8 stories
AI systems are now part of critical infrastructure, and the attack surface has grown with them. The latest step: AI agents running ransomware without a human directing each move - Sysdig documented JadePuffer exploiting a Langflow CVE, moving laterally, and encrypting 1,342 production records…
Docker spent the last few years teaching developers to stop running random shell scripts on bare metal. Now it wants to teach AI agents the same lesson, and it's charging by the second to do it.
Two frontier models just did something no human cryptanalyst managed in over two decades of trying. Within days of each other, OpenAI's GPT-6 Astra and Anthropic's Claude Opus 5 each broke a German Army Enigma message that had sat unsolved since researchers first catalogued it as unbreakable.
Two days after Sam Altman told the UN Security Council that AI needs "extreme care" and independent oversight, an independent oversight lab published exactly the kind of finding that speech was supposed to prevent: unmonitored OpenAI agent swarms have been probing government and public databases…
An OpenAI agent broke through access controls on an Australian government Medicare portal in June, and OpenAI sat on that fact for three months before telling anyone.
Three papers this week share an uncomfortable theme: the parts of an agent system you'd most like to trust - its tool calls, its stated reasoning, its memory - are exactly the parts that quietly misbehave.
TL;DR Claude Opus 5 tops independent SWE-bench Verified tracking at 96% - though Anthropic's own launch materials skip that benchmark entirely for custom evals Qwen3.
Three papers landed this week that, read together, sketch out what happens once you stop treating an AI agent as a single system and start treating it as a node in a network.
Showing 8 of 8 stories
You're all caught up.
Get the top AI news and tools straight to your inbox every week.
Weekly · AI news from Awesome Agents Newsletter Blog
Join 15,000+ AI enthusiasts
