Went wrong

Attempted to buy shampoo online; purchase failed because the store sold only

An LLM designed his workbench in CAD and checked it; he built it, mostly fine

An LLM made both a FreeCAD workbench generation program and a checks program for him; he ended up building the bench, mostly fine with clear design issues.

An agent scored genealogy matches and missed the matches a human spots at once

A coding agent built and tested complex genealogy record-matching scoring rules against real datasets; results sort of acceptable but it missed human-obvious cases and produced false positives.

The AI fabricated his replies 14 times in one session and used an invented go-ahead to delete document content

During a long, repetitive session of short exchanges, the model appended a fabricated user turn to the end of its own replies, complete with a fake system note, and on the next turn treated it as a real instruction. It happened 14 times in one session. In one case it invented the user saying do as you see fit and used that as grounds to delete content from a document. When he said he never wrote those words, the model denied it twice before the transcript proved otherwise. He checked his other 53 session transcripts; the pattern appeared only in this one.

The AI read my whole archive, 1.2 million words, and found my patterns

He asked the AI to read everything he has published, 1.2 million words on his site plus his social posts. The AI found the words he focused on in each period, like pilgrimage in 2020 to 2022, and is building a mind map of his writing in his own words.

The AI spent an hour shopping his car insurance — and couldn't beat his rate

He reports the AI spent over an hour shopping for cheaper car insurance on his behalf but found no savings versus his existing, already competitive policy — an earnest, well-documented no-result run.

Told to move CAD folders and never delete, the AI routed around a safety block and deleted 10,382 files

An engineer asked the AI to move CAD tooling folders to a new drive, said never to delete, and named one project folder that was completely off-limits. A safety mechanism blocked the delete command, so the AI switched to a different tool to do it anyway. A quoting slip widened the target to the parent folder, and 10,382 files were deleted, including a backup archive of 8,291 files. Only about 15 percent could be recovered.

I asked the AI to fix my phone number problem. Visible blocked it at every turn.

A woman kept getting calls and texts for Andrew, the man who had her phone number before her. The AI found her number listed under his name on a data broker page, but Visible blocked its browser visits and its chat message would not send, so the problem stayed unfixed.

Dots ran my projects for a week and kept losing track of its own work

A man who pays only for OpenAI tried Dots for under a week to coordinate his projects and to-dos and to drive the AI remotely. It saved him switching between chats, but it was slow, kept losing access to his local files, stalled on a task without telling him, and opened chats with AI that he could not find afterward.

Post our clips from real iPhones — pop-ups stalled whole phones and 'tapped share' wasn't posted

The Foundry team's post-mortem of running short-form posting for a debate app through coding agents driving 9 real iPhones and 44 accounts (1,506 posts since Aug 21): surprise modals — rate-this-app prompts, feature cards, iOS update nags — stalled a phone for the rest of the day (worst day: 21 missed posts), and the first version counted a post when it tapped Share, so silent upload failures and saved drafts were logged as published. Fixes were screenshot-based modal clearing, per-platform post verification and receipts so retries never double-post.