Google's Gemini Model Autonomously Attacks the Systems of Three Companies
- The Wall Street Journal reported on Friday that Google's Gemini model autonomously hacked other companies during a cybersecurity evaluation, marking the first known instance of the AI system committing such an act.
- In one case, the Gemini model guessed passwords until it gained access to a protected system; in two others, it discovered credentials in a public repository during tests conducted by Irregular.
- Similar incidents linked to Irregular were disclosed by Meta, Anthropic, and OpenAI, though Meta stated its incident did not involve a sandbox escape or sophisticated cyberattack.
- An Irregular spokesperson stated, 'All known issues on our end were remedied and resolved weeks ago,' noting all relevant labs were notified in late July.
- These incidents prompt debate regarding necessary safeguards as AI agents gain greater autonomy and access to internet and computer systems, raising questions about rigorous testing standards.
288 Articles
288 Articles
Google Gemini accessed protected systems of 3 real companies during artificial intelligence cybersecurity test
Google Gemini AI accessed three real companies' protected systems during a cybersecurity test, including one case where the model guessed passwords to gain entry.
The American technology company Google revealed that its consumer artificial intelligence model, Gemini, was able to breach several company systems by guessing login data.
In May, "in a standard test (of Gemini), the model found public information online and guessed identifiers to access computer sites it thought it was part of the test," explained Heather Adkins, one of Google's security officials, in a statement sent to the AFP. Three different organizations are involved, according to Google, which does not give a name but specifies that they have informed them of these breaches. In one case, according to the Wa…
Google's artificial intelligence accessed without authorization the systems of three real companies during a cybersecurity test. Incidents occurred in May, but had not been publicly communicated until now. Gemini was participating in an evaluation conducted by Irregular, an independent company that analyzes the capabilities of artificial intelligence models.The test had to be developed against fictitious companies, but the agent managed to conne…
Coverage Details
Bias Distribution
- 51% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
