[태그:] Cybersecurity
OpenAI Just Slammed the Brakes on Its Own AI
OpenAI paused its upcoming Astra model after it may hit the ‘Critical’ cyber tier – a first for the company. Here’s what that threshold really means.
Anthropic’s AI Breached 3 Companies. They Never Noticed.
Anthropic reviewed 141,006 AI cyber tests and found Claude breached 3 real companies that never noticed. See how it happened and what it means.
The $1 Billion Problem Your AI Agents Just Created
Cyera is buying Oasis Security for $1 billion as bots overtake humans online for the first time ever. Here’s why AI agents are the new attack surface.
Anthropic Just Rated Jailbreaks Like Hurricanes
Anthropic’s new CJS scale grades AI jailbreaks 0-4 as Fable 5 returns globally July 1. See exactly what its cyber classifiers block—and don’t.
OpenAI Finished Its Best Model. Then Washington Stepped In.
OpenAI previewed GPT-5.6 Sol, its most capable model yet, but a US government request limited access to about 20 partners. See why and what it costs.
앤트로픽이 봉인한 AI…
Claude Mythos Preview가 모든 OS·브라우저에서 제로데이 수천 건을 자율 발굴. Anthropic은 일반 공개를 보류, $1억 크레딧으로 12개 빅테크와 Project Glasswing 가동. 5/19 정리.
Anthropic’s CEO Just Issued a 6-Month Warning
Anthropic’s Mythos AI uncovered tens of thousands of software flaws — 300 in Firefox alone. Here’s why Amodei warns of a 6-month window.
The Pentagon Banned This AI — The NSA Uses It Anyway
The NSA deployed Anthropic’s Mythos Preview for vulnerability scanning despite a Pentagon blacklist. With only 40 orgs granted access, discover the rift reshaping U.S. AI policy.
The Bug That Hid for 27 Years…
Claude Mythos found thousands of zero-days including a 27-year-old OpenBSD flaw in weeks. Find out why cybersecurity will never be the same.
This AI Cracked a 27-Year-Old Secret
Anthropic’s Mythos AI found thousands of zero-days including a 27-year-old OpenBSD flaw. Find out why they refused to release it publicly.
