We’re seeing a new and subtle risk to AI adoption and productivity: AI arrogance.
It’s the confidence some users develop, often those championing new tools, where they assume AI can solve every problem. They believe their preferred platform, and the way they use it, is the answer to all challenges.
This mindset is very risky. It leads to overestimating accuracy, underestimating complexity, and sometimes missing the real needs of your business. We see it when people insist Copilot or Claude (or any tool, including ours) can do everything, ignoring the nuances and limitations.
The reality? AI is powerful, but not omnipotent. Success depends on clear-eyed evaluation, honest feedback, and recognising where human expertise matters. Lowering the barrier for entry is important, but so is tackling the “AI arrogance” that can blind us to better solutions, or to the value of simpler, more targeted tools.
Leaders have a critical role to play here. Evangelists are vital for building momentum, but they must be evangelising the right skills and approaches, not just promoting a single tool or platform. It’s essential for leaders to spot and stop evangelists who, however well-intentioned, risk holding others back by narrowing the conversation or stifling experimentation.
As we build and deploy AI systems, let’s stay humble, keep listening, and remember: championing a tool is great, but championing outcomes is better.
When it comes to AI, humility isn’t just a virtue, it’s a competitive advantage. That’s because AI is a powerful ally, but a poor dictator. The evangelists you need champion outcomes, not tools.
