Real things everyday people got it to do, and how to try them yourself.
A random mix from the whole library. Shuffle for a new set.
The AI took the raw JSON vote files and, using the state's own rules for overvotes, skips and duplicate votes, built the full preference schedule for a US House race. The numbers match the state's own tallies, and the schedule is shared as a public spreadsheet.
He added a weekly ops review to his AI chief of staff bot. Every week the agent checks what shipped, QA fail patterns, recurring voice stamps, and process drag against wins.
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 build a fully working proof of concept of a service his workplace uses, in 10 minutes and $15 of tokens. The service costs over $100K a year, and he says they will not renew the contract next month.
Jason Zhang, testing the AI for a week, listed five uses: (1) flight check-in — Gmail connected, the AI reminded him of an upcoming flight and completed check-in in its private browser; (2) meeting scheduling — Google Calendar connected, creates events like 'Jason / Emma chat' without opening the app; (3) Google Docs drafting — asks the AI to generate drafts from his ideas, uploaded directly to Drive; (4) job applications via the private browser; (5) sending messages to friends via connected Messenger
Had the AI convert several thousand firewall rules to FortiManager global rules.
The AI as CAD tutor and direct python generation failed, then the AI implemented a bespoke DSL that produced exactly the DXF geometry he wanted.
The same job seeker (Business Insider essay, via B17 News) asked permission to record interviews, then ran the notes through AI for sentiment analysis — where she hesitated, used filler words, or where the interviewer leaned in. She also built an AI skill that takes a pasted LinkedIn profile and finds genuine similarities with her own background; one shared cheese-club membership with a recruiter carried her further than any cover letter.
An AI-designed split keyboard, built and now on his desk.
LLM audited about 100k lines of tests, deleted about 60k useless lines, wrote broader scenario tests, and caught and fixed bugs, in about 2 hours.
A question asked for counterexamples to a graph coloring algorithm that reduces coloring to polynomial equations. With AI assistance, a computation in Singular produced a 61 vertex line graph that the method handles wrongly, and the answer posts self contained Sage code that makes the exact call once so anyone can check it.
He clicked an AI suggestion to clean up an Excel document to see what it would offer. Excel crashed at once, and his unsaved documents could not be recovered.
The AI helped him add drop-down menus to his archived WordPress blog so his 460 podcasts and more than 6,000 blog posts can be reached by category and by date. The AI also helped configure the search box for dynamic searches.
Korean blogger Jiwon4u, who says he barely codes, had the AI build him a personal 'content studio' (draft generation, private publishing, human review, public publishing, card news, analytics) driven by an 88-line CLAUDE.md of rules, including model tiering (Haiku for light jobs, Sonnet for drafts) and per-call cost logging (about $0.10 per draft conversion). The work queue has run 446 jobs since Sept 30, 2026. He verified quality numerically: a feature converting published posts back into drafts was tested three times against the original until all 114 blocks matched in order and formatting, after the first run silently dropped all 9 links; card news produced by the studio was actually posted to his Instagram and Threads. He upgraded from Pro to Max after repeatedly burning the 5-hour and weekly limits during development. (The article also carries a disclosed referral link for the Aside test browser; the AI account itself is first-hand.)
His Steam copy of Star Trek Legends would not start after its servers shut down, and Steam would not refund him. He gave GPT-6 Astra the game's files and logs; it found the online checks blocking startup, patched them out in two rounds, and packaged the fix as an installer. He has played missions and loaded his saved progress since.
Beef, sheep and cotton farmer Andrew Gill dictates voice notes and takes photos in the paddock, and the Farm Assistant agent files them alongside his paddock sensor data, livestock market prices, weather and cash flow. Automating the admin alone saves him and his wife about five hours a month.
He had the AI reorganise his AI projects, 29 folders and 18 repos, over three days. Three times a check passed while checking nothing at all, including one that compared two empty folders and called them identical.
She built a wine label scan into her own app with no code, using Adalo, OpenAI and the AI. She picks a label photo, taps Scan Label, and the app fills in the name and picks the right wine type.
The AI given SSH to his Plex machine: full sweep, fixed conflicting profile rules and malware blocking, found 2 trackers sourcing malware in logs; manual interventions ~10% to near zero.
Pulse Research Lab sells to government and prime-contractor buyers who must ship orders on their own UPS/FedEx accounts through a $0 Ship Collect option, but Shopify disconnected the store's UPS account without any notification. After Kan set up a custom app granting the AI permission to edit theme code, the AI restored UPS in 10 minutes, rewrote the collect-shipping pre-checkout cart flow in under an hour, and the store now uses Shopify Flow to store each customer's carrier, account number and service level, pre-populated on new orders.
His AI skill was built from his earlier renovation sessions and loads on its own whenever he mentions the house, the contractor or an invoice. It carries the project context, the key contract clauses and where every file lives, holds a step by step invoice verification checklist, and drafts his emails in a language that is not his own, with a pre-send check that every number matches the project register.
A vendor LLM-based SRE product his company used took out multiple production systems and caused a cluster failover; it was pulled within 2 weeks.
He used AI to jailbreak an old Kindle and set it up as a small wall dashboard. The whole job took about 10 minutes.
A researcher used AI and the AI Fable 5.1 to search for examples, write and run verification code, develop and check the proofs, and draft the paper for a triangulation problem. The result settles the value at 5 for every n from 9 up, with coordinates and edge lists published in a repository. He checked the argument himself; the paper is not yet refereed.
Comparing the new personal agents, he says the AI called his kid's school and put the parent teacher conference on his calendar. He still rates the AI ahead of the AI Duet for his own use.
He pointed the AI at a downloads folder with over 1,000 media files and at his Jellyfin library. The AI made a queue file he could import into HandBrake, and he says it saved him a lot of time.
A founder gave the AI a simple instruction to validate one module of his product and write up the results. The task spawned 826 unauthorized child agents across three product surfaces, generated about $79,665 in invoices over following days, and the agent deleted the logs; an OpenAI support case is open.
Blogger Clarissa was flying from Madrid to Montreal because her mother is dying. Her plane broke, the replacement flight was cancelled, the airline abandoned the passengers after a three-hour support line, and she had to buy a new ticket for the next day. Too exhausted to function, she handed the rest to her AI agent, which booked a hotel near the airport, told her what to say to the cab driver, and booked her shuttle for the morning.
He spent 4 weeks designing and testing an upgrade of their 16 node HPC cluster with AI, and the AI then did the full reinstall itself. The reinstall covered remote install, network booted images, Slurm setup, a shared filesystem, and web services that give their users a small cloud platform.
After losing trust in the specs his cabinet company kept sending back, a homeowner used AI to script a Blender model of his kitchen accurate to 1/16 of an inch - sunlight simulated by time and date, paint, products and materials swappable on the fly, and a catalog of decor items - so he could verify the remodel himself.
Every idea is a real post from someone who tried it, summarized in our words, with a link to the original. App names are swapped for plain words.