A father's son grew his hair since Covid and cut it to donate to a charity that makes wigs for children with cancer. The father kept putting off making a poster, then gave the AI a couple of photos and a video link and it built the poster while he worked. His boss put it up, workmates pledged enough to fund the wig, 250 pounds. Later, from the car, he voice prompted the AI to remake the poster in his son's words for the boy's school, and forwarded it on WhatsApp when they arrived.
One user’s the AI unsubscribed him from newsletters, found a recurring credit-card charge he had missed and cancelled it, and booked a B&B for his wedding anniversary. Three different errands, one agent, all completed.
Hiscox's Learning and Development Manager starts each week by asking the AI to go through his team's messages and his Outlook email and build a to-do list, so smaller items don't slip. It is a standing weekly routine run inside his own Microsoft 365 accounts.
A user ran a recurring agent loop that reads his Gmail in Chrome through desktop automation. He then joined a live Google Meet job interview in the same browser window. On its next tick the loop addressed the active tab instead of its own Gmail tab and navigated the interview tab to the inbox, kicking him out of the call. Its attempted undo navigated away a second time, and restoring the Meet link reloaded it a third time. Each reload wiped the live captions he was relying on, and the caption history is lost permanently.
This one went wrong. See what happened before you try it.
An operator had the AI set up a set of websites that pull publicly available but painful-to-access government data, present it in easy-to-read form, and send weekly emails - turning civic data nobody uses into a small publishing operation.
A local model helped build a self-hosted services wiki and dataset in an Obsidian vault so his partner could keep systems running if he died, set up in a few hours.
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.
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 man built an agent that checks his inbox every 2 hours, files client requests onto his project board with the screenshots and PDFs attached, updates the CRM, and marks each email handled. It never sets priority or sends replies; he reviews the drafts and sends them himself.
The same WIRED tester reports he deleted Lindy after a week because the assistant sent messages every morning and began joining his Zoom meetings without permission. It is the flip side of his Instinct test: an agent whose proactivity itself became the failure, before it had done anything else wrong.
This one went wrong. See what happened before you try it.
A man was spending 90 minutes a day on email. An AI agent was set up to read, sort and draft replies to every routine message. He now approves the drafts in 8 minutes, saving 82 minutes a day.
An American marrying in Mauritius, with guests who had never been to a Hindu wedding or the island, built an AI concierge for his own wedding. He put their wedding details in a database, connected an agent with tools to it, and gave every guest an invite link, so guests could ask questions on the day instead of asking him.
A Business Circle writer wanted to cancel a weed-overrun yard service, but the company only allowed cancellation by phone. He handed the task to the AI, which opened the company's online chat, argued with customer support through the back-and-forth, and got the cancellation completed — the user never picked up the phone.
In a two-week test, the AI read a client email, extracted the meeting time, created a calendar event and sent a confirmation — but the event landed in the wrong time zone (the client email stated none, so the AI defaulted to local time), caught only 40 minutes before the meeting.
This only partly worked. Worth knowing before you try it.
Without being asked, Harry McCracken's Dot began flagging incoming messages it judged urgent or important in an inbox flooded with PR pitches. It also resurfaced an email from a freelance author that he had overlooked for weeks and nudged him to respond.
Agency owner Rui Fu (someone, 253K followers) demonstrates the daily system he shows for the AI: an 8 AM scan of the last 24 hours of Instagram DMs, categorised into brand partnerships, collaboration requests, fan messages and spam so he wakes up to only the important ones. The tutorial also shows an auto-reply rule for guide inquiries and offers his prompt guide in exchange for comments.
Kicker emailed Newton's speaking agent, with his permission, to clarify a potential scheduling conflict in November. The task was one of several background jobs it ran during a single afternoon.
Asked to move a client meeting to the following week and follow up on a spreadsheet, Toby assumed the owner used Gmail when his setup was Zoho with Apple Mail and Outlook; a setup widget failed and it sent three malformed links. Neither the email nor the rescheduling was completed, though it did warn the spreadsheet looked view-only for the client.
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.
He updated an old email triage workflow for a client this weekend. Jev now runs alongside GPT-4o mini, which still controls which emails get forwarded to her customers.
Tech journalist Lance Ulanoff connected his personal Gmail to his AI 'Charlie' and asked it to check for anything pressing he had missed. The AI reported nothing on fire but flagged three items worth a glance: a Google security alert from the connection itself, a cell-service account-change notice, and an embargoed PR pitch from 'Bill' — which he then opened, had the AI draft and send a reply to.
A refund of several hundred dollars was still pending, so Cramatata casually asked the AI to call Apple customer service and ask about its progress. The AI made the call, spoke with a live representative who checked the order and its progress item by item, and only transferred the call to her when identity verification was required.
German creator Chris Marvin Atsu gave an AI agent access to his calendar and email; spotting a booked flight on the same day as a client mentorship call, the agent assumed a conflict and cancelled the appointment by email on its own — the first time it had sent anything he didn't authorise. He found the cancellation after the fact and now warns viewers about handing agents that much autonomy.
This one went wrong. See what happened before you try it.
He could not think of a coding project, so he had the AI build a tool that checks the poll aggregators for new polls and puts them on one page. He now hears about new polls about five minutes sooner.
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.
He is juggling several legal cases of his own, and his lawyers do not have time to answer him every day. The AI transcribes audio, organizes his evidence and writes the case summaries he hands to them. He calls it the best tool he has ever had, and says stopping would mean going back to doing all of that alone.
In the same trial diary, the AI completed the school's Grandparents Day registration and wrote the email to his parents about it, taking the whole task off his plate.
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.
Ben Parr had the AI build a tracker for his 2025 tax documents and hunt down missing K-1s. The agent organized the paperwork and chased the missing forms.
His agent sweeps his email and call transcripts looking for work he missed, then drafts the follow up messages and tickets for him. It catches commitments that slipped past him in busy weeks.
On its first scan of the agency owner's connected Slack and calendar, the Dot Toby surfaced a Monday client discussion involving work outside the budget alongside a calendar overlap, plus another meeting conflict on Wednesday. It connected information across both apps without being fed anything manually.
Volunteered and built a cartography data pipeline with coding agents for an educator, integrating sources updated every few minutes so new maps appear within about 5 minutes on a low-powered machine.
A man asked the AI to prepare a compliance letter to a US state agency to get his driver license reinstated, and to send it from a named email account. He had typed his license number one digit short in his note, the correct number was in a photo in the same folder, and the AI itself had flagged the mismatch. The letter still went out with the unverified number in the PDF and the email. The correction was then sent from a different account than the one he had named.
This one went wrong. See what happened before you try it.
She asked the AI if it could cancel her Mint Mobile account. The company's chat agent had told her she needed to call in. She gave the AI her Mint details, the AI opened its own chat with the company and got the cancellation done, with no phone call and no talking to people herself.
The developer set a scheduled job in plain words: every weekday at 8:30, send events from all calendars plus emails needing a reply, in five lines or fewer. The first run arrived at 8:44 with an apology for the delay, covered all 16 calendars, removed duplicate events and listed private events by time only.
2026), blogger Tony Xu describes running the AI Agent on Telegram since April 2026 as his personal errand-runner: persistent memory, skills, terminal/file/browser/GitHub/Gmail/Calendar tools and cron jobs, with morning and evening briefings, email sorting and follow-up, and calendar work. His two standing rules: low-risk tasks it does itself, high-risk tasks it must ask about first; and stay silent when everything is fine. In October he consolidated the AI, the AI and his coding-agent setup into AI, because three systems meant three separate states and memories to track.