You just used a
research instrument.
Only Humans Score isn't a novelty. It's a working study of the one question every organization now has to answer in private: where does the machine genuinely add value, and where does a human still have to decide? The game makes a room answer it out loud, with money and pride on the line. I build instruments like that — and I help teams answer the same question about their own AI.
What you just watched is the work
This whole platform is the résumé. Nothing here was bought off a shelf:
Models in production, safely
A generative image model wired into a live product behind a content-safety wall, rate limits, and a graceful fallback — the un-sexy parts that decide whether AI ships or embarrasses you.
Where humans must stay
The entire game is an argument about which calls a machine may make and which it may not. That's not a slogan here — it's the architecture. The same line has to be drawn inside your org.
End to end, by one person
Design, edge functions, offline PWA, payments architecture, WCAG-audited accessibility, IP and authorship strategy — the full distance from idea to a thing in your hands.
If your organization is building with AI, you should talk to me.
Most teams deploying AI right now are guessing at the exact question this game makes a dinner table answer in twenty minutes: what can the machine be trusted to do, and where does a human have to keep the pen? Get that line wrong in one direction and you ship something reckless; get it wrong in the other and you've spent a fortune to automate nothing.
I help teams draw that line on purpose — what to automate, what to keep human, how to ship it without it blowing up, and how to be honest with users about which is which. Same rigor you just watched run a game; pointed at your roadmap.
- AI strategy & the human-in-the-loop line — where models help, where they don't.
- Shipping AI features safely: moderation, guardrails, fallbacks, evaluation.
- Accessibility and trust — products people can actually use and believe.
- Prototype to production by someone who has done the whole distance.
$400 / hour
Straight advisory or hands-on. For research collaborations and pro-bono work — nonprofits, public-interest projects, anything that moves the human-vs-machine question forward — the rate is negotiable to nothing; the research is its own pay. Ask.
Hiring teams and press welcome too — interviews, advisory, or a one-off teardown of where AI fits your roadmap.
Sean McKendry · builder of Only Humans Score. The work above isn't described — you just used it.
Why your firm should pay the consult fee. Yesterday.
Because two clocks are already running against you, and both started before you finished reading this sentence.
Clock one: AI is in your practice whether you sanctioned it or not. Associates are already drafting with it. Courts have already sanctioned lawyers — real ones, by name, in the public record — for filing briefs full of cases the machine invented. That is the exact failure this whole site is built to prevent: knowing where the machine may hold the pen and where a human absolutely must. Get that line wrong once and it's not a bad quarter — it's a bar complaint, a malpractice exposure, and a headline with your firm's name on it. I draw that line for a living.
Clock two: trust is your only real product, and it's the one thing you can't bill back. Law is among the least-trusted professions in America — not because lawyers are the worst people, but because nobody ever taught the firm to speak human. Your jurors are humans. Your clients are humans. The person deciding whether to call you or the firm down the street is a scared human at a kitchen table. I spent a career as an ad-man making the case to exactly that person — and I've sat on the other side of the system myself, pro se, with no firm at all. I know precisely how your profession sounds to the people you need most. That perspective isn't on your masthead. It should be on your retainer.
- AI guardrails for legal work — where models help draft and research, where a human must verify, and the firm-wide policy that keeps you off the sanctions list.
- Tell your firm's story to actual humans — plain-spoken, trustworthy, no Latin where English will do. Juries and clients included.
- The client's-eye audit — what your intake, your bills, and your website feel like to a frightened non-lawyer, from someone who was one.
- One honest outside voice — the jester in the room full of yes, which (see the library) is load-bearing.
Same $400/hour as above — which, against one avoided sanction or one retained client, is a rounding error you'll wish you'd spent sooner. The honest pitch, held to the one rule: I'm not a lawyer, I won't pretend to be, and I'll never invent a credential. I'm the ad-man, the AI hand, and the human who's been on your client's side of the table. Bet on that. No net.
Not legal advice and not a claim to a license. A communications, AI-strategy, and client-experience consult — from the maker of the instrument you just used.