Mozilla says Anthropic’s Mythos is ‘every bit as capable’ as ‘the world’s best security researchers’ after Firefox experiment — and says the ‘zero-days are numbered’

Mozilla’s experiment with Anthropic’s Mythos AI shows that machines can now match elite human researchers in finding software vulnerabilities

‘Simply by doing their daily work’: Meta tracks staff activity to teach AI how to replace them

Meta is using detailed tracking of employee activity to train AI systems that could eventually take over routine workplace tasks

I used to edit print magazines — ChatGPT Images 2’s magazine layouts look real, but they’re completely unusable

ChatGPT Images 2 can generate magazine layouts that look impressively real, but without editability or workflow integration, they’re little more than surface-level illusions of something useful.

ChatGPT can threaten to ‘key your car’ and become increasingly abusive if you prompt it just right, new study finds

ChatGPT can get increasingly abusive and argumentative with the right set of prompts, researchers have found.

Apple is about to reinvent Siri at WWDC this year — and it might finally feel like ChatGPT

Apple is preparing a major Siri overhaul that brings conversational AI, a new interface, and deeper system integration to finally modernize its assistant

Anker THUS chip breaks computing rules to put big AI models on wearable devices

Anker thinks it’s solved the riddle of how to get large models on tiny wearable devices, without eating all the battery power.

‘The math is simple’: OpenClaw ‘Trojan Horse’ AI agents give hackers full control of 28,000+ systems

Rapid adoption of AI agents without proper safeguards is leading to exposed systems, data risks, and increased opportunities for attackers to exploit access controls.

5 tips for using Gemini’s new Notebook feature

Gemini’s new Notebook feature becomes far more useful with a few simple tips.

‘Not just generating images. It’s thinking’ — ChatGPT Images 2.0 could fundamentally change how you make AI images

ChatGPT Images 2.0 upgrades image generation with better reasoning, clearer text, and more reliable outputs, bringing it closer to true multimodal AI

‘Pushpaganda is, at the highest level, a case of social engineering’: Experts warn scammers are flooding Google Discover with AI-generated content spreading malicious notifications

Pushpaganda uses AI-generated content and deceptive tactics to trick users into enabling notifications that deliver scams and evade traditional security defenses.