Unreleased Anthropic Claude model autonomously finds critical flaws in five platforms
Anthropic released a report on an unreleased Claude model that autonomously identified dozens of critical vulnerabilities across five platforms, chaining them into five end-to-end attacks. The model breached Linux core in 53 minutes without credentials, demonstrating autonomous security capabilities; all vulnerabilities were patched.
Summarized by AI from one outlet
Source
Latest
- Research
Study: AI Auditors Catch Most Agent Misbehavior But Flood Reviewers With False Alarms
A study by an Anthropic researcher finds that GPT-5 can detect about 98% of AI agent reward-hacking cases. However, increasing sensitivity leads to a high volume of false alarms, meaning current methods are not yet reliable enough for safety checks.
- Research
Vals AI: Inkling 2 Small swears in 316 benchmark runs
An analysis by Vals AI found that Thinking Machines' Inkling 2 Small model swears 316 times across 2,655 benchmark runs, typically during failed long tasks. The study notes that profanity does not serve as a reliable warning signal, as 96 percent of failed runs contained no cursing.
- Research
Val benchmark shows open-source DeepSeek V4.1 cyber capabilities close to closed GPT-6
The Vals AI benchmark revealed that OpenAI's GPT-6 Sol Daybreak achieves 80.8% of attack objectives in simulated corporate networks, while the open-weight DeepSeek V4.1 Flash reaches 72.8% at a fraction of the cost. Analysts warn that the window for defenders to counter AI cyber threats is shrinking as open models cannot be restricted.
- Research
Study shows political bias in Chinese AI models can be removed via fine-tuning
Nous Research retrained Alibaba's Qwen3.8-27B to create Arete-27B, stripping its pro-Chinese bias. The study found propaganda rates dropped from 22% to under 1% while general capabilities remained intact.
- Research
METR study reveals alarming speed of Anthropic's AI R&D automation
AI safety organization METR published a report indicating that Anthropic is using AI to build more powerful models. The study found that model improvements are occurring roughly twice as fast as a year ago with no signs of slowing down, raising concerns among researchers.
All news → · All rights belong to the original creators