💩 GPT-4 Hacker — Should We Be Worried?
Anonymous researchers tested whether AI can hack websites.
The GPT-4 neural #Network was tested on new vulnerabilities that haven't been patched yet. Out of 15 such security holes, #GPT-4 found and exploited 8.
💩 The secret to success lies in a smart approach. The AI created "sub-agents" to quickly perform specific tasks without wasting extra resources.
💩 Now, developers have to decide: limit the capabilities of the AI or allow it to operate at full potential, risking security?
