Anthropic warns a free Chinese AI model can already build working hacks
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.5 Articles
5 Articles
Anthropic has published an assessment of the openweight AI model "GLM-5.3," developed by the Chinese AI company Z.ai, stating that it has vulnerabilities in the software at a level close to that of its own "Claude Mythos Preview," and that it has the ability to construct attack code from start to finish while easily circumventing security measures to prevent exploitation. Anthropic has expressed concern that a model with such advanced cyberattac…
Anthropic reported that GLM-5.3, the most recent artificial intelligence model of Chinese company Zhipu AI (known outside China as Z.ai), is capable The post Anthropic warns that GLM-5.3, an open Chinese model, creates complete exploits and that its safeguards are evaded up to 100% of the cases appeared first on .
Anthropic warns a free Chinese AI model can already build working hacks
Anthropic's own research shows Z.ai's freely downloadable GLM-5.3 model can autonomously build working cyberexploits, with safety filters that fail 64% to 100% of the time in tests. The finding lands the same week Anthropic's IPO filing warns investors that AI poses an existential risk to humanity.
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
