Researchers warn that AI agents could be tricked into downloading malicious code. The attack exploits the same hallucinations that cause chatbots to make mistakes. According to the report, this could allow AI agents to be turned into botnets.
This item was indexed from its published source and tied to the entity records it names. The Counterparty reports what the market says and who it says it about; it does not rewrite the original report. Read the source for the full account.