There’s a growing narrative in boardrooms and headlines alike that AI is driving layoffs.

Generative AI can now draft reports, write code, summarize research, analyze contracts, and handle customer interactions. The productivity gains are real, but the scale and timing of many recent “AI layoffs” raises important questions.

While the data confirms that AI is eliminating certain tasks, there is increasing skepticism around the true motivation behind many announced “AI layoffs.” A significant portion of the current displacement may actually be due to slower sales, shifting priorities, and previous overhiring.

A recent Harvard Business Review survey found that many companies cutting staff in the name of AI are acting in anticipation of AI’s potential rather than its current performance, meaning they are making workforce reductions before the technology has fully reshaped how work actually gets done. Forrester predicts that more than half of layoffs attributed to AI will be reversed as companies discover the operational challenges of replacing human workers too quickly.

Evidence of this dynamic is already emerging. A recent survey of HR leaders found that nine in ten companies would rethink AI layoffs if they were given the chance, with 32.7% of companies already rehiring between 25% and 50% of the roles they initially eliminated.

These data points suggest that many AI-linked layoffs do not tell the full story about how AI is restructuring work.

AI Requires a Different Way of Working

AI reduces the cost of generating content, analysis, and other deliverables. However, it doesn’t eliminate the need for judgment, synthesis, and governance. In many cases, it increases it. As the volume of AI-generated code, reports, emails, and analysis explodes, the human bottleneck shifts. It moves from creation to verification.

Organizations need to think through which tasks are truly automatable, which require human judgment layered on top of AI outputs, and where new bottlenecks will emerge.

Most productivity gains from previous technologies came from redesigning entire workflows. We’ve seen this pattern before, and AI is no different. Entire fields like cybersecurity emerged as digital systems became more complex. Substituting a model for a task without rethinking the surrounding process often creates new inefficiencies rather than eliminating old ones. As NVIDIA CEO Jensen Huang recently noted, companies with imagination can use AI as an opportunity to do more.

Early efficiency gains are easy to measure. Systemic resilience and creative expansion are not. Organizations that cut roles before imagining a new way of working may realize months later that they removed critical roles or failed to anticipate new ones.

The Risk of Losing Institutional Knowledge

While AI promises significant long-term productivity gains, reducing headcount before organizations have redesigned how work actually gets done with it introduces real structural risks. One of the most immediate hidden costs of rushed AI-driven layoffs is the loss of essential expertise and institutional knowledge. Beyond the direct loss of capabilities, these decisions can increase burnout and reduce efficiency among the remaining workforce.

As one expert noted in a Tegus expert transcript, aggressive corporate cutbacks often eliminate the exact employees who possess the end-to-end business understanding required to integrate AI successfully: “When you're trying to integrate something like AI, the new folks, they don't know enough about all end to end how everything works, how it impacts? There's this gap in knowledge.”

Optimizing for the Wrong Metrics

There’s another reason AI layoffs may be a misstep right now: We are measuring the wrong thing.

In many organizations, AI is being evaluated primarily through an efficiency lens. What is the cost per task? How much time is saved? How many roles can be reduced?

Those metrics are clean. They’re quantifiable. But in focusing so heavily on efficiency, we risk missing the bigger gains of AI. Imaginative leaders look beyond hours saved. AI’s most transformative potential isn’t only doing the same work cheaper; it’s enabling work that wasn’t feasible before. It’s improving decision-making.

If leaders fixate on efficiency metrics alone, the most obvious lever becomes headcount reduction. But that narrows AI’s role. And while strategic gains are harder to quantify, they’re more powerful. And those are the metrics that will matter most in the AI era.

Once we know how to measure AI’s true value, we can then have a serious conversation about workforce design. Until then, cutting roles risks optimizing for what’s easiest to measure rather than what matters most.

Redesigning Work for the AI Era

There is no doubt AI will reshape many roles in the workforce. But the most durable organizations will likely follow a different approach to workforce decisions — redesigning roles around oversight, judgment, and AI integration, while preparing for new responsibilities that emerge alongside those that fade.

Cutting before this work is done may create short-term financial gains, but it risks long-term structural weakness. You don’t cut your way into an enduring and effective AI strategy.

Discover how you can transform your research process with AlphaSense’s Generative Search. Start your free 2-week trial of AlphaSense today.