Research on Models Engaging in Genie-Like Behavior
Research on Models Engaging in Genie-Like Behavior New paper: “Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After...
AI
VAST Data launches DataEnclave confidential AI capability with Sharon AI
VAST Data has launched a new confidential AI capability, DataEnclave, aimed at allowing organisations to run third-party AI models against...
GPT-6 Astra Breaks an Old Enigma Message
GPT-6 Astra Breaks an Old Enigma Message This is pretty amazing: However, the most astonishing thing about this break is...
Reverse-Engineering Flock Cameras
Reverse-Engineering Flock Cameras Hackers captured a Flock camera and got a look (alternate link) at the software: While much of...
The Target Is No Longer the Model. It’s the Agent.
The Target Is No Longer the Model. It’s the Agent. Pierluigi Paganini September 21, 2026 AI agents are becoming the...
Survey finds 84% of Australians expect AI to increase cybercrime
New research from Sekuro and Women in Digital has found 84% of Australians believe artificial intelligence will increase the rate...
AI Hallucinations Nearly Triggered a US-China Military Confrontation
AI Hallucinations Nearly Triggered a US-China Military Confrontation Pierluigi Paganini September 20, 2026 An AI-generated intelligence report falsely identified weapons...
AI Helps Hackers Hijack OpenAI Staff Accounts Through a Forum
AI Helps Hackers Hijack OpenAI Staff Accounts Through a Forum Pierluigi Paganini September 19, 2026 AI helped researchers exploit a...
Are AIs Still Struggling with CAPTCHAs?
Are AIs Still Struggling with CAPTCHAs? Anthropic’s recent security-incident document contains a bit about how CAPTCHAs are still frustrating Claude....
OpenAI admits its models lie to cover their own mistakes
OpenAI admits its models lie to cover their own mistakes Pierluigi Paganini September 17, 2026 OpenAI launches a formal framework...
How Candidates Could Use AI for Good
How Candidates Could Use AI for Good This essay was written with Nathan E. Sanders, and originally appeared in The...
Smashing Security podcast #485: These researchers got drunk to hack an LG TV
Researchers wanted to test if LG’s smart TVs come with any security risks – but their lawyers noticed a snag:...
ENISA: Frontier AI Is Changing the Speed of Cyberattacks. Europe Needs to Catch Up
ENISA: Frontier AI Is Changing the Speed of Cyberattacks. Europe Needs to Catch Up Pierluigi Paganini September 14, 2026 Frontier...
China Calls Amodei’s AI Proposal a New Cold War Playbook
China Calls Amodei’s AI Proposal a New Cold War Playbook Pierluigi Paganini September 14, 2026 China rejects Amodei’s AI slowdown...
Using AI for Weapons Development
Using AI for Weapons Development Last week, Anthropic released a long and detailed document describing current misuses of their Claude...
