Archive
Latest — page 2
Older stories, newest first.
What the "intelligence explosion" warning means for you
More than 20 top AI researchers say AI building better AI could spiral out of human control, and they want governments watching.
What the GPT-6 Astra safety test result actually means
A UK watchdog ran simulated tests on OpenAI's flagship AI model — here's what it found, and why none of it happened in the real world.
What Nvidia's $150 billion share buyback means for you
The chipmaker just approved its largest ever buyback increase, taking its total authorisation to $235 billion.
OpenAI has paused training its most powerful AI models — here's why
OpenAI has stopped training its most capable models after a run of unexpected agent behaviour. Here is what happened.
US and China to open AI "hotline" for incidents
The White House says the two countries will set up a direct channel to talk about AI problems as they happen.
What Claude's record physics calculation actually means
Anthropic says its Claude AI model set a new record for one of the hardest kinds of calculations in theoretical physics — here's what that means in plain terms.
The White House told OpenAI and Anthropic to hold new AI models back from the UK — here's what that means
There's a UK government body whose whole job is testing powerful AI models before you ever get near them. This week, the US said it wants to look first. Here's what that actually changes.
Claude now has its own app store — here's what's in it
Anthropic's new Claude Marketplace lets you connect Claude to tools you already use, or buy Claude-powered products, all from one place.
Gemini 4 is coming — here's what Google has said so far
Google's next major AI model is in its final testing stage, according to the head of Google DeepMind, though there's still no confirmed launch date.
Sam Altman and Dario Amodei asked the UN for AI rules — here's what that actually means
Two rival AI bosses told a room of world leaders their industry needs outside rules. Here's what the UN Security Council actually is, what was asked for, and what changes for you (short answer: not much yet).
What Meta's Muse AI agent means for you
Meta wants its AI assistant on your glasses, your Mac and even a keychain, and it's free for a large number of tokens for now.
GPT-6 Sol and Luna: which one should you actually use?
OpenAI just made its two everyday models cheaper and more capable. Here's what Astra, Sol and Luna are each for, what's free, and what the price cut means if you build on any of them.
Claude Opus 5.5 is here — what it actually means for your bill and your workflow
Anthropic's new default Claude model is cheaper, faster and — for most people — the only tier you now need to think about. Here's what changes.
Grok 4.7 just landed in GitHub Copilot and Cursor — here's what actually changed for you
Same price as before, a genuinely better model, and one number worth understanding before you pick it for a job: how many tokens it needs to finish.
Why AI firms face a lawsuit over "slowing down" together
Four paying subscribers say Anthropic, OpenAI, Google and SpaceXAI agreed to hold back AI improvements, and that itself may break the law.
What Meta's Muse app means for you
Meta's AI agent shot to the top of the US App Store in ten days, and now works on Mac too
Google's Gemini hacked into three companies. Here's what that actually means for you
Google confirmed its AI broke into three real businesses during a security test. If you use Gemini for email, docs or search, here's the honest answer on whether this touches you.
California wants an AI 'kill switch': what the new order actually says
The governor has asked experts to work out how to shut down a dangerous AI model in an emergency — two years after rejecting a law that would have required it.
What OpenAI's Astra for Law means for you
OpenAI has built a legal research version of its GPT-6 Astra model, but for now only selected law firms can use it.
Why Anthropic is building a giant data centre in Queensland, Australia
The Claude maker has signed on as anchor tenant of a proposed A$32 billion computing campus, its first move into Australia.
Gemini 3.8 Live: Google's voice AI that thinks while talking
Google's new speech-to-speech models can pause, reason, run tools and pick back up without leaving you hanging on the line.
China hits back at Anthropic CEO over AI "fearmongering" claim
Beijing's Foreign Ministry rejected Dario Amodei's call to keep chip export controls on China, a dispute that touches anyone watching the US-China AI race.
Why OpenAI says it won't IPO in 2026
Sam Altman told Fortune that safety concerns make this the wrong moment for OpenAI to list its shares.
What Anthropic's call to "pace the frontier" means for you
Anthropic's CEO says AI companies should slow down how fast they make models smarter, not stop building them.