AI is genuinely reshaping how companies operate, and some of this year's hardest calls are about it. That is what makes the dishonest uses so costly. A growing number of leaders are using AI not to make a decision or explain one, but to avoid owning either. It happens at two moments. Before a decision is made, AI can make the call so no one has to answer for it. After a decision, AI can take the blame so no one has to defend it. Both are choices, and both come back on the organization.
After the decision: AI as the reason that tests well
Framing a restructuring as an AI transformation signals discipline and forward motion. It tells investors the company is building toward the future rather than cutting to survive. When AI actually drove the decision, that is a fair account, and the workforce recognizes it as true because it matches what they saw.
The problem is when AI was not really the driver. The same words that reassure investors reach the workforce, and inside the building they read differently. The employees who watched the work know why the cut actually happened. When they are told it was AI, and they can see it was something else, they do not hear a strategy. They hear a leadership team reaching for a machine's name rather than owning the real decision.
Even the industry is calling this out. Jensen Huang, whose company Nvidia sits at the center of the AI economy, said it does not make business sense to claim AI is replacing workers at the scale executives keep implying. He called the leaders who lean on that explanation lazy. When the person with the most authority to confirm the AI story is the one casting doubt on it, some of that story is doing something other than informing.
The stated reason is not a fact leadership reports at the end. It is a choice, and a workforce remembers the reason longer than the announcement. When the reason is not true, they know, and they carry that forward.
Before the decision: AI as the one who makes the call
The same avoidance shows up one step earlier, in decisions leaders hand to AI so no person has to own the outcome. Hiring is the clearest case. An AI screener decides who advances and who gets rejected, and it lets leadership treat those rejections as the neutral output of a tool rather than a set of choices the company made.
They are still the company's choices. A federal court ruled this year that an AI hiring vendor must face claims its software screened out applicants based on age, race, and disability, and it ordered the vendor to identify every employer that enabled those features. The tool did not absorb the responsibility. It concentrated it, because one biased system shapes decisions across every company that used it.
The pattern matches the layoff story exactly. In both, a leader puts AI between themselves and a decision they would rather not answer for. The difference is only where the avoidance sits and how the bill arrives. Blaming AI for a cut is a communications failure the workforce feels as broken trust. Handing hiring to a biased screener is a decision the company answers for as legal exposure.
What both cost
When the stated reason does not match what the workforce experienced, people do not assume leadership simplified for a broad audience. They conclude leadership was unwilling to tell them the truth. That conclusion attaches to the next announcement, and the one after it. Trust does not break loudly in that moment. It erodes quietly, and it is expensive to rebuild.
The hiring cost is more concrete and just as delayed. It shows up as complaints, as discovery, as a company name in a story about AI discrimination, long after the tool was quietly switched on. In both cases, leaders who reach for AI to avoid a hard moment pay for it later, at a higher price than honesty would have cost.
Owning the decision instead
Owning a decision does not mean disclosing everything or sharing what is genuinely confidential. It means a person stands behind the call and the reason given for it. When AI is the real driver of a restructuring, the honest move is to say so plainly and prepare the workforce for what it means. When AI screens candidates, the honest move is to treat every rejection as the company's decision and to audit the tool as if the company made each call itself, because it did.
Before leadership settles on the language of an announcement, or switches on a tool that will decide who gets hired, someone should ask the question that matters most. In twelve months, when the workforce and the record show what actually happened, will we be able to stand behind what we said and did?
The technology is not the problem. Using it to avoid owning a decision is. That choice costs more than the ownership it was meant to escape, and it costs it later, when it is hardest to repair.
I work with leadership teams on the decisions and the reasoning behind an announcement, before the language is set and before the call is made. If your organization is facing a difficult decision and wants to own it well, reach out at daphnescott.com.