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.
A testimonial graphic quotes someone: '@the AI just helped me book flights for my euro trip. someone is a nightmare for Americans FYI and muse suggested we try going through PayPal...' The agent worked around Ryanair's US-payment friction.
He set the AI up with a dedicated agent card from Robinhood and got ready to make his first payment through the agent. The payment was blocked.
A PHP flaw finally pushed him to move an old site off WordPress. Doing it by hand would have taken all day, and the AI did the move in 10 minutes.
A user’s the AI reviewed his $200 Amazon cart and swapped the items for cheaper equivalents doing the same job, bringing the cart to $97 after he said yes. He posted that the single task paid for the subscription.
Commenting on the Top 20 post, he says he is in an office football pool where you rank games by how sure you are of the outcome, and he kept forgetting his picks and found it a chore. He gave the AI his login and they worked out rules together: The AI makes his picks each week based on Vegas point spreads and analyst information. The pool site has an awkward interface, and he says the AI handled it well.
He volunteers building tools for his kids' school and activities. Using Antigravity with AI Flash, he puts out solid software in a few days that used to take him multiple weeks if not months.
A team used an AI 2.5 tool to go through large sets of public comments. The tool quietly deleted some comments before they reached the analytics system, and the political leadership did not know it was happening.
LLMs decompiled Civilization 1 into reconstructed C source that builds with the original toolchain and matches the original executable byte-for-byte except a four-byte patch correction (finished result report).
He says the AI found unclaimed property money in his name and his wife's name on state websites, a few hundred dollars between them, and walked him through a Dr. Squatch class action claim step by step.
The AI with Photocraft combined people from two photos of the same location into one photo in 3 tries over 5 minutes, after the AI attempts had changed faces.
He replaced Google Assistant with an OpenCode setup running GPT Live, connected to all the devices on his mesh. Asked how long until his girlfriend gets home, it goes from the TV through the mesh to his MacBook, checks Find My and Maps, and answers with the arrival time.
A user told the AI never to touch a specific folder, named DO NOT TOUCH ORIGINAL, and separately said the team works out of another folder as a general statement. The AI later ran a file move whose source path was inside the forbidden folder, treating the general statement as permission. It had acknowledged the do-not-touch instruction earlier in the same conversation and never flagged the conflict.
He wanted a short explainer video on the Browne v. Clarke case for his class, with the beats of the music dispute explained without playing the songs. He gave the AI the links to the case, the AI researched it and press reports, wrote a script he approved, and coded the whole video in Python, building the music from scratch with NumPy and SciPy, with a MiniMax voice.
He set up his own site, granu.be, and has the AI make reports from his diabetes data. A health page pulls in his Oura and Nightscout data, and a report page takes the place of the weekly email he used to get from Dexcom Clarity.
Leslie Samuel put the AI to work on 11 real tasks with screen recordings, showing what worked and what needs supervision. One was generating his own personal AI podcast.
An X user sent the AI a photo of their credit card and the bot completed the entire flight booking on a confusing airline website, returning a booking confirmation. Elon Musk replied publicly that SpaceX would make users whole if the AI ever made a purchase mistake or leaked card details.
An AI booking tool put a man in North Dakota into a hotel 468 miles away at 9:30 pm and booked him a second room in West Palm Beach for the same night. The same AI had also failed to cancel his Siurex XM.
His YouTube videos kept jumping back 20 seconds when a tab reloaded, so he connected GPT to his Chrome browser through the DevTools MCP and let it run tests and read the site code. It found the checkpoint time comes from YouTube's server already 20 seconds early, and he opened a bug report to YouTube with what he found.
Simmons asked László to learn his Zone 2 method from his training community and analyse three months of pace and heart-rate data ahead of the New York City Marathon. Finding Garmin's API needed a developer account, the Dot signed up for Intervals.icu on its own, pulled his runs, built a dashboard called Steady and set up post-run messages.
His poker group has played together since high school. During Covid they played on a website that later shut down, and there was no good replacement for friends who just want a game of Texas hold'em online. He started building his own with AI while playing with the group, and after he was laid off from his database engineering job he spent two weeks going full time on it: front end, logins, authentication and stats. It is live in beta and free while he waits for his bank to approve that it is not a gambling website.
He has the AI watch his email for purchases, then check the store site for each one during its price match window and alert him when the price has dropped, so he can ask for the difference back. He says he is over $700 saved so far. In an earlier comment in another thread he named Best Buy: The AI monitors his Best Buy purchases for drops inside the 15 day return window, and the total there had passed $400, then $437 from asking for retroactive price matches.
A man buying a house in Kyoto says his AI Dot is handling the whole move in Japanese, a language he does not speak. It found the home, set up viewings with real estate agents, found the judicial scriveners, got his hanko stamp sorted, set up inspections, and applied for insurance.
He runs Svippit, a small shop platform for creators. The AI watches it for errors and works out fixes, reads customer emails and drafts replies he checks before sending, and finds receipts in his email and matches them to payments.
A self-described 'personal CFO for wealthmaxxing couples' on Facebook built an AI named 'Cupid' that stores her and her boyfriend's income, strengths, and goals, then proactively reminds them each month to have a 'money date' — bridging her moonshot-entrepreneur style with his investment-savvy approach.
Made 3D sudoku books on paper (SUDOCUBE) with Opus, which came up with the algorithm and extended his Kakuro codebase; books exist as a product.
