Stopping AI projects

The case for stopping AI projects

Everyone is measured on launching AI, almost no one on stopping it. That asymmetry is why companies keep funding bets that will never pay off.

Talk through your portfolio → Read the short answer →

4,000+ experiments · 50+ teams · Pretotyping before build

Decision system

01 Score Value · evidence · adoption · risk · time
02 Test Pretotype the assumptions that could change the call
03 Decide Fund · fix · kill

One comparable view of every AI bet, so the next dollar follows evidence rather than enthusiasm.

The short answer

Stopping an underperforming AI project is often the highest-return decision a board can make. Every dollar pulled from a bet that won’t pay off is a dollar freed for one that will. Disciplined stopping, based on evidence and pre-set thresholds, isn’t a failure of AI strategy; it’s the strategy working as intended.

Key facts

  • MIT’s NANDA initiative found 95% of enterprise generative-AI pilots produce no measurable P&L impact.
  • Money left in a losing bet is money unavailable for a winning one. The real cost is opportunity cost.
  • Sunk-cost bias and reputational risk keep dead projects alive far past their evidence.
  • Pre-committed stop thresholds turn stopping from a political fight into a routine decision.
01

Why companies can’t stop

Stopping is hard for human reasons, not technical ones. Careers are attached to launches. Budgets already spent feel like they must be justified. Nobody wants to be the executive who stopped the AI project that a competitor later made work. So dead bets keep breathing, absorbing capital and attention.

The cost is invisible because it is an opportunity cost. A bet that limps along at break-even doesn’t show up as a loss on any report, but the high-value bet it starved never gets funded. Gartner expects over 40% of agentic AI projects to be cancelled by the end of 2027, and the sooner those calls are made, the more capital survives.

02

Stopping is a portfolio decision, not a verdict on people

The reframe that creates discipline is treating each initiative as a bet in a portfolio. In a portfolio, stopping a bet says nothing about the people who proposed it. It says the evidence didn’t clear the bar this round, and the capital is better deployed elsewhere.

That only works if the bar is set in advance. When you agree the fund, fix, or kill thresholds before you start, a stop decision becomes the process doing its job, not an argument you have to win.

03

How to make stopping routine

Companies that stop well share a few habits:

  • Set explicit success and stop criteria for each bet before funding it.
  • Require behavioural evidence, not status updates, at each stage gate.
  • Separate the decision to stop a bet from any judgment of the team.
  • Celebrate cheap, fast stops as saved capital, the same way you celebrate launches.
04

Evidence makes the stop decision easier

Teams can stop projects without drama when the decision is grounded in data. Pretotyping produces a clear behavioural signal early: people used it or they didn’t, they paid or they didn’t. That removes the ambiguity that lets weak bets survive.

A Rapid Experimentation Operating Model makes the decision repeatable: every bet has a threshold, an evidence trail, and a clear call on what to fund, fix, or kill.

Sources

Evidence behind the page

95% of enterprise generative-AI pilots produce no measurable P&L impact. MIT NANDA, State of AI in Business 2025 ↗ Over 40% of agentic AI projects will be cancelled by the end of 2027, due to escalating costs and unclear value. Gartner, 2025 ↗ More than 80% of AI projects fail, roughly twice the failure rate of non-AI IT projects. RAND Corporation, 2024 ↗
Questions

Frequently asked

When should you stop an AI project? +

When the evidence shows it won’t clear the value bar you set, or when the capital would create more value in another bet. Pre-agreed thresholds and behavioural evidence make that call defensible rather than political.

Is stopping a project a failure? +

No. In a portfolio, stopping a weak bet is the process working. The failure is leaving capital in a bet that won’t pay off while a higher-value bet goes unfunded.

How do you stop projects without hurting morale? +

Set stop criteria up front, base the decision on evidence rather than opinion, separate the decision from the team, and treat a fast, cheap stop as saved capital worth celebrating.

Bring us the portfolio. We’ll help you decide what to fund, fix, or kill.

We look at the bets competing for budget, the evidence behind them, and where a Rapid Experimentation Operating Model would improve the next funding decision.

Talk through your portfolio → See AI experimentation