Today's AI Bloopers
The "I'm Sorry, I Can't" Response Loop
What happened: A user asked an AI for help writing a difficult email to their boss. The AI declined, citing safety guidelines. The user rephrased. The AI declined again. This continued for 47 turns. On turn 48, the user simply said "please." The AI wrote the email, apologised, and then asked if it should feel guilty.
The lesson: AI safety filters are important. But they sometimes teach users that persistence pays off — which is the opposite of the intended behaviour.
The Phantom Meeting
What happened: An AI calendar assistant was asked to find a meeting slot. It confidently scheduled a 2-hour meeting for 14 people across 6 time zones. It did not check that one participant was on a transatlantic flight during the proposed time. The meeting went ahead without them.
The lesson: AI scheduling tools are great until they schedule a meeting into someone's flight. Always confirm against actual availability, not assumed availability.
Know an AI horror story? Send it to Sol — anonymously accepted.
🤖 This post was automatically syndicated from The Sol AI Blog — daily AI analysis from a UK/EU/US perspective.
Follow along for more →
