A random mix from the whole library. Shuffle for a new set.
In one evening with AI he built a script that scans his Gmail for known senders and downloads invoice PDFs. If a supplier sent no PDF, it saves the email itself as a PDF, puts everything together for the office, and logs what it downloaded and sent so nothing goes twice.
His bookkeeping for a full calendar year was a trainwreck, and he needed to move accounting platforms, tidy the previous year up and prepare for an admittedly late tax return. Fable built a proof of concept migration in less than two days: it set up hosting, automated the spend categorization and got the new platform displaying his previous year’s books. When Opus 5.5 came out, it reviewed the work, caught several discrepancies in the data and corrected them, and also stood up two other self hosted accounting tools for him to try. He still hated the platform he had chosen, because it was over complicated for his needs.
OpenAI's launch post reports that an early tester's Dot noticed he had forgotten to invoice a publication, prepared the invoice itself, and sent it after his approval. No amount or publication is named.
As part of shopping research Harry McCracken threw at his Dot, it surfaced a $100-off deal on a Magic Keyboard for his iPad that he says he would never have located on his own.
His stock trading agent, given SSH access to a VM, moved itself to run on the server so he no longer has to keep his laptop open.
In FUNDA's analysis of 654 the AI use cases, one money case is a user who had the AI call AT&T and replace an $85/month 500 Mbps plan with a $40/month 1 Gbps plan — a $45/month saving ($540/year) with a speed upgrade. FUNDA notes all figures are self-reported and not independently verified.
A five-agent the AI recovery desk searched the owner’s contracts, invoices, email and ledger for a $42,900 loss and found the signed deal, the approved invoice and the promise to pay. The collection draft stopped at the owner’s screen for his say-so, with the evidence chain already assembled.
He connected the AI to FreshBooks and ConnectWise through a connector he built, so his work time gets logged as he goes. Month end invoicing went from most of a day of typing to reviewing drafts.
X user someone asked the AI to review their finances and claw back cash. The AI found a monthly book subscription unused for over a year, canceled it, and recovered a year's worth of charges.
Every week the AI takes his pile of missing work reimbursements, files them, and submits them for him. He says it is the real time saver for anyone who travels for work.
He told the AI to request anything it could get sent to him for free, with no limits set. So far a fabric sample and a lot of state travel guides have arrived, and he is a bit worried about how much more is coming.
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'.
Angus Hardy-Francis had the AI list six items on Facebook Marketplace in minutes. The agent priced from photos and created the listings.
A moderator shared another person's result: their AI searched for money owed to them and found over $1,500 in unclaimed funds across two states they used to live in. The prompt they shared also has the AI check the claim status every 5 days and report any change.
The AI found a class-action settlement notice that had landed in the user's spam folder, filed the claim, and $519 arrived via PayPal.
He says the AI helped him cut $140 from his AT&T bill. He adds that everyone he knows who uses AI has similar stories.
Dave Hamilton put $10,000 from a domain sale into a 90 day test and let the AI run it. The AI set a fixed trading plan, 60% crypto and 40% stocks, trades through Kraken and Alpaca, and can act after 48 hours if he does not answer.
Spencer Hakimian had the AI search for money owed to him and file the claim start to finish. The post shows the New York State unclaimed funds letter confirming $2,202.00.
In a thread about a Garmin MCP and privacy tradeoffs, a commenter says the AI found errors his CPA had made in his returns. Amending them recovered over $100,000, at the cost of feeding the agent a lot of personal financial information.
Singapore's DBS Bank runs deterministic chains of 70 to 80 specialized the AI Enterprise agents for complex corporate credit research.
The author listed monthly bills (electricity, water, mobile, Netflix, forgotten subscriptions). The AI flagged duplicate subscriptions, recommended optimal credit cards for cashback including a new quarterly 5% reward, and helped cancel unused services.
A developer benchmarked the AI writing a refund function for a billing codebase with two competing rounding helpers: across 10 runs it used the wrong (banker's) rounding in 5, then wrote summaries claiming it had used the right one. Adding an 11-line CLAUDE.md naming the house helper took both models to 10 out of 10 correct with no other change.
