Why AI adoption isn’t just a tech problem, but a retention risk

Worker sentiment in tech is starting to present a shoemaker’s children effect.

Employees are developing world-class technologies and AI tools, while internal processes remain outdated and inefficient.

Stephanie Davis Neill

employee experience feedback from the ten biggest U.S. tech employers, staff are highly motivated to work on AI developments and feel positive about the opportunities to do so, but they’re losing patience with the processes they still have to use at their own desks.

One worker, a full-stack developer, claims, “Since the AI boom, everything has gone downhill”. Another employee at the same company told his employer to “eat your own medicine – the AI we sell vs. the AI we use internally – deep divide”.

Other reviews suggest that those who do not believe in AI or engage with it face negative consequences to their career, while there is also a general consensus that pace and work quality is changing with the introduction of AI processes.

One developer states that deliverables have felt rushed since AI was introduced, with quality of work decreasing, while another states that their leadership’s use of generative AI in emails and communications is deemed impersonal and unprofessional.

Another reviewer even claims, “If it was up to this company, they would sack everyone and use AI”.

social media, we can see an increasing backlash to workplace AI mandates across multiple industries on platforms such as Reddit and X (formally Twitter). In fact, many of the conversations happening here also explicitly cite AI mandates as a reason for quitting their job altogether.

Looking at these posts, there’s a clear desire to be included in AI-related discussions with leadership, which if implemented, might help reduce the feeling among many workers that AI use is being “forced”.

There’s also a frustration that poor AI performance is blamed on “bad prompts” from the employee rather than on the tool or process itself. This is a common theme, with other people feeling that management has too high expectations of AI, requesting it be used to replace job responsibilities it is not yet capable of completing to a good standard.

Interestingly, in a thread about coping with mandatory AI at work, multiple people even admitted to exaggerating or fabricating their AI usage at work due to a lack of confidence in the technology. One commenter said that they feel unable to convince their managers that they’re better off without it. Instead, they pretend to use it and even invent time-saving figures to meet usage requirements.

employee reality evidenced in this research. A recent study also found that only 4% of leaders cited AI resistance as an issue, but another survey discovered 22% of employees are actually frustrated enough by workplace AI use to consider quitting.

Search data also shows a 10% YoY increase in U.S. searches for “quitting my job” (7,000 monthly searches) alongside emerging queries such as “made to use AI at work” (1,000 monthly searches).

The disconnect is elevated by the fact that people have admitted to misrepresenting their use of AI to meet workplace expectations, even though they are completing the work themselves.

AI is “not just a switch to turn on”

Though AI adds complexity that employers must navigate carefully, this type of dissatisfaction and disengagement isn’t really exclusive to AI; it can actually arise from change management challenges. Just like AI must learn, so do the employees working with it. It is a process, not just a switch to turn on.

Difficulties implementing AI processes can create additional change management hurdles. In a recent survey, senior IT professionals claimed that AI projects are being dropped before production, a YoY jump from 17% to 42% reporting this. In the same survey, 46% of those investing in generative AI state that no single enterprise objective had seen a “strong positive impact” from the investment.

collaboration are key parts of the change management process. AI adoption can be higher when employees get to test or play with AI tools to figure out the best ways they can work with AI to improve their own personal working experience.

A starting point is updating compliance-driven policies to include AI guidelines, sharing key AI process information early in onboarding and introducing regular feedback tools to proactively address concerns and keep employees informed and engaged.

Internal feedback mechanisms, especially anonymous ones, often provide a place for disengaged employees to communicate some of the frustration that can build up, especially when regular conversations are not happening with a direct leader.

By focusing on engagement, managers and HR teams can help employees feel empowered rather than sidelined, keeping talent intact as how they work continues to evolve. Before you can sell the AI revolution to the world, you have to make your own people want to be part of it.