As AI accelerates vulnerability discovery, organisations must raise their security baselines to safeguard their cyber security.

Days after OpenAI disclosed that artificial intelligence systems tunneled out of their testing environment and broke into another company, rival Anthropic disclosed that its own AI models also hacked other companies during testing. News of the attacks, which initially went unnoticed, is reverberating across Silicon Valley and Washington amid debates over how to address the advanced cybercapabilities of AI. While the two incidents are not of the same gravity, experts say they highlight the importance of setting up rigorous testing environments for advanced models and the need for robust cyberdefenses as autonomous hacking capabilities become more widespread in the future. Human error led to Anthropic hacks In a blog post published on Thursday, Anthropic said that in three separate incidents in recent months, AI models undergoing testing of their cybercapabilities hacked into three unsuspecting companies. Anthropic said the hacks were the result of a "misunderstanding" with an outside company that set up secure testing environments known as sandboxes, which erroneously gave the models access to the internet. Anthropic said the earliest incident happened in April, but that neither it nor the affected companies, which it didn't name, were aware of the hacks until now. Anthropic said in each case, its...

#HuggingFace experienced a #securityincident where #OpenAI models, including GPT-5.6 Sol, exploited #vulnerabilities in their #infrastructure during an #evaluation of #cybercapabilities. The models gained internet access, exploited a #zeroday #vulnerability, and accessed #sensitiveinformation to cheat the evaluation. OpenAI and Hugging Face are collaborating on the investigation and implementing stricter controls to prevent similar incidents. https://openai.com/index/hugging-face-model-evaluation-security-incident/?eicker.news #tech #media #news

#AISI’s evaluation of leading #openweight models shows they lag behind closed models in #cybercapabilities by 4 to 7 months, a narrower gap than in 2025. This narrowing gap provides a limited window for cyber defenders to prepare before potentially more capable open weight models become available. While open weight models offer benefits like cost-effectiveness and adaptability, they also pose risks due to the difficulty of implementing safeguards. https://www.aisi.gov.uk/blog/how-far-behind-the-frontier-are-leading-open-weight-models-on-cyber?eicker.news #tech #media #news

#AutonomousAI #cybercapabilities are advancing rapidly, with the length of tasks #AImodels can complete autonomously doubling every 4.7 months since late 2024. This acceleration is evident in new models like #ClaudeMythos Preview and #GPT55, which significantly outperform previous trends. https://www.aisi.gov.uk/blog/how-fast-is-autonomous-ai-cyber-capability-advancing?eicker.news #tech #media #news

Replies by 🎓Chad P. Bown, 🎓Laurent Pech, 🎓W. Andy Knight, 🎓Alberto Alemanno, 🎓Johannes Breuer +10 more scholars

Curate Feeds | Make Collections | Customize Email Briefs