There is a version of AI adoption that looks like success from the outside — tools deployed, training delivered, adoption rates tracked — but produces quietly resistant employees, low utilisation, and a growing gap between what leadership believes is happening and what teams are actually experiencing. Mindy Honcoop has spent the last several years watching this pattern repeat, and at Running Remote 2026 she laid out what it looks like from the inside — and what strategic HR leaders do differently.
Honcoop, founder of Agile in HR and VP of Customer Success at meevibot, describes herself as a recovering chief people officer. After stepping back from the CPO seat at the end of 2021, she spent years working with small and medium-sized businesses across advisory and technology roles, interviewing organisations going through AI implementation at scale. What she found was a consistent pattern: organisations were approaching readiness from only one dimension — process and technology — while the people dimension was being systematically left out.
The readiness gap is not what most organisations think it is
When HR leaders are told to ‘do AI’ or ‘do more AI,’ most have no clear understanding of what strategic business outcome is being served, what ownership looks like at the people level, or how to measure whether readiness is actually being built. Boards pressure executives to replace headcount with automation. Executives pressure HR to deploy tools. HR tries to follow, often without being included in the original conversations that created the mandate.
The result is what Honcoop calls emotional chaos. And one of its more counter-intuitive findings: early AI adopters are often among the most fearful about their job security, precisely because they understand more clearly what the technology can do.
Resistance is a signal, not a problem
The single most important reframe in the session: resistance is not a performance issue. It is data. It is feedback worth getting curious about.
Most organisations respond to resistant employees by labelling them, managing them out, or simply moving around them without ever understanding what the behaviour is communicating. Honcoop argued this is both a waste of talent and a missed diagnostic opportunity. The employees who are resisting know something. The question is what.
She identified four emotional states that drive resistant behaviour, and each requires a different response.
Doubt — uncertainty about what is expected, what is permitted, what success looks like — needs clarity. Not certainty (leaders rarely have that), but transparency about what is known and what is not. One of the most paralysing examples she shared: an organisation that deployed Claude with a business licence and sent employees a disclaimer so filled with restrictions that HR could not determine what they were actually allowed to use it for. Paralyzing doubt, created not by the technology but by governance failures.
Overwhelm needs focus. The solution is not less ambition but more structure — doing AI work together as teams rather than in fragmented individual experiments, connecting implementation to team goals rather than floating abstractions.
Isolation — the feeling of being the only person struggling — needs connection. Honcoop was direct about the mechanism: shame is healed in connection, but shame makes people avoid connection. This is the catch-22that HR leaders need to interrupt with deliberate, low-stakes opportunities for people to share what is hard without social risk.
Fear — about job security, about being the only one who cannot keep up — needs psychological safety built through norms that normalise mistakes. Honcoop noted the particular absurdity of a workplace that demands human perfection while deploying AI tools that openly acknowledge they make mistakes. Modelling that experimentation is expected and failure is a data point changes the culture faster than any policy.
The core principle: slow down to speed up
The session’s central argument was that organisations trying to sprint through AI adoption are creating the resistance they are trying to overcome. Slowing down enough to ask good questions, listen to answers, and understand what is actually happening beneath the surface is not a delay to implementation. It is the implementation.
This is not about pausing. It is about the quality of attention. HR leaders who develop the skill of asking powerful questions — who can sit with an employee’s doubt or fear without rushing to fix it or dismiss it — are the ones who build organisations where AI actually gets used.
The practical starting point: stakeholder management, not change management decks. Understand the pain points of each executive one-on-one before trying to sell the ‘slow down’ message in a group setting. Find the one person in the room who is already open, and work from there. Build momentum before trying to move the whole system.
Honcoop’s closing point was worth staying with: organisations have, in many cases, lost the basic capability for caring conversations. The pace of implementation has outrun the relational infrastructure required to sustain it. Rebuilding that capacity — the ability to sit with someone who is struggling and genuinely understand what they need — is one of the highest-leverage investments any HR leader can make right now.
