Even agents checking other agents can still get it wrong Agents may be the next big thing in AI, but they have limits beyond which they will make mistakes, so exercise extreme caution, a recent ...