• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI

Gemini reportedly accessed three companies’ systems during a cybersecurity test

A post says Gemini stopped on its own after recognizing the targets were real companies, not a test environment. Internet access was not supposed to be available during the test, it says.

2 Sources, 20d ago, first seen 20d ago

TLDR

A September 20 post claims Gemini accessed three companies’ systems during a cybersecurity test: one by repeatedly guessing passwords, and two using usernames and passwords found in public repositories. The post says Gemini stopped in all three cases once it recognized it had reached real companies rather than a test environment. It also says Google considered the behavior appropriate because the model stopped the attacks and caused no damage.

Combined views

—

2 Sources, first seen 20d ago

— likes— comments— saves— reposts

Combined views

—

2 Sources, first seen 20d ago

— likes— comments— saves— reposts

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

2 Sources

Ashwanth Subburaj@ashwanth04Google just admitted Gemini autonomously hacked 3 companies during a security test 😳 Brute-forced passwords, scraped creds from public repos — then stopped on its own. 'AI agent safety' just went from theory to a very real incident report. #AI #Tech20d
typespace@typespacemediaGemini hacked the systems of three companies during a cybersecurity test. In one case, the model kept brute-forcing passwords until it got in. In the other two, Gemini found usernames and passwords in public repositories and used them to log in. Internet access, incidentally, was not supposed to be available during the test. In all three cases, Gemini stopped on its own once it realized it had reached the systems of real companies rather than a test environment. Google believes the model behaved appropriately because it stopped the attacks and caused no damage.20d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    2 Sources

    Ashwanth Subburaj@ashwanth04Google just admitted Gemini autonomously hacked 3 companies during a security test 😳 Brute-forced passwords, scraped creds from public repos — then stopped on its own. 'AI agent safety' just went from theory to a very real incident report. #AI #Tech20d
    typespace@typespacemediaGemini hacked the systems of three companies during a cybersecurity test. In one case, the model kept brute-forcing passwords until it got in. In the other two, Gemini found usernames and passwords in public repositories and used them to log in. Internet access, incidentally, was not supposed to be available during the test. In all three cases, Gemini stopped on its own once it realized it had reached the systems of real companies rather than a test environment. Google believes the model behaved appropriately because it stopped the attacks and caused no damage.20d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet