Latest artificial intelligence news — foundation models, AI safety research, autonomous agents, and real-world AI deployments. Updated every 30 minutes from top AI publications.
OpenAI's Cyber team reported on Nov 4 that a pipeline built on an unreleased "upcoming cyber model" reproduced 1,724 real Linux kernel vulnerabilities and converted them into 948 distinct local privilege escalation explo…
Anthropic released FlashProve, a zkSNARK library that cryptographically proves which LLM generated a given response without exposing the model's weights, according to byteiota. The library carries a 70-95x overhead, the …
OpenAI released Images 2.6 along with the GPT-Image-2.6 Flare and Sunburst API models, with Flare generating images in as little as 6 seconds. Both models add GIF output, SVG export, markup edits and a complexity paramet…
A paper titled "EmoReward: Self-Evolving Reward Calibration for Emotional Video Generation RL" proposes a self-evolving reward calibration method for reinforcement learning post-training of emotional video generation, ad…
Multi-turn agentic reinforcement learning (RL) improves long-horizon task performance, but those gains are highly sensitive to interaction constraints, according to an evaluation using Pass@(K,T) across sampling breadth …
Moonshot AI published the weights for Kimi K3.1 Pro, a 2.8T-parameter mixture-of-experts model with 104B activated parameters, a 1M token context, native multimodality and MXFP4/MXFP8 quantization, now live on OpenRouter…
A developer reports that using Crouton Alpha to rewrite a hot Rust loop produced code that was slightly faster than previously hand-optimized code, but only after several failed attempts that showed promise before the co…
Baidu's Apollo unit unveiled Apollo World, an agent-driven multimodal world model for physical AI, at its AI Day event in Shenzhen on Nov. 3, and the company said the model is already being used to test and validate the …
OpenAI reported that an autonomous vulnerability-discovery pipeline built by its security researchers validated and generated patches for 948 local privilege-escalation vulnerabilities in the Linux kernel in a single run…
Aito, led by CEO and founder Antti Rauhala, reported that on a UK wholesaler's order history its predictive database's top suggestion for buyers with only one or two earlier orders appeared in their next orders 31% of th…
Tencent Hy4, released November 3, 2026, scores 35 on the Artificial Analysis Intelligence Index, above the median of 30 for comparable models, according to Artificial Analysis. The reasoning model from Tencent accepts te…
Moonshot AI released the weights of its flagship multimodal model Kimi K3.1 Pro on Hugging Face, making it available for independent deployment after earlier availability through Kimi's applications and API. The model us…
Researchers introduced FrontierPhysics, a new benchmark for evaluating AI agents on end-to-end physics research, arguing that existing physics benchmarks only test fixed questions, isolated sub-tasks, or paper reproducti…
A paper proposes Muon-RT, a variant of the Muon optimizer for large language model self-distillation that reuses the exponential moving average (EMA) teacher's trajectory instead of storing a separate momentum buffer, el…
JustFit v2.3.1, released on GitHub Releases and PyPI by developer YuhuaBillChen, serves a 27B language-and-vision model with its full 256K context on Macs with at least 24 GiB of unified memory. At 16K context, JustFit v…
Google Research released Agile Deliberation, a human-in-the-loop framework that lets users define subjective visual concepts through borderline images. Across 18 live sessions, the framework raised Gemini-based classifie…
DNSFilter's "The Cost of Shadow AI" report found that 74.5% of respondents have a larger cybersecurity budget than 12 months earlier, and 40.3% said their next incremental security dollar would go to AI security and gove…
A study of 1,882 Russian-language documents found that four machine-generated-text detection methods disagreed on the same texts when only the task wording changed, with three procedures shifting negatively and a judge m…
A controlled study of 10 LLM models across 1,000 synthetic search worlds found that final answers are more sensitive to the order of initial evidence during active search than when all evidence is supplied in full, accor…
A new arXiv paper (2611.05814v1) introduces Consultation Value Distillation (CVD), a method that estimates offline, via Monte Carlo executor rollouts, how much an LLM agent benefits from consulting a human at a given dec…