News

Like Humans, AI Agents Can Be Scammed

Developing human-like intelligence in machines is all very well, but it leads to the development of human-like failings too. Social engineering can prompt agents to share funds or sensitive data, according to new reports; they can be hacked too, if not properly secured. A recent incident saw a cryptocurrency user trick an agent named Freysa into sending $50,000 to their account from a prize pool it was managing.