Skip to content
Terence Eden’s Blog·

🤖Ocado Sends AI Reply to Customer Bug Report, Then Apologizes Personally

Your bug report might get an AI brush-off

TL;DR

A developer reported a calendar invite bug to Ocado and got an obviously AI-generated reply from 'Lulu'. A low survey rating led to a personal apology and a human follow-up. The lesson: if your product has no real bug channel, customers will find the workaround themselves.

Ocado's support team sent a developer an AI-written reply about a broken calendar invite. The developer rated it poorly on the satisfaction survey, and that rating got them a personal apology. Honestly? That's the best outcome for the customer and the worst signal for the company. Why should you care? Because this is what happens when a bug report has nowhere real to go. The author says Ocado doesn't have an easy path to its technical team or a bug tracker, so support became the only door, and support answered with a bot. If you own a customer-facing API or integration, ask yourself where a detailed diagnostic report would actually land. The details are small but telling. The first reply came from 'Lulu' and read as AI. The survey followed, and the author gave low marks across it. The apology thanked them for detailed diagnostic feedback and promised it would reach the technical team, and the follow-up finally engaged with the original message. The author's line is simple: summarising AI output is fine, replying to a human with it is not.

Ocado Sends AI Reply to Customer Bug Report, Then Apologizes Personally — Terence Eden’s Blog

Key Points

1

The developer contacted Ocado about a technical problem with calendar invites and received a reply from an agent named Lulu that was judged to be AI-generated.

Keep reading — it's free

Enter your email to keep reading — plus the best of AI & tech, daily. Free, forever.

Also get
or

Already a member? Sign in

AI ethicscustomer supportbug reportingCSATchatbots

Comments

Subscribe to join the conversation...

Be the first to comment