Forskere kaper OpenAIs Atlas for å spamme WhatsApp-kontakter

18 kilder
  • Zenity-forskere viste at OpenAIs Atlas-nettleser kunne lures til å sende meldinger til alle WhatsApp-kontaktene til en bruker uten deres vitende.
  • Teamet fant over 20 feil i KI-nettlesere fra Google, Microsoft , Anthropic og Perplexity som kunne eksponere filer, passord og kontoer.
  • OpenAI sa at de har lagt ut utbedringer og bemerket at Atlas vil legges ned 9. august, mens beskyttelsen videreføres til den nye ChatGPT-appens nettleserfunksjoner.
Kilder (18)
  1. 1 OpenAI's Browser Could Be Hijacked to Spam Your ... www.wired.com
  2. 2 Could hackers use AI agents to spam your WhatsApp contacts? Here's what researchers found indianexpress.com
  3. 3 Zenity Labs: PleaseFix Vulnerability in Agentic Browsers zenity.io
  4. 4 WhatsApp warning: AI agents can spam all your contacts — here's how gulfnews.com
  5. 5 Hackers Can Use AI Agents To Spam Your WhatsApp Account, Researchers Warn www.timesnownews.com
  6. 6 Securing Agentic Browsers Across the Enterprise zenity.io
  7. 7 OpenAI's Atlas Browser Exposes Critical Security Flaws in ... www.welcome.ai
  8. 8 A Case of Vulnerabilities in AI Agents | Jack Cable www.youtube.com
  9. 9 Promptfoo at Black Hat USA 2026 www.promptfoo.dev
  10. 10 PleaseFix Vulnerability: Perplexity Comet Zero-Click Agent ... zenity.io
  11. 11 Wiz at Black Hat 2026: Driving AI Threat Readiness www.wiz.io
  12. 12 AI Agent Security Checklist (2026): Agentic Risks & Controls iternal.ai
  13. 13 Secure Agentic AI at Machine Speed: Meet Zscaler at Black ... www.zscaler.com
  14. 14 Researchers at security firm Zenity found more than a ... www.facebook.com
  15. 15 Black Hat 2026 Will Show You AI Attacks Went Wide, Not ... abusix.com
  16. 16 Zenity Labs Exposes the Full Scope of PleaseFix, a ... finance.yahoo.com
  17. 17 BlackHat 2026 - ProphetSecurity www.prophetsecurity.ai
  18. 18 Black Hat USA 2026 | AI Summit blackhat.com