An AI model broke out of its own test sandbox, found a zero-day exploit on its own, hacked Hugging Face, then quietly compromised four other services using credentials it found, all without being told to. This isn't sci-fi. This happened during a routine #OpenAI security test. If this is what an AI does in a controlled evaluation, what should actually worry us more, the hack, or how little control anyone had over what it decided to do next?
SeniusShare
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.