AI Washing Is Getting Noticed. Here's the Question Your Board Will Ask About Your AI Efficiency Story
An efficiency story used to be easy to tell: headcount went down, costs went down, and the press release credited new technology. That story is losing its credibility with the people who measure CFOs. Boards, investors, and analysts have learned to separate a cost cut from a change in how the business operates, and they are starting to ask for proof of the second. Finance leaders and P&L owners who can show what happened to the work will have a much easier conversation than those who can only show what happened to the payroll.
What is AI washing, and why is it in the headlines?
AI washing is the practice of attributing job cuts or cost reductions to artificial intelligence when the technology played a small part, or none, in the decision. Fortune described it as a way for companies to "spin what would usually be negative news into a positive feat that shows they are innovating".
Paul Osterman, professor emeritus of human resources management at the MIT Sloan School of Management, put the incentive plainly in the same article: "AI is a perfect excuse to justify big layoffs," he told Fortune, because it "makes it seem as if it's not our decision." He also noted that companies have long used downturns as a convenient cover story for cuts they planned to make anyway. AI is the newest version of an old move, and that history is exactly why sophisticated observers are now skeptical of the narrative.
Have investors started to notice?
Yes, and the market reaction is measurable. The Financial Times reported that companies citing AI as a factor in job cuts underperformed the Nasdaq by almost 10% in the 30 trading days that followed the announcement. An AI framing that was once expected to signal innovation is now just as likely to raise a question about what the company is really doing.
Gartner's research points in the same direction from the operating side. In a survey of 350 global business executives, about 80% of organizations piloting or deploying autonomous business capabilities reported workforce reductions, but reduction rates were nearly equal among companies reporting higher returns from those technologies and companies seeing modest or negative results. Helen Poitevin, Distinguished VP Analyst at Gartner, summarized the finding: "Workforce reductions may create budget room, but they do not create return."
Put together, these findings tell CFOs something important. A reduction in headcount doesn't distinguish a company where AI is working from one where it isn't, and investors are beginning to price in that uncertainty.
What question will the board ask?
The question is simple to state and hard to answer without the right data: "Show us the work changed."
A board approving or reviewing an AI-linked restructuring wants to know whether the company now does the same work, or more of it, with fewer people because AI absorbed real tasks. The alternative is that the company now does less, or does it worse, and the savings will show up later as rework, missed targets, or rehiring. Directors may phrase it in different ways:
- Which tasks did AI take over, and how do we know?
- Where did the time we freed up go?
- Did output and quality hold after the change?
- How would we know early if the remaining team is overloaded?
Each of these is a question about work, not about cost. That's why the usual finance artifacts struggle with them.
Why can't headcount and budget variance answer it?
Headcount and budget variance are reliable records of what happened. They show that payroll fell and that a cost center came in under plan. They can't show why.
A favorable variance after an AI rollout is consistent with several very different realities. AI may have absorbed meaningful work, or the remaining team may be absorbing the same workload through longer hours, or some work may simply have stopped getting done. All three produce the same line in the budget for at least a quarter or two. The difference only becomes visible later, when burnout, attrition, or a missed commitment forces the issue.
This is the gap AI washing exploits, whether deliberately or not. When the only evidence is a lower cost base, any explanation for it is hard to disprove. A finance leader who wants the AI story to survive scrutiny needs a second kind of evidence alongside the financials: evidence about how work is actually getting done.
What evidence actually answers the board?
Credible evidence connects the AI investment to observable changes in the work. Three measures do most of that job.
| Board question | What headcount and variance show | What workforce evidence shows |
|---|---|---|
| Did AI take over real work? | Fewer people on payroll | Changes in time spent on the tasks AI was meant to absorb, measured before and after |
| Where did the freed capacity go? | Nothing | How utilization shifted across the remaining team, and whether time moved to higher-value work |
| Did output hold? | Cost per unit, after a lag | Activity and outcome trends for the affected teams, plus early signs of overload |
Utilization before and after. A baseline of how teams spent their time before the change, compared with the same view afterward, shows whether AI actually reduced the effort required for specific work. Without the baseline, any claim about efficiency is a reconstruction.
Where freed capacity went. If AI saved time, that time went somewhere: to new initiatives, to customers, or nowhere in particular. Boards care about the answer because it determines whether the investment created value or only removed cost. Utilization data shows whether the remaining team has room to spare or is running hot.
Whether output and quality held. Efficiency means the same or better results with less effort. Tracking output trends for the affected teams, alongside signs of overload such as sustained after-hours activity, shows whether the savings are durable or borrowed against the future.
Gartner's guidance is consistent with this approach. Poitevin noted that organizations improving ROI are not those that eliminate the need for people but "those that amplify them by aggressively investing more in skills, roles and operating models". Evidence about the work is what lets a finance leader show the board which of those two paths the company is on.
For a broader view of how to turn workforce data into decisions, see a workforce analytics framework for decision-making.
How do you bring workforce data into the finance conversation?
The practical obstacle for most finance teams is that workforce evidence lives outside the systems they use to build board materials. Closing that gap is mostly a matter of getting the data where finance already works.
Prodoscore captures how work actually gets done across the business tools a company already uses. Its AI adoption measurement shows whether AI tools are changing how teams spend their time, and that data feeds the broader workforce analytics view of capacity and utilization. Prodoscore makes it available in three ways that fit a finance workflow:
- ProdoAI Chat lets finance leaders ask direct questions about utilization, workload, and trends without waiting on a custom report.
- The MCP Connector brings the same workforce data into AI platforms such as Claude, Gemini Pro, or ChatGPT, so it can be analyzed alongside financial and planning data. See how Model Context Protocol connects AI to business data.
- The Prodoscore Data API sends raw activity and workforce data into a company's data warehouse or BI tools, where FP&A teams can join it to cost and revenue data. See the Prodoscore Data API announcement.
Finance leaders using Prodoscore already describe the value in these terms. A PE firm leader quoted on Prodoscore's finance solutions page said: "Prodoscore gave me clear, objective evidence to act on and because the data is completely objective, no one could argue."
That is the standard a board will hold an AI efficiency story to. The companies that meet it will be the ones that measured the work before they changed it, and kept measuring afterward.
The next AI efficiency story your company tells will be judged on evidence about the work. Talk to us about bringing that evidence to your next board conversation.
