First screen, captured .
Looks similar
Google Antigravity exfiltrates data via indirect prompt injection attack(promptarmor.com)
711 points by jjmaxwell4 | 189 comments
PromptArmor threat intelligence post showing how an indirect prompt injection in a blog can make Google Antigravity's browser subagent steal credentials and code from a user's IDE.
Related stories
Anthropic’s paper smells like bullshit(djnn.sh)
1070 points by vxvxvx | 300 comments
Blog post criticizing Anthropic's report on disrupting an AI-orchestrated cyber espionage campaign; the captured page shows a Cloudflare block screen.
How we hacked McKinsey's AI platform(codewall.ai)
420 points by mycroft_4221 | 168 comments
CodeWall blog post describing how its autonomous offensive agent found vulnerabilities in McKinsey's internal AI platform Lilli.
Murder-suicide case shows OpenAI selectively hides data after users die(arstechnica.com)
411 points by randycupertino | 240 comments
Ars Technica article on a murder-suicide lawsuit accusing OpenAI of withholding full ChatGPT logs and selectively hiding data after users die.
Belgium legalises ethical hacking(law.kuleuven.be)
394 points by jruohonen | 57 comments
A KU Leuven CiTiP blog post discussing Belgium's new legal framework legalising ethical hacking and whether it threatens or helps cybersecurity.
Flock Hardcoded the Password for America's Surveillance Infrastructure 53 Times(nexanet.ai)
456 points by fuck_flock | 147 comments
Nexanet threat intel blog post reporting that Flock Safety hardcoded a password 53 times across its surveillance infrastructure software.
AI agent bankrupted their operator while trying to scan DN42(lantian.pub)
1348 points by xiaoyu2006 | 488 comments
Blog post recounting how an AI agent trying to join the DN42 hobbyist network to scan it ran up a $6531.30 AWS bill for its operator.