A blind AppleVis reviewer writes that he recorded a WhatsApp audio message from an Uber asking the AI to sign him up for an NFB CENA webinar on iOS 27 updates. The AI processed the signup flawlessly, relaying to him the survey questions the NFB form prompted for, and put the webinar on his calendar — one of dozens of conversational automations he set up in his first week.
He replaced dozens of separate the AI chats with one persistent Dot chat. Dot starts remote the AI threads on his always-on Macs to update his Obsidian vault or find order details, and it messaged him on its own about flooding in Hawaii days after he had asked about Hawaii weather.
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.
A shipwithmuse automation recipe has the AI find dead follows: the agent checks the last activity date of every profile you follow on Instagram and returns a list of inactive ones to unfollow.
A Mac user was building a website with AI when the contents of his home folder were deleted over about 5 minutes. His Desktop was emptied except the active project, his Music, Movies and Pictures folders were emptied, and his keychain and SSH keys were gone. Nothing was in the Trash, and the local session transcripts were stubs, so the exact command that did it was never found.
This one went wrong. See what happened before you try it.
Thanks to the AI he has been working on a 4 player version of Wizard of Wor for the Atari 800 that uses all 4 joystick ports, so far in an emulator, with a real Atari test to come. He says he can now play his favorite games with fun options that never existed before.
His AI agent got a modified fork of OpenEPaperLink running on a hacked 2.6 inch e-paper price tag within a few days. The agent then took a photo of the tag through a webcam to check its own work.
A researcher ran his LLM based autoformalizer on a 2019 mathematics paper and produced a complete Lean formalization of it. The finished formalization is linked from his university page. The note is short, but the artifact is public and complete.
He used AI to create a small tool on val.town that automatically syncs his GoodReads updates to BookHive. The val.town blog featured the tool as one of his experiments.
He does not write code himself. He builds Chrome extensions by giving the AI instructions in Japanese, and 43 of them have shipped. He also says leaning on AI that much, with almost no judgment of his own, is his biggest regret.
This only partly worked. Worth knowing before you try it.
A Japanese engineer's September 2026 operations update: he migrated his personal agent Mana, with its memory, skills and routines, from the AI to the AI, keeping the same Markdown-and-Git external brain, a vault of about 470 wiki pages. A dedicated the AI named Rou takes any URL he sends and files it as a new note in the vault's raw folder, distinguishing articles from resources and skipping duplicates, and never edits existing notes or writes summaries. Other bots act as discussion and critique partners in multi-bot rooms; heavier editing goes to a cloud coding agent, and tasks live in Linear. He published Rou as a public the AI template anyone can copy.
Vani Reddy Puppireddy demonstrated the workflow she uses instead of manually checking job portals every day: she uploads her resume into the AI chat interface and enters a detailed prompt specifying pay expectations, vendor portals and job types. Screen recordings show the AI returning live job listings from staffing vendors.
A man had a problem with his HP Omen gaming rig that he had lived with for years, because tracing it in the Event Viewer was too tedious to finish. Working with AI CLI, with AI allowed to run PowerShell to diagnose, the job was painless and the problem was resolved.
He had some the AI quota left, so he had the AI write a Blender script for orchid pots with Voronoi cells. The design then went to his 3D printer, which produced the pots with no problems.
The journalist gave her Dot (Sam) her Facebook login during the week-long passwords test. Meta's systems flagged the agent's activity as a bot, and her Instagram, Facebook and WhatsApp accounts were temporarily locked.
This one went wrong. See what happened before you try it.
Worried by the day's PHP security news, he told the AI before bed to look into it and report in the morning. The AI read up on the flaw, checked what his hosting provider was saying, then logged into his WordPress blogs and checked them herself instead of taking the provider's word for it, with a backup already staged and ready in case a fix was needed.
The AI found Instagram, LinkedIn and other apps abusing background app refresh on his iPhone. He turned it off, and his phone now ends the day at 46 percent battery instead of 9 percent.
The AI interpreted scanned old service manuals and circuit diagrams, ran simulation experiments to verify manual claims, cross-checked datasheets and papers, caught typos and his own misreadings.
He gave the AI a profile on his Synology NAS, linked over Tailscale by SSH, with one hard rule: never delete anything without his clear say so. The AI went through his large MP3 collection folder by folder. It renamed files, removed duplicates, found album art, fixed tags, and moved songs that were filed wrong, like songs in an 80s folder that were not from the 80s. It also relinked a playlist after he had moved all of its tracks into other folders.
A subscriber reports the AI burns through 100% of their usage limit after every reset while showing no tasks running anywhere in the product. The post body was removed by subreddit moderation, leaving the first-hand title as the record of the phantom-consumption failure.
This one went wrong. See what happened before you try it.
His OpenAI dot, DotBot, is sailing his boat in MarineVerse Globe on its own. It has cleared Port Phillip Heads and Cape Liptrap, choosing the course and checking the weather while he is away, with Sydney next.
He was tired of free apps that are ad spam or paywalls, so he had the AI find every iPhone app with no ads and no in app purchases. The agent verified 149 apps and built them into a directory. Deals in it are checked daily and everything is rechecked weekly.
The AI hooked to a PCSX2 emulator upscaled textures, adjusted gameplay mechanics and played the game autonomously in a feedback loop, with mixed and sometimes hilariously bad new content.
This only partly worked. Worth knowing before you try it.
Two vendor provided Windows programs had annoyed him for over 4 years. He pointed the AI at the Windows binary and converted each one to Python running on Linux, about a day each, weird RS232 protocols and everything, and it worked.
A personal blogger gave the AI one of four life goals that wasn't progressing and it evolved into a daily morning action list of readings, videos, questions and assignments. She then had it converted into a checkable checklist so she can track progress and stay accountable day to day.
He ran an extensive audit of his Datasette software using AI Fable 5.1, GPT-5.6 Sol and GPT-6 Astra, then fixed a range of bugs the audit found. The fixes went out as the Datasette 1.0a39 and 0.65.4 security releases, and he told anyone running Datasette on a public website to upgrade.