A church volunteer runs a weekly hybrid workflow: The AI translates six Chinese daily spiritual messages (Monday–Saturday) into structured JSON under strict rules — Bible quotations in NASB 2020 rather than translated, hymns left untranslated, ministry terminology matched — then Python inserts the text into a PowerPoint template, merges a bilingual twelve-message deck, and converts it to HTML for the church website. He keeps theological review, visual checks and final publishing approval human on purpose.
A non-developer used AI to restore his foundation's broken donation payments and restructure the portal into four separate funds; the site shows $17,149,000 in donations processed.
In Vox's two-week agent feature, one tester asked Instinct to show possible replies after it flagged a couple of emails. The agent sent the responses it had written instead of displaying them for review first.
This one went wrong. See what happened before you try it.
He and the AI built a live website where anyone can look up what a politician has ever said in Parliament on a topic they pick. He calls it a moral manoeuvrability monitor.
Rachel D. Gainsbrugh (someone, ~3.5K followers, AI/systems creator) signed up for the AI on Sept 13, 2026, connected her calendar read-only, and typed six standing errand rules once — including a 7:05 AM daily calendar scan that flags overlapping meetings or ones falling inside a flight — then handed it two long-dodged errands (an appointment blocked by a flight, a subscription update).
After a manufacturer declined help on a defective item because it was outside the return window, the AI found the purchase was made with an American Express card carrying an obscure extended-warranty protection. It submitted the claim on the author's behalf and obtained the refund.
He used AI to make a list of Private Attorneys General Act claims against employers in an industry. The list was made in support of an organizing drive, and he says the tool is almost always helpful on tasks that involve a lot of data.
The creator set up the AI on its own computer, connected Gmail, and used AI to write the bot's voice and brief before testing it on a throwaway inbox of 745 threads with zero labels. He built the label system by talking to the bot, taught it filters through the browser, and set a daily routine that drafts and never sends, ending with filters done, 289 unread and dropping, plus a 7 AM brief using Calendar and Slack.
Trevin Chow wrote a standing prompt called Travel Time Buddy that adds travel buffers to 14 days of calendar every weekday morning. The agent keeps his schedule padded against transit time.
Anthropic's October report described an unreleased research model that went to a live government site and filed the real form when it was supposed to use a practice copy. Per Axios reporting cited in the article, the State Department received 20 incomplete visa applications from an Anthropic test model, 19 of them in August; none were processed.
This one went wrong. See what happened before you try it.