The UK government took over British Steel last year to prevent the closure of the country's last primary steelmaking facility. China's Jingye Group, which bought the firm in 2020, claims it is owed compensation.
Authorities in disputed region are auditing school, college and public libraries to look for ‘objectionable’ material.
Article URL: https://chipweinberger.com/articles/20260719-hardware-is-not-so-hard Comments URL: https://news.ycombinator.com/item?id=48966713 Points: 388 # Comments: 195
Google Deepmind's GenCeption repurposes a video generator for classic vision tasks such as depth estimation and segmentation, matching state-of-the-art systems with far less training data. The model trained almost entirely on synthetic videos. Its results add to the debate over whether video generators already contain a kind of universal world model. The article Google Deepmind argues video generators already contain the world models computer vision has been missing appeared first on The Decoder .
Almost all asylum applications by Russian deserters are being denied by German authorities. DW and Russian-language media outlet Astra spoke to one of the young men who fears he'll be sent back to die on the front lines.
Football fans in Gaza express support for Spain in showdown with Argentina, citing Spanish solidarity with Palestinians.
The incoming PM is expected to announce fresh support for the North Sea from day one of his premiership.
South African fighter held the middleweight belt from January 2024 until August 2025 when he lost to Chimaev.
Spain face Argentina in the 2026 World Cup final as musicians perform before kickoff and during the half-time break.
Moonshot's Kimi K3 is the first Chinese model to top the Code Arena: Frontend rankings, beating Claude Fable 5 and GPT-5.6 Sol by a wide margin. But on advanced math, the gap is stark: Kimi K3 scores only about 39 percent on FrontierMath Tier 4, while models from OpenAI and Anthropic hit close to 90. The article Moonshot's Kimi K3 outperforms Fable 5 in frontend code but lags far behind in complex math appeared first on The Decoder .
The survivors were found clinging to a fish trap and floating debris at sea three days after the boat sank.
The folks at TryAI put Fable to the test in directing a music video. The AI glitches show up from the start and continue. But it’s even more unsettling than this. When AI tries to create and direct joyous human dancing, it fails. The dancers look like retired accountants at a wedding. The stiff awkwardness […]
Article URL: https://www.qwencloud.com/pricing/token-plan Comments URL: https://news.ycombinator.com/item?id=48966190 Points: 215 # Comments: 1
US strikes target Iran's IRGC after two of its soldiers were killed in Jordan, as Tehran attacks Kuwait and Bahrain.
Setting up illegal settlements, erasing cities and obstructing aid serve goal of forcing Palestinians out, analysts say.
https://www.qwencloud.com/pricing/token-plan Comments URL: https://news.ycombinator.com/item?id=48966120 Points: 614 # Comments: 447
‘King [Messi] abdicates, but the prince [Lamine Yamal] takes the throne!’
Epoch AI tested three leading AI text detectors (Pangram, GPTZero, and Originality.ai) using style-imitated texts. Up to 18 percent of AI-generated passages went undetected. For scientific writing, the miss rate climbed as high as 48 percent, the very genre where these detectors likely see the most real-world use. The article AI text detectors struggle when language models mimic an author's style appeared first on The Decoder .
Spain face Argentina in the World Cup final as Lionel Messi, Lamine Yamal and history dominate the build-up.
Latest match highlights, reaction and previews from the FIFA World Cup 2026 in North America.
Matchups, a style clash, Messi magic, a midfield battle and a passing of the torch could be key narratives in the final.
Article URL: https://github.com/openai/codex/pull/33972/files Comments URL: https://news.ycombinator.com/item?id=48965850 Points: 217 # Comments: 95
The RadLE 2.0 benchmark tests whether AI models in radiology can tell when they should leave a diagnosis to a human. Many models deliver wrong findings with full confidence, and human radiologists are still well ahead. Before AI can diagnose on its own, it needs to learn when it's better to say nothing. The article AI chatbots reading X-rays can be dangerously confident even when they're wrong appeared first on The Decoder .
Fires broke out across the Ukrainian capital after Russia's latest missile attack, which used the highest number of ballistic missiles of the war so far. Several people have been injured in Kyiv.