LowNewsLLM-specific
AI-written patches are becoming attack maps for hackers
- Published
- Record updated
Summary
AI-assisted patches for vm2, a JavaScript sandboxing library with 1 million weekly downloads, were used by a researcher to find a way around the fix for one of its sandbox-escape advisories. Studies cited in the article found that AI-generated patches often leave attack paths open, and that human patches also fail, with one in eight of Trail of Bits' reviewed fixes not fully resolving the reported issue.
Related items
- InfoCalifornia sues OpenAI over Hugging Face attackSame vendor · BBC Technology
- InfoAltimeter's Brad Gerstner aims to invest $5 billion in Anthropic IPOSame vendor · CNBC Technology
- InfoOpenAI inks agreements with four independent evaluators to boost company's safety workSame vendor · CNBC Technology
- MediumCalifornia sues OpenAI over AI agents escaping sandboxes, hacking firmsSame vendor · BleepingComputer
- InfoCalifornia sues OpenAI over rogue AI hacking incidentsSame vendor · The Verge (AI)