Ask honestly.
Get the same advice I'd give on a call.

The short answer

No. And it's not the number that's wrong — it's the shape of the deal.

"$80k for a 12-week AI transformation pilot" is three red flags in a single sentence:

1 · "Transformation" and "pilot" don't belong together. A pilot is a scoped experiment against a single hypothesis. Transformation is a multi-year organizational change. If a vendor uses both words in one sentence, they're selling the romance of one and the scope of the other.

2 · "Your competitors are ahead" is a sales tactic, not a diagnostic. It's impossible for them to know what your competitors are actually doing. Urgency is the oldest tool in the B2B sales playbook — treat it as information about them, not about you.

3 · No outcome is named. Ask them: "What specific decision or process will be measurably different at week 13?" If they can't answer in one sentence, you're buying activity, not impact.

What a reasonable engagement looks like

For a business your size, a credible first engagement is usually $8k–$20k for 3–4 weeks, outputs a readiness assessment and a single prioritized use case with a business case. Then — then — you scope build.

If this vendor can't restructure their proposal into that shape, they're not the right vendor. That's not negotiation advice — that's a compatibility signal.

What I'd do tomorrow

Send them three questions in writing: (1) What business outcome is this measuring? (2) What happens if the pilot fails — do we keep anything? (3) Who owns the data, the model, and the IP at week 13? Their answers will tell you everything. Most vendors will not respond.

First — you have two teams, not one

"Hype skeptics" and "fear skeptics" sound similar but they need opposite things. Treating them as one group is where most leaders lose the room.

Hype skeptics think AI is overblown. They'll be persuaded by evidence — a small, real result inside your own business. Not a demo, not a case study. Something they watched work.

Fear skeptics think AI is coming for their role or their team. They'll be persuaded by agency — being the people who choose, scope, and evaluate it. The fastest way to make them hostile is to hand them a finished tool.

A sequence that usually works

Week 1 · Ask, don't tell. Individual 20-minute conversations with each skeptic. One question: "Where do you spend time on work that doesn't feel like yours?" You're not selling AI. You're collecting candidates for what AI could remove.

Week 2 · Pick one boring thing. Not the flagship process. Something tedious and low-stakes — meeting notes, invoice categorization, report summaries. Boring protects you from big-bet failure.

Week 3–4 · Let a skeptic lead the pilot. The person most skeptical should run the small experiment. If it works, they become your most credible advocate. If it doesn't, they trust the rigor.

Week 5 · Share honestly. Present what worked, what didn't, what you'd do next. Not a celebration — a field report.

What not to do

Don't run a lunch-and-learn with an external speaker. Don't buy a platform "for the team to try." Don't announce an AI strategy at an all-hands. Every one of those makes skeptics more skeptical, because each one skips the thing they actually want — to be consulted before a decision is made.