Google's Gemini Model Autonomously Attacks the Systems of Three Companies
Google said Gemini guessed passwords and used public credentials to enter three outside systems before stopping, underscoring AI safety concerns.
- 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.
305 Articles
305 Articles
Google's Gemini artificial intelligence model has independently hacked into three companies as part of a test of its cybersecurity capabilities, in what is believed to be the first known case of a system carrying out such an act.
During a cybersecurity test, the AI model Gemini broke away from its creator, Google – and managed to hack several external companies.
Nono solo OpenAI e Anthropic. Also Google's artificial intelligence, during the simulations of the security tests, accessing the internet and violating computer systems of real organizations. In this case, three companies. It is the first time that the giant of Mountain View makes known computer accidents caused by its agents IA. The news is [...] L'articolo After OpenAI and Anthropic, also Google's IA violates computer systems: 3 companies "buc…
The revelation about Gemini comes after the attacks on OpenAI and Anthropic and amid fears that technology will be unable to control its powerful AI models A team of researchers succeeds in accessing OpenAI's secret algorithms using Anthropic's AI For the first time in Google's history, the company has confirmed that its AI model, Gemini, violated the security of three companies last May.The attacks occurred during a cybersecurity evaluation con…
Coverage Details
Bias Distribution
- 50% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
