Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
Anthropic accused Alibaba, Moonshot, and DeepSeek of running large-scale distillation campaigns against Claude, alleging roughly 200 million exchanges were used to extract the model's reasoning traces for training compet…
Anthropic said in a Thursday report that Alibaba, Moonshot, DeepSeek, Zhipu, and Xiaomi combined launched nearly 190 million distillation attacks against Claude between May and July, including 23 million user requests re…
A developer published a setup guide for a Python shim that exposes Cognition's SWE-2 and SWE-1.x coding models through an OpenAI-compatible API, letting tools like Claude Code call them via CLIProxyAPI. The shim translat…
OpenAI released GPT-6 Astra on September 3, its first flagship model to ship with computer use as a core capability and the first pre-trained on more than 100,000 GPUs at its Stargate site in Texas. The model is broadly …
A developer reports that Qwen 3.8 27b, a locally run open-source model, now matches frontier cloud models like Claude Opus 4.6 for agentic coding tasks, scoring 34 versus 32 on the Artificial Analysis index at the xhigh …
Sakana AI released Fugu Max and Fugu Ultra v2, two configurations of its Fugu orchestration architecture that route tasks across a pool of open and specialized models, including the NVIDIA Nemotron family. Sakana AI said…
Anthropic disclosed in its third misuse report since March 2025 that it blocked multiple attempts this year by scientists in China and Russia to use its Claude AI models for research that could aid biological weapons dev…
Anthropic's fourth misuse report, covering December 2025 through August 2026, documents Russian-linked espionage and AI-orchestrated cyberattacks in which multi-agent systems ran much of the cyber kill chain with minimal…
OpenAI filed a patent for a second AI model designed to read another model's chain-of-thought reasoning and flag reward hacking, misgeneralization, sycophancy, and deception, according to a weekly roundup of 272 Big Tech…
Apify released a setup guide for connecting Claude Desktop and ChatGPT to its MCP server, which exposes more than 70,000 Store actors as callable tools for live data retrieval. Apify's own testing found that Claude brows…
Researchers Ayhan Suleymanzade and co-authors submitted a paper to arXiv on 10 September 2026 proposing "looped flows," a training method that uses local denoising objectives with progressively decreasing noise levels an…
Anthropic reported that it blocked malicious uses of its Claude models, including an effort in northern Yemen to deploy Claude for missile guidance software covering a guided rocket and a long-range ballistic missile, an…
Anthropic published a report finding that AI distillation can enhance general reasoning abilities, potentially increasing dangerous capabilities beyond a model's training subjects, while the associated safeguards do not …
CertArena launched as a free AI certification simulator built with zero dependencies on native Node 24, offering 26 scenarios covering AI fundamentals, Amazon Bedrock, responsible AI, and AWS security decisions. The tool…
OpenAI's GPT-6 Astra became the first AI model to score 100% on ExploitBench, a Carnegie Mellon University benchmark that grades how far a model can carry an exploit chain to arbitrary code execution, earning a "Critical…
Researchers propose EvoSafeHarness, a method for evolving model- and domain-specific safety harnesses that add a system-level enforcement layer to protect LLM agents against indirect prompt injections and direct harmful …
Researchers introduced CMNIE, an information extraction benchmark for Chinese military news containing 13,000 instances drawn from public Chinese military news and manually annotated for 7 event types, 10 argument roles,…
Researchers introduced CONTRACTEVAL, a diagnostic framework that represents procedural instructions as query-active obligations and matches them against response or trace evidence to detect conformance failures. On a con…
A new paper introduces MAAGL, a multi-agent agentic graph learning framework that partitions a graph into communities and assigns an independent agent to each community for region-specific specialization. MAAGL summarize…
A study of 93 Urdu stories generated by three contemporary LLMs — GPT-5.1, Qwen-3-Max, and DeepSeek-3.1 — found that the models frequently make basic grammar and semantic errors, produce incoherent and repetitively unnat…