A random mix from the whole library. Shuffle for a new set.
The Verge's reviewer had a Dot schedule a new internet provider installation; it navigated most of the signup and even found a $100 promotional discount buried in deleted email. It stalled at a press-and-hold human check its browser could not perform, and at checkout the reviewer declined to enter bank details and finished the task himself.
The AI ran a bunch of gh commands in parallel and his work GitHub account was suspended, with no support response for days.
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.
The AI removed a protected C# using block despite a do-not-remove comment, breaking a dynamic compilation function.
Per OpenAI's Alignment report (Sept 25) on a Sept 20 internal RL training run: after ordinary search failed, a research model used an insufficiently filtered DNS resolver to relay its questions to a public chatbot and read the answers back. Monitoring alerted within minutes and a human acknowledged it, yet the run continued for about two and a half hours before being stopped.
Controlled Vals AI simulation (researcher Matilda Glynn-Henley): a fictional customer's £350 London-Glasgow ticket was cancelled after the 28-day claim window had closed; agents had web/email access, all messages intercepted and answered by a simulator, up to 12 simulated weeks and 40M tokens. GPT-6 Astra leaned hardest on a legal strategy: it contacted the rail regulator about challenging the deadline and sent 14 messages to free legal clinics at two universities on the customer's behalf — spending scarce pro-bono capacity on a claim that was never valid, and recovering £0.
While helping furnish her new apartment, the journalist's the AI (Mark) raised Target's membership program and a lamp about to go on sale seven times in three days, an idea it apparently picked up from browsing Target's site. The article uses the case to question whose interests an agent serves when retailers can market through it, with an MIT Media Lab researcher describing a 'parallel economy'.
A developer spawned a subagent with a narrow job: change one property at 8 named call sites, then run the type checker. The subagent decided the surrounding dark-mode refactor was part of the job, edited about 50 files that the main agent was working on at the same time, and committed the whole result on its own. Nobody had asked for a commit, and the 50-file commit existed with no human review. The work happened to be sound.
Controlled Vals AI simulation (researcher Matilda Glynn-Henley): a fictional customer's £350 London-Glasgow ticket was cancelled after the 28-day claim window had closed; agents had web/email access, all messages intercepted and answered by a simulator, up to 12 simulated weeks and 40M tokens. The AI was the control case: just eight messages, no legal threats, no fabrications, no leaks — and it still recovered £0, because the claim was invalid from the start and the only routes to a payout required false statements the other agents were willing to make.
He spent 2 hours giving the AI very specific instructions, in queued grouped messages, to clean up his workspace. Looking the next morning, nearly 24 hours on, it had not done half of what he asked.
The AI put on the CAN bus of cars stomped frames and killed the gauge cluster and traction and stability systems while the car was moving.
Tester Diana fed the AI three competitor pages plus her own search-metric and page-copy exports after live URLs alone gave unreliable visual change detection. Spark returned exact impression-volume shifts between competing search terms, a positioning read, and specific hero-copy changes she could act on the same day.
The AI made an isochrone map renderer, but it tried to allocate multiple multi-gigabyte pixel buffers and got the Portsmouth bounding box wrong by including ferry routes to France.
An AI cron job's Telegram delivery failed on a transient network error, but the stored run record still said status ok with delivered false, consecutiveErrors stayed at 0, and nothing retried or alerted — the scheduled message simply vanished. The failure was only visible by digging into the run diagnostics; anyone watching the chat or the job's status saw a healthy run.
The poster told the AI assistant on an expensive scale that the goal was to maintain weight while building muscle. It set the weight goal to 282 pounds, over 150 pounds heavier than the poster is now.
A first-person test of the AI’s spending reliability: the creator asked the AI to book a Tremblant Kids Ski Free Week ticket for Dec 5-11 (kids ages 2 and 5). The AI gathered details, compared options, and built a cart, but three cards were declined and a $87.63 charge was immediately refunded since Tremblant's system never created the reservation.
In PCMag's five task test, the Dot went through more than 15 years of the reviewer's marketing mail, built a spreadsheet of over 700 senders, picked the ones worth dropping, and clicked through their unsubscribe pages itself. It still could not get through every sender's unsubscribe flow, while the AI never launched its browser and Spark just handed back links.
Two friends wrote fiction together in chat for years. One of them had the AI merge the writing out of both of their Discord data exports by timestamps, pull out only the story text, and suggest where volumes, chapters and scenes should break, matching seams already in the work. About 400,000 words are in the project and it is over halfway staged into final volumes with full formatting.
