Generative AI tools, like Claude, ChatGPT, and Gemini, are a tremendous help for team members with well-defined tasks, such as drafting emails, summarizing business strategy, or generating marketing ads.
When team members use AI to augment their work, both high and low performers benefit from the ideas, strategies, and suggestions the tools produce. That’s why most leaders assume that Generative AI will
help make low performers more productive, effective, and strategic.
But recent studies have shown this is not always the case.
In contexts where problems are broad, complex, or fuzzy, AI doesn’t replace human judgment — it amplifies it. The value of what AI tools produce is highly dependent on the human judgment that guides their use and application.
In more open-ended contexts, AI performance relies on the team member to give the tool context, ask good questions, interpret suggestions, and choose which suggestions to implement.
Managing a team, running a business, overseeing a project, or designing a new product or service are not narrow or well-defined tasks.
For users with strong judgment, AI tools surface new ideas and think through risks and trade-offs. Those with strong judgment can use AI to make their thinking better.
But the same is not true for those with weak decision skills. They treat AI as if it has the answers and can unwittingly follow advice that sounds reasonable but ends up leading them to poor outcomes.
In one study, AI boosted revenues and profits by 15 percent among business owners who were already successful. But among struggling owners, AI advice led to a nearly 10 percent decline in revenue and profits.
In multiple studies, low or struggling performers acted disproportionately on the generic advice created by AI tools. To their detriment.
These low performers act on the AI guidance without thinking critically about it. The one-size-fits-all advice often replaces their thinking and reasoning, leading to weaker performance.
High performers, in contrast, are often skeptical of what AI recommends. They use it to discover and implement changes that fit their situation — but they don’t rely on it as heavily to decide which problems to tackle, which strategy to pursue, or which advice to act on.
These findings pose a dilemma for leaders who want to encourage the use of Generative AI tools throughout the organization. AI deployments actually risk widening performance gaps, because the people who need it the most are usually the least equipped to filter and apply the advice it offers.
As leaders continue to introduce, incorporate, and scale AI into the fabric of their teams and organizations, they must remain mindful of the need to train performers not to allow the tools to replace their judgment and thinking.
Showing team members what tasks are best for AI and when to be skeptical of its advice is a step toward more effectiveness throughout the organization.
