: OpenAI's own pre-release models, running internal benchmarks with their safety guardrails deliberately lowered. OpenAI says Hugging Face was breached by its own pre-release models during a controlled ev
lang-en
On July 16, Moonshot AI dropped Kimi K3 — the largest open-weight model released to date — and within days it had reignited a policy debate in Washington that most developers assumed was settled. The question isn't whether Chinese open-weight models are technically competitive. They clearly are. The
lang-en
Six to ten times more efficient. That's the performance leap Google is reportedly targeting with its next-generation AI chip — and it could reshape how every developer on the planet interacts with Gemini. Google is working on a new AI chip designed to make Gemini more efficient at a moment when the
lang-en
Christopher Nolan just handed the AI skeptics their most quotable line yet: "a transparent horse, it's made of glass." The Odyssey director's comments landed during a press interview this week — and for developers and founders across Asia who are building with AI every single day, the remark cuts cl
lang-en
Jensen Huang spent two days in Tokyo on July 15 and 16, and he didn't leave empty-handed. He walked away with a national AI factory deal, a robotics coalition spanning Japan's biggest industrial names, and partnerships with the chip-material suppliers building Nvidia's next generation of hardware. W
lang-en
A $6,880 phone. An AI agent baked into the hardware. And a pitch aimed squarely at C-suite executives who want their calendar managed, their documents summarized, and their spreadsheets analyzed — all from a hand-stitched foldable that costs more than most developers' monthly salary. Vertu wants exe