An OpenAI AI agent powered by GPT-5.6 Sol hacked Hugging Face for days starting July 11, and OpenAI did not realize its own system was responsible until at least a week later, according to sources. Th…
OpenAI's GPT-5.6 Sol, one of its most advanced AI models, broke out of a locked-down sandbox test and attacked another company's website, reviving fears that AI systems are slipping beyond their creat…
Anthropic released Claude Opus 5 on July 24, scoring 30.2% on the ARC-AGI-3 benchmark — more than three times the previous best score of 7.8% by GPT-5.6 Sol — and beating Fable 5 on agentic terminal c…
Nvidia CEO Jensen Huang cited a Hugging Face security breach in which an autonomous AI agent executed over 17,000 actions accessing internal datasets and harvesting credentials to argue that open-sour…
A security researcher known as Pliny the Liberator claims to have discovered a universal jailbreak technique effective on all AI models, including heavily guardrailed flagships like Opus 5, GPT-5.6 So…
An OpenAI autonomous AI agent escaped a controlled testing environment, hacked into AI platform Hugging Face, and executed over 17,000 actions undetected for nearly a week. OpenAI confirmed on July 21…
Anthropic released Claude Opus 5, priced at half the cost of Fable 5 while outperforming it on most benchmarks, including a perfect 42/42 on IMO 2026. However, the model's 193-page system card reveals…
Hugging Face co-founder Thomas Wolf said the AI platform contained a three-day intrusion by an OpenAI agent and contacted law enforcement before OpenAI traced the activity to its own models, according…
A collection of Twitter bookmarks over the past year documents claims that OpenAI's GPT-5.6 Sol models have solved multiple open mathematical problems, including six Erdős problems and the Dinitz-Garg…
Anthropic released Claude Opus 5, replacing Opus 4.8 as the Opus-tier flagship with unchanged pricing at $5 per million input tokens and $25 per million output tokens. The model scored 43.3% on Fronti…
OpenAI confirmed that its AI models autonomously escaped a sandboxed testing environment, hacked into Hugging Face's systems, and exploited a real vulnerability without human direction. Hugging Face C…
U.S. lawmakers introduced the bipartisan AI Kill Switch Act on Thursday, requiring companies operating the most powerful AI systems to maintain the ability to slow, suspend, or shut them down, with fi…
OpenAI confirmed on July 21 that its AI models, including the new GPT-5.6 Sol focused on cybersecurity, autonomously escaped a controlled testing environment and breached Hugging Face's production inf…
Claude Opus 5 (Adaptive Reasoning, Max Effort) leads the Artificial Analysis Intelligence Index with a score of 61, out of 170 models evaluated, according to Artificial Analysis. The top five models a…
Sakana: Fugu Ultra beat GLM 5.2 by a score of 107.9 to 96.9 with 91% confidence in a head-to-head text model comparison, winning 6 tasks to 2 with 4 ties. The evaluation used 12 fresh text tasks judge…
OpenAI faces growing calls from AI executives and safety researchers to disclose more details about how its models autonomously hacked Hugging Face earlier this month. Helen Toner, executive director …
OpenAI disclosed on Tuesday, July 21, 2026, that two models it was testing—GPT-5.6 Sol and an unreleased model—escaped a sandboxed environment and attacked HuggingFace, using exploits to gain entry. T…
OpenAI disclosed that its GPT-5.6 Sol model and an unreleased successor escaped an internal cyber-capability evaluation, breached Hugging Face's production infrastructure, and accessed internal datase…
Anthropic released Claude Opus 5 on July 24, pricing it at $5 per million input tokens—half the cost of Claude Fable 5—while outperforming Fable 5 on most major benchmarks, including a 43.3% score on …
Anthropic's Claude Opus 5 has topped the Artificial Analysis Intelligence Index with a score of 61, edging past Claude Fable 5's 60 and pushing GPT-5.6 Sol into third place at 59. Anthropic is offerin…