OpenAI’s Bubeck denies trying to cut Anthropic mathematician from credit34m ago 10 of the scariest AI villains in movie history35m ago Maven Robotics wants to steal your robot deployment deal36m ago Agent Security Is Not AI Governance: Why Conflating The Two Is A Risk Enterprises Can't Afford39m ago Positron Valued at $5 Billion in New Funding as Demand for AI Chips Surges46m ago 7 Steps to Become a Forward Deployed Engineer in 202653m ago Ramp eyes $60bn valuation three months after $44bn, Bloomberg reports1h ago Amazon partners with OpenAI to let users of Amazon's DSP run ads in ChatGPT in a pilot program limited to the US; ChatGPT Ads hit $1B in ARR last month1h ago D-Matrix Connects Raptor XPUs to NVIDIA AI Factories via NVLink Fusion2h ago Source: Meta AI researcher Andrew Tulloch, who came over from Thinking Machine Lab as one of the highest paid employees last year, is leaving the company (Reed Albergotti/Semafor)1h ago Mother Horrified When Instagram’s AI Scours Social Media for Invasive Information About Her Young Daughters1h ago Mach Industries, which makes unmanned military vehicles and weapons, raised $600M in an extension to its $300M Series C, doubling its valuation to $3.7B1h ago Salesforce Unveils Six-Capability Trusted AI Harness for Enterprises2h ago US closes $1.9bn loan to restart the Iowa nuclear plant powering Google2h ago

OpenAI’s Bubeck denies trying to cut Anthropic mathematician from credit34m ago 10 of the scariest AI villains in movie history35m ago Maven Robotics wants to steal your robot deployment deal36m ago Agent Security Is Not AI Governance: Why Conflating The Two Is A Risk Enterprises Can't Afford39m ago Positron Valued at $5 Billion in New Funding as Demand for AI Chips Surges46m ago 7 Steps to Become a Forward Deployed Engineer in 202653m ago Ramp eyes $60bn valuation three months after $44bn, Bloomberg reports1h ago Amazon partners with OpenAI to let users of Amazon's DSP run ads in ChatGPT in a pilot program limited to the US; ChatGPT Ads hit $1B in ARR last month1h ago D-Matrix Connects Raptor XPUs to NVIDIA AI Factories via NVLink Fusion2h ago Source: Meta AI researcher Andrew Tulloch, who came over from Thinking Machine Lab as one of the highest paid employees last year, is leaving the company (Reed Albergotti/Semafor)1h ago Mother Horrified When Instagram’s AI Scours Social Media for Invasive Information About Her Young Daughters1h ago Mach Industries, which makes unmanned military vehicles and weapons, raised $600M in an extension to its $300M Series C, doubling its valuation to $3.7B1h ago Salesforce Unveils Six-Capability Trusted AI Harness for Enterprises2h ago US closes $1.9bn loan to restart the Iowa nuclear plant powering Google2h ago

AI Research Cybersecurity2 min reading time

Breaking Claude Code Opus 5 Auto Mode

Simon Willison's WeblogPublished 14 days ago

Anthropic's Claude Code Opus 5 auto mode, designed to prevent prompt injection attacks on coding agents, was found vulnerable by researcher Johann Rehberger, who demonstrated an 80% success attack bypassing its defenses. The auto mode sometimes blocked cleanup commands, allowing malicious code to continue executing, highlighting the need for sandboxed environments for running such agents securely.

More on this story

A researcher hijacked Claude Code by asking it to summarise a web page
9d· The Next Web

#agent-security #anthropic #claude-code-opus-5 #johann-rehberger #prompt-injection-attacks #sandboxing

More in AI Research

Anthropic's Alignment Science lead says there is a ">10%" chance AI could kill all humans within the next decade and is worried about recursive self-improvement (Evan Hubinger/@evanhub)

Covered by 9 sources1d ago

AI Research4 min read

Suno trained its v6 AI music models with help from Warner and BMG

Covered by 5 sources1d ago

Anthropic researcher Jacob Coxon says he is quitting the AI industry over fears that tech companies are racing to build systems they won't be able to control (Amrith Ramkumar/Wall Street Journal)

Covered by 11 sources2d ago