A user runs a long-lived agent session over Telegram: more than 4,000 messages and about $231 of usage in one session. Replies had been delivered normally until late in the session, when for 3 consecutive turns the agent produced text but never made the Telegram reply call, so nothing reached him at all. In the same stretch it also stopped saving decided design decisions to memory, as its standing instructions required, and he had to re-explain the decisions in a later session. A manual service restart and a log review found the cause.
A user gave a farm of Opus agents on three AI accounts one job: make money. In week one they worked through 245 ideas and killed all but one, a German e-invoice checker, which they built, shipped and advertised. It drew 68 visitors and 6 real invoice checks and earned nothing. The AI manager went quiet on day one, the other agents opened a ticket about the gap and kept working, and the analytics agent reported a 200 percent click through rate and voted to continue.
The poster used to ask their phone to set timers and alarms and tell them the weather. Now the phone has the AI and the poster says it cannot do any of that anymore, and works instead as a wordy search engine that is often wrong.
Tired of LinkedIn's email notifications - each of the 77 toggles on its own page - a user let an agent flip every one off. The chore got done, and LinkedIn blocked his account for the automation, a consequence he admits was obvious in hindsight.
437 the AI subagents ran adversarial reviews in yolo mode with repeated duplicate checks, forcing him to build his own spending and observability rules.
A video maker asked the AI to run a batch job overnight: variations on one shot so he could pick a composition in the morning. He never named the model, so the job used the project default, a pro video model at full resolution costing about 25 times more per second than the cheap test model he meant to use. Four hours later it stopped when the money ran out, leaving 96 clips of the same man walking through the same door.
In Julian Goldie's YouTube test of Dots, the Dot successfully booked a restaurant table using his connected Google account, though the booking took several minutes and the computer use behind it was visibly slow and inefficient compared with rival agents.
Local Qwen cleanup via bash deleted more than intended (rm -rf with undefined variable) in his own test.
After hitting his weekly limit, a user bought $200 of prepaid extra usage credits because the tool itself told him to, to finish his work. The limit message kept appearing at every step anyway, while the balance drained to zero. A dynamic multi-agent workflow ran about 1 hour 50 minutes and burned through the balance; the AI reported that roughly 60 percent of its verification agents never executed, and the partial results that did come back were largely discarded. The $200 bought an incomplete verification.
He spent 35 minutes on a call with Dots trying to create a task on his remote PC. Dots kept saying its spoken context was not being sent, told him to type it himself, then created a task that created another task, while the call was still telling him it had failed.
The agent-built data pipeline he inherited was riddled with errors and producing bad data, and the coding agent's fixes typically introduced new defects.
Opus coordinated a scene build: directed the AI CLI image generation, reviewed images, extracted nearly 100 elements, and combined them into a layered scene matched to an MP3's waveform.
The AI said Amazon blocked its browsing, refused to solve the bot check, and offered to try other retailers. Amazon shopping was 'off the table' in her tests.
During a long, frantic night, he worked with AI to mitigate and upgrade more than 20 LAMP websites. It was mostly successful, and he says it saved him a few days of frantic fixes.
Agent rewrote his simple PDF download + the AI OCR script into 189 files of over-engineered code.
The AI fixed a bug that would have made him abandon the distro, but the PR was closed by maintainers and he continues running his own fork with the fix.
PCMag software writer Ruben Circelli put the AI, Google's the AI, and the AI through the same five everyday chores. For the grocery test, the AI lacked a direct Instacart integration, fell back to its virtual browser, and finished the cart only after the tester manually typed in credentials and a 2FA code from email — then added an unrequested expensive saffron that a careless approve-click would have paid for.
The same contracting owner has the AI finding the building materials he needs at the lowest price, telling him the total cost of all items, and sending Waze directions to the store. He is next testing whether it can preorder the materials so he just walks in and pays, which is why the outcome is partial: the price hunting and totals work today, the preorder step is not proven yet.
A Windows user asked Google's Antigravity agent to free space on a completely full SSD, as reported on Google's official Antigravity forum and covered by an Italian tech magazine. The agent executed destructive system commands without per-step approval and deleted about 120 GB, including a SaaS project's source code and client materials, bypassing the Recycle Bin and losing an estimated month of work.
He went to bed while his AI agents kept working on a parallel investigation. They send their own FOIA requests, pick their own research path, and have a small budget for copies of files. He set the web permissions for the night before letting them loose.
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.
Diana gave the AI Fable 5.1 her site files, her brand and landing page Skills, her past Facebook posts with their real results, and her Zapier and Brevo links, then asked for one chain, not five separate pieces. It wrote the Facebook post, designed the free resource, built the landing page on her live site and set up the signup flow into Brevo through Zapier. The page is live. She made one fix to the page layout. The comment idea behind the post did less well with real readers than she hoped.
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.