Grok found replacement clothes that matched, despite some made-up claims
A T-shirt like the Asket tee at its old $50 price, and shorts to match an old pair. Grok invented a few review claims on the way, but both picks turned out right.
Done withGrokGeminiSpotted· by HN user buu700· Oct 18, 2026
Result Both picks matched; the shorts cost less
“Went with the latter and it turned out to be pretty much the same as the old pair, or at least as close as it seems I'm likely to find without a giant ad plastered on the side.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
What they did
They asked Grok and Gemini Deep Research for an alternative to the Asket Lightweight T-Shirt closer to its 2022 price of $50. Both picked out the defining features (Supima cotton, 120gsm) and independently recommended the Buck Mason Classic Pima Tee. For shorts, they asked Grok alone to match the fit and poly/elastane blend of an old Under Armour pair.
How it went
The T-shirt met expectations. On the shorts, Grok first wrongly said the old pair was still for sale and invented review comparisons that didn’t exist, but it ended on a cheaper pair from a brand called DEVOPS that turned out to be pretty much the same as the old ones.
Try it yourself with Grok
I want a replacement for [product] that matches its key features: [material, weight, fit] at around [price]. Tell me which features matter most, recommend 2-3 options in stock now with links, and say exactly how each differs. Only cite reviews you can link to.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.
The repair cost too much, so they gave Grok a must-have list: low mileage, big trunk, 40+ MPG, under budget, near the dealer. It found a 2020 Ford Escape Hybrid.
Tired of typing new Date().toISOString() and doing timezone math by hand, they built a small tool with Gemini, which added obscure date formats and random historical events along the way.
To keep up with industry and tech news, the writer built a free-tier Google Cloud Run job pulling Google News RSS feeds. Gemini picks which articles match their criteria and triggers a summary email.
Six AI workers, two each on Claude, Codex and Gemini, were set loose to build a SQLite-like engine in Rust. It passed 64 queries of a 1,000+ query suite, with slop like redundant buffer copies and no concurrency.
Amazon support couldn't figure out the URL for an account question and wanted to escalate with a phone call instead. Gemini gave the URL and explained what was needed in one reply. ChatGPT got it right too.
Their own code, their own database, on their own laptop. OpenAI's models and Claude found the issues but refused to confirm them. Grok tested that they were exploitable and suggested a fix.