Lately, leaders have been increasingly focused on maximizing the use of AI tools so that the team can deliver projects as quickly as possible. Skills and bots are being used, workflows are being automated, tasks are being worked on in parallel, etc. And this is slowly creating a problem that many leaders aren’t even aware of. Team members are losing focus, becoming overwhelmed by the amount of information and pressure, and the team is becoming demotivated. Also, often they start working on tasks just to get them out of the way and meet their required targets.
A typical day for a developer now looks something like this: they are assigned to several PRs with too many files changed, review PRs through their AI agents, spend the entire day sending prompts and checking what the AI has done, work on their own tasks with AI, and mostly just read its answers and ideas. We spend the whole day reading and reading, and somewhere in all that reading, we lose focus on the bigger picture of the project.
I’m noticing that, despite having access to more information than ever, developers and other team members are often missing answers to some fundamental questions: Why are we doing this? How does my work contribute to the bigger picture? And how does my task fit into the existing system? Somehow, in organizations, it has become expected that developers should figure out all of this information on their own - or have Claude Code explain it to them.
- Make sure every team member understands their task and why it matters.
- Make sure everyone understands the main idea behind the project, how the system works, and how each person contributes to the project’s success.
- Make sure documentation and tasks are clear and easy to understand, rather than being just AI-generated text.
- Have regular check-ins with team members to talk about their satisfaction, motivation, and overall well-being.
- Take a human approach with team members and show them that they are the most important part of the team.
- Support team members in their growth and actively create opportunities for them to learn and improve.
Giving the team a clear picture of the project has always been part of a leader's job. A leader needs to take time to explain what the team needs to do. The best way to do this is by actually talking to people, because a long Slack message or a ticket full of AI-generated text won't do the job. Sit with the team, draw the flow, and answer their questions. It may feel slower at first, but it saves time later, when developers are less likely to go in the wrong direction or accept AI suggestions that don't fit the project.
For this to work, the leader has to be the person who understands the project completely. They need to understand the whole flow and where the project is going. If the leader also only sees pieces of the project, or only looks at dashboards and AI summaries, there's nobody left on the team who understands how everything works together. When everyone else is lost in the details, the leader should be the one who still sees the whole project.
Leaders should also pay attention to what the team reads. Documentation and tickets are the basis of everyone's work. If they're full of AI slop, with long sentences and fancy words that sound smart but don't explain anything, people will either skip them or misunderstand them. A leader should make sure that docs and tickets are written in simple language that everyone on the team can understand. Even if AI helped write them, someone should read them, clean them up, and make sure they say something useful. A short ticket that everyone understands is worth much more than a long one that nobody reads.
Leaders also need empathy for developers who are struggling with AI adoption. Some developers feel that their skills are becoming less valuable, and many are starting to dislike their jobs because they spend more time reviewing AI output than building things themselves. Leaders need to listen, support them, and help them find a way to work with AI while still feeling that they are in control of their work.
I think the main point is that team members need to feel like they're creating something bigger than just their individual tasks. When developers understand the project and see how their work contributes to it, they start caring about the result. That's when AI becomes a tool that helps them do better work. Without that understanding, AI becomes a tool that does the task for them, and the developer just accepts whatever AI writes and moves on to the next ticket. Quality drops and nobody really owns the result.
A leader can't stop AI from producing too much information. But a leader can help the team understand what matters. A team that understands what it's building will always use AI better than a team that's just waiting for AI to finish the job. A motivated team can build a good project with AI as a tool. With an unmotivated team, AI will end up being the one building the project.