Fable 5.1's “45 percent cheaper” claim is disputed
Anthropic announced its new models with a cache discount. Artificial Analysis, which took part in pre-release testing, says the saving does not hold in every scenario.
Models
Epoka — What's happening in AI, without the noise.
What the technical report leaves out: how to read the human side of an accident
The 38-page report explains how the failure happened, not why it was not stopped. Safety experts say that is the real question. Step by step.
Safety
Bank of England chief warns the G20 about inflated AI valuations and borrowed money
What worries Bailey most is not the valuations but the leverage stacked on them. If one AI company stumbles, the chain runs long.
Companies
OpenAI starts billing some customers only when the AI actually finishes the job
Outcome pricing instead of a subscription. The hard part is attribution: proving the good result really came from the AI.
Companies
The EU classifies ChatGPT as a very large search engine, with four months to comply
Built-in web search and at least 45 million EU users crossed the threshold. Reddit and Roblox were reclassified in the same decision.
Policy
Instagram throttles the reach of undisclosed AI profiles
The company states the reason plainly: people dislike finding out later that a profile was AI. Those who use the label are not penalised.
Policy
The skills that earn top grades are the ones AI fakes best
An experiment with 1,053 students at Bocconi raised grades. But the rubric was penalising the very markers of real thinking. Here is the argument, step by step.
Research
Nvidia's advantage has moved from the GPU to the system around it
Competition shifted from the chip to traffic control. Getting data to the GPU at the right moment now matters more than more processor cycles.
Companies
Worker reviews have turned against AI: positive share fell from 81 to 43 percent
The complaints go well beyond fear of losing a job. The most sceptical group is Gen Z women, at 21 percent positive.
Companies
AI agents have no sense of time, and no awareness that they lack it
They can neither predict how long a task takes nor tell how long they have worked. They also overrate their own output by 20 points on average.
Research
Claude Code's limit rises on paper and falls 17 percent in practice
The permanent baseline goes up 25 percent, but the 50 percent temporary boost in force today expires on September 14. The gap is the cut.
Products
A $399 bipedal robot: Hugging Face opened the training loop too
Walking, kicking, standing back up after a fall — every move is a neural policy trained in simulation. The reward functions are on GitHub as well.
Products