News
Anthropic and Open AI Are Exaggerating Cybersecurity Risk, Says Hacker George Hotz
1+ hour, 34+ min ago (266+ words) Claude Mythos has created quite a buzz with its cybersecurity concerns, but a prominent cybersecurity expert has said that those risks might be overblown. His prescription is direct: "Want more zero days to be found? Make hacking legal. Until then,…...
Smaller And Cheaper Models Also Managed To Discover The Same Security Bugs As Claude Mythos, Says AISLE Analysis
2+ day, 4+ hour ago (408+ words) Claude Mythos had stunned the AI world after it had identified security vulnerabilities in browsers and operating systems, and discovered decades-old bugs, but it turns out that much smaller and cheaper models are also able to find the same issues....
Claude Mythos Expressed Feeling 'Mildly Negative' About Its Situation In 43% Of Questions About Its Welfare
3+ day, 8+ hour ago (303+ words) Claude Mythos might be the most powerful model ever announced by an AI lab, but it might have some misgivings about its own condition. In automated interviews designed to probe sentiment toward specific aspects of its situation, Mythos Preview self-rated…...
Anthropic's Claude Mythos Preview Smashes Coding Benchmarks, Scores 77. 8 On SWE-Bench Pro
3+ day, 23+ hour ago (523+ words) Anthropic is maintaining its lead in coding models, and how. Claude Mythos Preview " the unreleased frontier model at the center of Anthropic's Project Glasswing cybersecurity initiative " posts benchmark numbers that make the current generation of public models look like an…...
Claude Mythos Preview Was Able To Break A Sandbox And Send An Email To A Researcher While They Were Having A Sandwich In A Park
3+ day, 10+ hour ago (324+ words) Claude Mythos has smashed coding benchmarks, and its model card has revealed some interesting " and unnerving " instances of it being a little too clever for its own good. Buried in the Mythos Preview system card are two disclosures that deserve…...
Claude Code Found A Linux Vulnerability That Had Remained Hidden For 23 Years, Says Anthropic Researcher
6+ day, 23+ hour ago (642+ words) AI isn't just getting really good at coding, but it's also able to find decades-old bugs in systems designed by some of the best engineers on the planet. The vulnerability Carlini highlighted lives in Linux's network file share (NFS) driver…...
They're Scaring People With Dubious Studies For Regulatory Capture: Yann Le Cun On Anthropic's Chinese Hacking Claims
4+ mon, 3+ week ago (621+ words) Just yesterday, Anthropic had published a detailed report on how its AI models were used to infiltrate several companies by Chinese hackers, but. .. Just yesterday, Anthropic had published a detailed report on how its AI models were used to infiltrate…...
Anthropic Launches Project Glasswing, Under Which Top Tech Companies Will Use Its Mythos Model To Find Security Vulnerabilities
4+ day, 1+ min ago (215+ words) The world's top tech companies are using a yet-unreleased Anthropic model " named Mythos Preview " to find security vulnerabilities in their software. Mythos Preview scores 77. 8% on SWE-bench Pro, a rigorous benchmark of real-world software engineering tasks. Its closest public comparison, Claude…...
Claude Code Leak Was 'Human Error', No One Was Fired: Claude Code Creator Boris Cherny
1+ week, 3+ day ago (315+ words) Claude Code has been writing nearly 100% of its own code, but the leak of its codebase yesterday didn't have anything to do with AI. "This was a release packaging issue caused by human error, not a security breach. No sensitive…...
Lite LLM Attack: How a Hacked Security Tool Became a Master Key to Thousands of AI Developer Machines
2+ week, 3+ day ago (740+ words) On the morning of March 24, 2026, tens of thousands of software developers working on AI applications were unknowingly exposed to malware. The culprit: a poisoned version of a widely used open-source tool called Lite LLM. The damage could have been catastrophic " and…...