you're paying real money, so here's exactly what you get, what it costs, and how to get your money back. plain language. no 40-page maze.
last updated: june 14, 2026
the setup ($199, once): we remote in and install a private, local AI on the computer you already own, pick a model that fits your machine, lock down privacy, and leave it working. one-time. you own it after.
managed (optional, $12/mo): after a setup, we keep it sharp — model upgrades, a quarterly checkup, priority human help. it's an add-on, not required.
teams: $499 sets up to 5 machines, +$99/mo optional managed.
prices are in usd. payments run through stripe — we never see or store your card. the setup is charged once. managed bills monthly until you cancel. that's the whole billing story.
we check your machine before you pay. if it genuinely can't run a useful local model, we tell you and you owe nothing.
if we somehow miss that and your computer can't run what we promised, you get a full refund of the setup fee. the setup also comes with 30 days of support to make sure everything stuck — if it didn't, we fix it or refund it. no drama, no script.
cancel anytime. email us or say the word and we stop billing — no further charges after your current month, no cancellation fee, no "are you sure" maze.
here's the part that matters: when you cancel, your AI keeps working. it lives on your machine. you just stop getting upgrades and a human on call. we're a mechanic, not a landlord — we can't and won't shut anything off.
the AI runs on your computer, not our servers. we sell a service (setting it up, keeping it current), not a subscription to the software itself. nobody can remotely disable it, expire it, or hold it hostage. that's the entire point of the brand.
the models we install (llama, qwen, mistral, gemma, and others) are open-source and made by third parties under their own licenses. we pick the best fit; we don't own them.
a local AI is a tool, not a professional. don't treat its output as medical, legal, or financial advice. open models are excellent for most everyday work and still trail the big clouds on the hardest reasoning — we're upfront about that line and won't pretend otherwise.
we'll always recommend the best model for your machine, but we can't guarantee a specific speed or answer on hardware we don't control.
to the extent the law allows, our liability for any claim is limited to what you paid us. we provide the service in good faith but can't be responsible for how you use your AI or for third-party software and models.
these terms are governed by the laws of Ontario, Canada. if we change them, we'll update this page and the date at the top.
fuckbigtech.ai is operated by Kinetic Labs Inc.
52 Birch Hill Lane, Oakville, Ontario, L6K 2N9, Canada.
questions, refunds, anything: [email protected]
see also our privacy policy — short and honest, same as this.