ProPublica: Anthropics KI finner feil raskere enn Microsoft klarer å rette dem

18 kilder
  • Anthropics Claude Mythos-KI har avdekket hundrevis av kritiske feil i Microsoft -produkter, noe som går raskere enn ingeniørenes evne til å tette dem, ifølge ProPublica.
  • Microsoft lanserte sikkerhetsoppdateringer for over 600 sårbarheter den 14. juli, en rekord, hvorav minst to allerede er utnyttet i angrep.
  • En tidligere NSA-topp som rådgir Anthropic, advarer om at standard prioritering kan undervurdere risikoen fordi Mythos kan kombinere mindre feil til store angrep.
Kilder (18)
  1. 1 Anthropic's New AI Model Can Identify More Software Bugs Than Ever. Microsoft Is Struggling to Fix Them Fast Enough. www.propublica.org
  2. 2 Microsoft Patches Record 622 Flaws, Including Two Zero- ... thehackernews.com
  3. 3 The July 2026 Security Update Review - Zero Day Initiative www.thezdi.com
  4. 4 Microsoft Fixes Record 600+ Vulnerabilities In July 2026 ... www.linkedin.com
  5. 5 Hitachi uses Claude Mythos to cut code security reviews to hours jp.ibtimes.com
  6. 6 AI cyber threat 'months, not years' away, intelligence ... www.euronews.com
  7. 7 Looming AI-fueled threats require urgent ... www.cybersecuritydive.com
  8. 8 Anthropic Is Finding Bugs Faster Than Microsoft Can Fix Them ... fourweekmba.com
  9. 9 Anthropic's Mythos Finds 231 Microsoft Vulnerabilities Faster ... aigovernance.com
  10. 10 Microsoft July 2026 Patch Tuesday: 569 Vulnerabilities Fixed stackcyber.com
  11. 11 “Five Eyes” Intelligence Alliance Warns AI Models Pose ... www.democracynow.org
  12. 12 Anthropic: Claude Mythos identified 10000+ software flaws www.helpnetsecurity.com
  13. 13 Five Eyes Warns AI Models Capable of Devastating Cyber ... www.linkedin.com
  14. 14 Anthropic's Mythos And The Cyber Good Guys vs Bad www.forbes.com
  15. 15 Anthropic Claude Mythos and the 2026 Cybersecurity ... www.radware.com
  16. 16 Anthropic says Mythos has already found more than ... www.reddit.com
  17. 17 Microsoft Patches a Record 570 Security Flaws krebsonsecurity.com
  18. 18 AI could breach government and business defenses in ... www.cnn.com