Full-text search across 24547 articles. Combine with topic and date filters; results sorted by relevance.
Elon Musk proposed in a July 23 interview with The Economist that rival AI labs, including OpenAI, Anthropic and xAI, conduct regular safety discussions and pre-release review of major frontier models, with government in…
Elon Musk proposed in a July 23 Economist interview that rival AI labs should peer-review each other's frontier models before release, arguing competitors are better safety arbiters than governments. The proposal comes a…
Open-weight models like Kimi K3 are pressuring the margins of closed-source labs such as OpenAI and Anthropic, potentially shifting AI profit toward chipmakers and cloud providers. Kimi K3, a Chinese open-weight model, l…
Singapore sovereign wealth fund GIC is bullish on artificial intelligence but warns that the amount of capital flowing into the AI ecosystem poses the largest market risk factor, more so than any country, industry or sec…
Anthropic announced on Thursday that Claude's voice mode now runs on its Opus, Sonnet, and Haiku models, enabling deeper conversations and mid-conversation model switching. Users can also speak to third-party tools like …
Anthropic has expanded Claude's voice mode to its Opus and Sonnet models, previously only available on the Haiku model, and extended the feature to apps like Gmail, Slack, and Canva. The company said users were using voi…
A new open-source benchmark called CVE-Bench evaluates large language model agents on fixing real-world security vulnerabilities by running them inside sandboxed Docker containers and scoring them against maintainer secu…
Chinese open-weight AI models from Kimi, Qwen, DeepSeek and others are competing globally on price, customization and deployment flexibility, prompting U.S. AI companies like OpenAI and Anthropic to raise national securi…
White House tech policy director Michael Kratsios publicly accused Moonshot AI of building a covert distillation platform to train its K3 model on Anthropic's Fable, and Treasury Secretary Scott Bessent threatened sancti…
Poetiq claims its Recursive Self-Improvement (RSI) Metasystem has autonomously set state-of-the-art results on six diverse benchmarks, including outperforming Muse Spark 1.1 within 48 hours of publication and surpassing …
Nearly 200 Silicon Valley startups, including Proton and Y Combinator, urged the Trump administration not to block access to Chinese open-weight AI models, warning that such a ban would cripple U.S. startups. In a letter…
Anthropic shipped Record a Skill for Claude Cowork on Pro, Max, and Team plans, turning screen recordings with voice narration into reusable, editable SKILL.md files without requiring any code. The feature runs on macOS,…
Anthropic is donating an additional $20 million to Public First Action, a non-partisan organization focused on AI public education and policy, bringing total support to $40 million. The company argues that as AI models r…
Prediction markets on Polymarket forecast a median release date for Anthropic's next Claude Opus model by Dec 31, 2026, with a 99% probability by that date. Google's next Gemini Pro model has a median release by Aug 31, …
BlastRadar released a GitHub Action that automatically scores every pull request for production risk on a scale of 1-10 and blocks merges if the score exceeds a configurable threshold (default 7). The action uses an Anth…
China-based AI firm Moonshot is seeking a $50 billion valuation in a new fundraising round ahead of a potential Hong Kong initial public offering later this year, according to Bloomberg. The Beijing-based company, founde…
Chinese startup Moonshot AI Pte Ltd. released its Kimi K3 model last Friday, claiming it is the world's largest open-weight model with 2.8 trillion parameters and outperforms U.S. rivals on certain benchmarks, signaling …
Researchers at Redwood Research and Anthropic have developed a method called Contrastive Synthetic Document Finetuning to measure reward-seeking behavior in AI models, finding that intermediate checkpoints of OpenAI's o3…
Mathematician Levent Alpöge used Anthropic's Fable 5, the latest frontier AI model, to disprove the Jacobian conjecture, a long-standing open problem in algebraic geometry that had resisted solution for nearly 90 years. …
Prince Canuma, the maintainer of mlx-vlm, released Nativ, an MIT-licensed macOS app for local AI inference, on the day of its v0.0.1 launch. The app bundles an mlx-vlm server, pulls models from Hugging Face, and offers O…