Companies Rolled Out AI With a Layoff Threat Attached. Now 29% of Employees Are Sabotaging It.

A 2,400-person survey by WRITER and Workplace Intelligence found that 29% of employees admit to actively sabotaging their company's AI strategy.

Not passively ignoring it. Not dragging their feet. Sabotaging it. Entering proprietary company data into unauthorized public AI tools. Using unapproved third-party applications. Deliberately generating low-quality output to make AI look less effective. Refusing outright to use mandated tools.

Among Gen Z workers, the number is 44%.

These are not hypothetical risks flagged by a security vendor trying to sell you something. These are employees, at companies that have already deployed AI, telling a survey they are working against their own company's technology strategy. On purpose.

Most of the conversation around AI adoption focuses on tooling, pricing, and productivity metrics. Almost nobody is talking about this. And the reason nobody is talking about it is that nobody wants to frame AI resistance as a rational response to a real threat. It is easier to call it a change management problem. It is easier to blame training gaps.

It is not a training gap. It is a trust gap. And the trust gap is about to become a brand problem. We have written repeatedly about how AI deployment decisions that look like operational wins become brand liabilities when the human cost surfaces. This is the most vivid example yet.

What the Sabotage Actually Looks Like

The WRITER data breaks the sabotage into specific behaviors and none of them are subtle.

Employees are feeding company data into consumer AI tools the organization does not control. Not because they are careless. Because the approved tools are slower, more restricted, or do not exist yet. Mimecast's State of Human Risk 2026 report found that 80% of organizations worry about sensitive data leaking through generative AI. 60% still have no specific strategy to address it.

The scale of unauthorized usage tells the rest of the story. Up to 65% of employees are bypassing IT entirely to use unsanctioned AI tools. The average enterprise experiences 223 data policy violations per month related to AI usage. IBM's research found that shadow AI was a factor in one out of five data breaches, adding an average of $670,000 per incident to breach costs.

But the WRITER data shows something beyond shadow AI. 29% are not just using unauthorized tools. They are deliberately undermining the rollout. Generating bad output to make the tools look broken. Feeding garbage in to get garbage out so they can point at the results and say "see, it doesn't work."

That is not a skills problem. That is an employee base that has decided the AI strategy is a threat and is acting accordingly.

Why They Are Doing It

The same WRITER survey tells you why. 60% of executives plan to lay off employees who cannot or will not use AI. 77% say non-adopters will be excluded from promotions and leadership roles. 92% of the C-suite admit they are actively cultivating a class of "AI elite" employees while the rest of the workforce watches.

Employees are not stupid. They see the two-tier system forming. They hear the language about "AI-proficient" workers getting promoted 3x faster. They understand that when the CEO talks about "leaner teams powered by AI," what the CEO means is fewer of them.

A Kyndryl survey of over 1,000 senior business and technology executives found that 45% of CEOs say most of their employees are resistant or openly hostile to AI. Only 14% have aligned their workforce, technology, and growth goals. 95% have invested in AI. The money went in. The people did not follow.

The resistance is rational. When a company tells employees "adopt this tool or lose your job" while simultaneously telling investors "this tool will let us operate with fewer people," every employee in the building can see both messages. The sabotage is not confusion about the technology. It is people protecting their jobs the only way they can. And when 44% of your youngest workers, the ones you need most for the next decade, are the ones fighting hardest, you have a generational trust problem that no all-hands meeting is going to fix.

The Part Nobody Wants to Call a Brand Problem

We cover brand for a living. And we are telling you: this is a brand story.

Glassdoor's Worklife Trends 2026 report identified a growing "employee-leader disconnect" as one of the defining dynamics of the year. Their 2025 word of the year was "fatigue," reflecting exhaustion with layoffs, AI disruption, and top-down mandates. Career opportunity ratings on Glassdoor fell from 4.1 in 2020 to 3.5 in 2025.

That sentiment is not staying inside the building. Your employer brand is being shaped right now in Glassdoor reviews you have not read, in Reddit threads you do not know exist, and in the summary a candidate just received from ChatGPT when they typed your company name into an AI search bar. 75% of job seekers research a company's employer brand before applying. A single viral Glassdoor review can suppress application rates for months.

When 29% of your workforce is at war with your AI strategy, that war leaks. It leaks in how your customer support team sounds. It leaks in the quality of the work your team ships. It leaks in the Glassdoor reviews your candidates read before deciding whether to take your call. It leaks in how your employees talk about the company at dinner, at conferences, on LinkedIn. We wrote about Klarna learning this the hard way when it cut 700 jobs for AI and then hired them back. The internal damage was already done.

Startup Fortune reported on how Meta's AI mandates are breaking employee trust internally, with workers feeling squeezed between aggressive AI adoption targets and layoff anxiety. Every AI tool starts to carry a second meaning. A coding assistant is not just a coding assistant. It becomes a signal that fewer engineers may be needed. A tracking dashboard is not just a data source. It becomes a reminder that management is studying workflows for replacement potential.

Your employees know what the tools are for. And when the tools feel like surveillance, the response is sabotage. When the sabotage becomes culture, the culture becomes your brand.

The Security Problem Is the Brand Problem

The security numbers put a price tag on the brand risk.

IBM's global study of 600 organizations found that only 37% had detection or governance policies in place for shadow AI. Deloitte's 2026 State of AI in the Enterprise found that worker access to AI tools rose 50% in 2025, but only one in five companies has a mature governance model for how those tools are actually used.

If your employees are feeding customer data into unauthorized tools, you have more than a security incident waiting to happen. You have a trust incident. A breach that traces back to an employee using ChatGPT to process customer records because the company's approved tool was too slow becomes a PR crisis. A brand story. The kind of headline that follows a company for years.

And the breach data says it is coming. One in five organizations has already experienced a breach involving shadow AI. The average cost premium is $670,000. The reputational cost does not show up in the IBM report but every brand leader reading this knows it exists and knows it compounds.

The Fix Is Not a Webinar

Stop treating AI resistance as a change management problem. It is a trust problem. The fix looks completely different.

If 29% of employees are sabotaging AI, the question is not "how do we train them better." The question is "what did we do to make them feel this threatened." And the answer, for most companies, is that they announced an AI strategy in the same breath as a layoff plan and expected loyalty in return.

Provide tools that actually work. Healthcare organizations that provided approved AI alternatives saw 89% reductions in unauthorized use. Samsung reversed its ChatGPT ban and gave employees sanctioned tools instead. The pattern is consistent across industries: give people something good and they will use it. Block everything and they will go around you.

Be honest about what AI means for their careers. If roles are going to change, say so with specificity and a timeline. If roles are going to be cut, do not pretend the AI tool is there to "empower" them. Employees can read a balance sheet. Treating them like they cannot is what creates the adversarial dynamic in the first place.

And measure what matters. Only 29% of executives report meaningful ROI from generative AI despite near-universal deployment. If you cannot prove the tools are delivering value, you do not have standing to demand adoption. You are asking employees to risk their careers for a strategy that your own C-suite admits is not working yet.

The companies that get this right will not be the ones with the best AI tools. They will be the ones whose employees actually want to use them. Culture drives that. Leadership drives that. And when both fail, brand absorbs the damage.

29% of your workforce just told you what they think. You can call it a training problem. You can hire a change management consultant. Or you can listen to what your own people are saying and fix the thing they are actually reacting to.

Brand is not just what you say to the market. It is what your people say about you when you are not in the room. Right now, 29% of them are saying it with sabotage.