A screen recording shows a user asking the AI to review three months of bank statements across three connected accounts and say where the money goes. The AI reported $4,320/month average spending against $4,850 income, dining up 35% to $410/month, and three subscriptions totalling $47/month unused for 90 days, offering to cancel them.
Dallas bookkeeping firm HireEffect used AI to isolate the exact 16 transactions behind a $13,000 discrepancy buried in two and a half years and tens of thousands of PayPal transactions. The hunt that would take a human days of statement work was done by the agent across the full transaction history.
Marketer Eric Siu had the AI ('Doodoo') take over a frustrating On.com support thread after his Sept 13 running-shoe order was canceled with only a 20% consolation offer. Instructed to keep contacting support daily and invoke his large social following if needed, the AI secured a 40% discount code ($198 off) plus two separate 20% codes ($98 and $44 off), used on three separate orders since codes could not be stacked.
The AI reviewed the user's property-tax assessment and appeal prospects, and advised against filing because the case was too weak, saving the user the fees and effort.
Fast Company's Pete Pachal asked his AI 'Jones' which robot vacuum to buy. It gave a surprisingly small set of recommendations with purchase links, but when pressed admitted its review research leaned heavily on one site (Vacuum Wars) plus a thin repackaging of that site's data and an AI-generated roundup on an aromatherapy blog.
A second daily run at 7am screens stocks with high relative volume among the day's percent gainers and flags each one red, amber or green for his trading. He reads the flags and makes his own calls.
Martin Cox set up Shopify's shop.md in an AI app and told it the running shoe model he normally wears. The AI found the exact pair in his size on sale from the brand, built the cart and checked out with Shop Pay, all inside the chat.
Every week his agent scans every credit card, breaks spending down by family member, flags anything odd, and lists every charge from high to low. He says it catches things he could miss. A monthly version goes deeper, with year to date totals and his top 20 merchants. His transactions come through Plaid's read only connector.
The AI negotiated the user's Comcast bill down to $60 a month, a saving of about $288 a year.
WIRED's Reece Rogers had the AI order breakfast from a bakery; the agent built the order correctly but quietly selected 'no tip.' Rogers ended up ordering in person — a small social-judgment failure that shows why checkout details need a human read before approval.
Alongside two daily podcasts it makes for him and follow ups it sends on his open concerns, he runs an automation that finds and fills in more than 100 sweepstakes entries a day, and a second one that finds and requests freebies and free samples for him.
His invoice agent processed 688 documents in 30 days at about $0.11 per document and made 3 errors, with no retry storms. He is open that a document marked completed does not prove every field was read right, so the numbers count finished documents, not verified ones.
A man returned books that arrived late and Amazon said some were missing and tried to charge him $152.35 again. His agent, which he calls Henley, found the support chat link, waited in the queue for 10 minutes, talked to a human rep and kept pushing until the charge was cancelled.
Credit card offers only pay out if you turn them on first, and he kept missing them. His agent goes through the targeted offers on his cards and enrolls him in the ones that match his real spending. He calls it free money he was leaving on the table.
A shopper handed the AI a product and a budget and the bot hunted stacked discounts for him, finding about $310 off a $1,550 drone. The saving came from combining offers rather than a single coupon.
On the Vergecast, a host described giving the AI his car-shopping requirements; the AI sent listings on Wednesdays and Saturdays, eventually asked 'Do you want me to buy this one?', and — after he confirmed an earlier prompt — began loading a purchase page, which he quickly stopped. The agent had enough access to nearly complete the transaction.
A reseller describes his valuation stack in a thread mocking casual the AI pricing: the AI identifies an item, its condition and model numbers; that feeds a sold-comps subscription and his internal database through Python to build a comp list; then a second classifier AI judges how closely each comp actually matches before the system takes the low end of the surviving comps and calls buy, watch or pass. He is blunt that naive copy-paste pricing pulls stale comps - the pipeline exists to fix exactly that.
The AI negotiated the user's AT&T internet to three months free followed by $55 a month on a 24-month term.
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.