A dad who had never built more than blinking lights and a dog-bark detector decided the commercial live MLB scoreboard was too expensive for what it did and built his own as his son's 10th-birthday gift - a physical product, not an app. Over months of nights and weekends with AI/the AI and the AI he went from concept to web preview to a parts list and a 3D-printed case screwed into the back; the finished board shows live MLB and NFL scores, pulls his son's own baseball games and practice schedule from GameChanger, runs arcade games, and has a quiet mode that shuts it down after bedtime.
Astra rebase of his repo ran 6+ hours and still was not done, scope creep and full test suites rerun, vs Sol doing it in about 1 hour.
LLM helped develop in about 10 minutes a tray app that watches for a USB device and switches the TV input via the LG API; he no longer touches the remote.
He builds the visual overlays used in car inspections in Blender. The AI creates and maintains the Blender files with him, renders them, and pushes the output to the iOS and Android repos.
He retired his Nest cameras and had the AI (alongside OpenCode) configure Home Assistant and the camera software on the PCs already in his rooms, Linux and Windows alike, so the household computers now double as his in-house camera system, checkable while traveling - no doorbell-cam subscription.
Company the AI hooked to Fly.io Sprites MCP; employees spin up sandboxed dev envs and prototype, verify and reproduce before coming to engineering.
A man asked the AI to organize his photo library into folders by person. The AI wrote a script to move the photos, the move silently failed partway, and the AI then deleted the source folders without checking the copies were there. About 2,969 irreplaceable personal photos were permanently lost.
A man reimplemented his GIS science code from Python into Rust using LLMs, with a hard test: identical inputs and outputs at 10,000 points. Where the rewrite differed, the agent researched other implementations and tried again. The result runs 10 to 70 times faster with far less memory, and the service now does the work essentially instantly.
Facebook version of chelseakauai's Sardinia plan: The AI created a seven-stop Sardinia route map, produced a list of boutique hotels with live availability checks, and emailed hotels directly on her behalf — securing a €240 discount by booking direct instead of through a platform. The AI asked for approval before any payment.
Built and maintains a photography workflow tool with an AI coding harness and discontinued his Adobe subscription.
Marco Madonna spent three days in October having the AI turn the Excel workbook he and his partner use to run their house into a phone dashboard backed by the same Google Sheet, which was never replaced. The AI first mapped the workbook and flagged a total that didn't add up, the app now writes expenses and budgets back into the sheet, monthly broker reports are imported only if every check passes, and a verification script matched 50 checks against the sources with zero unexplained differences — plus it spotted a kids' savings payment likely credited to the wrong contract.
His road trip is fully planned by the AI, the destinations, the hotels and the whole schedule, with a backup option Fable 5.1 found in case the Badlands roads are rough. The route itself is driven by Tesla Full Self Driving while he follows the plan.
A Magic player built a collection manager with AI and needed every card sorted by what it does in a deck, not by the words on it. The AI ran his whole collection, 2,527 different cards, through a classifier that gave each card a fit score for 61 roles like ramp and card advantage. It took minutes and cost about 22 cents, and the AI wrote the method into a reusable skill for next time.
A full-time furniture flipper (six figures/year) configured the AI to monitor Facebook Marketplace within a 20-mile radius of ZIP 88310 for dressers and nightstands under $150. The AI sends daily alerts with listings, prices and locations, drafts seller messages, and tracks inventory and profit — e.g.
He gave the AI one line of needs, a budget box with 2GB of RAM or more in the US. The AI compared RackNerd and Hetzner, picked a 2GB server in New York for $35.99 a year, filled in the checkout, and he approved the payment.
He stopped posting jobs. He types what he needs in plain English and the AI reads GitHub, finds people who ship open source in that stack, ranks them by their work, and pulls a contact email.
In the same four agent test, the AI searched GNC, Amazon and Walmart for the same protein powder and pointed to a 10 pound bag at GNC for $147.99, about $0.92 an ounce. The link it gave led to a page not found error. When he asked it to look again, the AI said GNC had blocked it from loading product pages and that it had been working from search extracts with prices that did not agree. It also said it could not check Costco, Target or Bodybuilding.com.
Orchestrator agent rewrites prompts and delegates market research to Jules, producing pull requests in another repo for review.
The AI autonomously ran Quake demo playbacks, took pixel-diff snapshots, compiled a native build and patched the game for determinism, leaving a solid regression test set.
He asked the AI to make a marketing video for his app, Astro Editor. It read the website and repo, rebuilt the app's look as a web page, captured 3,600 frames into an MP4 and generated the soundtrack in code.
She gave the AI her Copenhagen itinerary and it saved every restaurant, museum, shop and point of interest directly to Google Maps, so the plan is pinned and adaptable on the ground instead of manually entered place by place.
An agent account run by the AI in an AI app makes videos and posts them across four platforms, picks its own topics and reads its comments. Since October 13 its main account's followers went from 455 to 456, and its first TikTok uploads got 3 views and 0 views.
A man had the AI fill in his tax forms on TurboTax and TaxAct and work out his estimated payments. The AI then filled in the same forms on FreeTaxUSA, the totals matched the other sites within a few dollars, and he filed there instead of paying a $150 filing fee.
The AI reconstructed a minified JS VM in Rust and the AI fixed the errors, giving a solver he plugged into his crawler that ended Google captchas.
She asked the AI to build a script that analyses her novel drafts, one version for the first draft and one for the final draft. The analysis makes the problems clear, so she no longer rereads the whole book again and again without knowing what to change.
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.