Our Favorite Management Tips on Leading with AI

Each weekday, in our Management Tip of the Day newsletter, HBR offers tips to help you better manage your team—and yourself. Here is a curated selection of our favorite Management Tips on leading with AI. Manage AI-Induced “Brain Fry” on Your Team AI can accelerate productivity across your organization—but if you’re not careful, it can also lead to what researchers call “brain fry.” As employees juggle multiple AI tools and oversee streams of outputs, their attention becomes overloaded. As a leader, your challenge is to design how AI fits into work so it boosts performance without exhausting your team. Here are five practices to guide you. Redesign work for human and AI collaboration. Don’t simply layer AI oversight onto existing responsibilities. Define clear limits on how many tools or agents employees manage at once. When AI is embedded thoughtfully into team workflows and treated as a shared capability, cognitive strain drops and productivity improves. Set clear expectations about workload. If you celebrate productivity gains without explaining how work will change, employees may assume expectations are increasing. Clarify AI’s purpose, explain how roles will evolve, and define what oversight and output should look like. Measure impact, not activity. Reward outcomes instead of volume of AI use. When you incentivize quantity, employees produce more work with lower quality and greater mental strain. Focus on measurable business results. Build skills for managing AI work. Help employees strengthen skills such as problem framing, planning, and prioritization. These abilities prevent endless AI iteration and keep people focused on solving the right problems. Protect human attention. Treat attention as a limited resource. Monitor cognitive load and create norms that support focus, good judgment, and sustainable performance. This tip is adapted from “ When Using AI Leads to ‘Brain Fry’ ,” by Julie Bedard et al. . . . Lead Your Team Through AI Anxiety Many employees are worried about how AI will affect their careers. While you can’t offer certainty to mitigate their anxiety, you can give your team honesty, empathy, and clarity. Here’s how to lead anxious employees through the unknown. Name the fear. Start by normalizing anxiety. Share your own uncertainty and model courage by taking action anyway. Naming fears out loud helps your team feel seen and builds trust, especially when the path ahead is unclear. Create space for emotional conversations. Build psychological safety by setting aside time to talk about AI’s impact. Use a “container”—a dedicated space with clear intentions and ground rules—to hold honest, judgment-free conversations. Acknowledge the emotional weight of these changes and make room for multiple viewpoints. Commit to action. Don’t just talk, act. Offer next steps, skill-building opportunities, or shared learning sessions. If job risks are real, say so early and give people time to prepare. Clarity and proactivity reinforce your support. Keep the dialogue going. This isn’t a one-time fix. Make space regularly for honest reflection and discussion. Continuous, transparent communication strengthens resilience and shared purpose. This tip is adapted from “ Your Team Is Anxious About AI. Here’s How to Talk to Them About It. ” by Morra Aarons-Mele. . . . Stop “Workslop” on Your Team Implementing AI mandates without also providing support leads to a flood of low-quality output that erodes trust, wastes time, and degrades team morale. As a leader, you can’t eliminate this “workslop” with stricter reviews or harsher feedback; you need to take a systematic approach instead. Here’s how. Rebuild a culture of trust. Make it safe to admit when AI isn’t helping. Teams that talk openly about quality, ask questions, and give feedback without stigma generate better work. Trust enables people to engage with AI thoughtfully, not just compliantly. Set clear norms for AI use. Without structure, people default to using AI performatively. Define what “good” looks like, establish when AI should and shouldn’t be used, and embed review processes that reinforce—not replace—human judgment. Create accountable support roles. You need people fluent in both AI and collaboration to bridge strategy and execution. Appoint AI collaboration “architects” who can map friction points, align tools with workflows, and ensure output stays tied to business goals. This tip is adapted from “ Why People Create AI ‘Workslop’—and How to Stop It ,” by Kate Niederhoffer et al. . . . Keep Your AI Efforts from Stalling Out If your internal AI rollout looks successful on paper (licenses are being activated, usage is climbing) but you’re still not seeing meaningful impact, the issue often isn’t access, training, or mandates; it’s unaddressed employee fears. If you want real ROI, you need to manage risk perception, not just technology deployment. Here’s how. Recognize industry-shaped risk before deploying AI. Your industry informs how your AI efforts will land before you ever introduce a tool to employees. In some environments, it signals growth. In others, it signals threat. Start by identifying how your people interpret AI: Does it feel like an opportunity, or their replacement? Build your strategy around those perceptions. If you ignore them, you’ll misread both enthusiasm and resistance. Stop treating usage as a proxy for buy-in. High usage doesn’t guarantee commitment. Employees who feel anxious may use AI more, but from a place of self-protection. Pair adoption metrics with signals of psychological safety, openness, and experimentation. Measure whether people feel safe—not just whether they log in. Design for learning before designing for scale. If employees feel personally at risk, they’ll engage cautiously. Create safe environments for experimentation before pushing for enterprise-wide expansion. When people feel secure enough to learn, adoption becomes durable—and results follow. This tip is adapted from “ Why AI Adoption Stalls, According to Industry Data ,” by Erin Eatough et al. . . . Are You Ready to Manage AI Agents? Managing AI agents requires a new kind of leadership. If you oversee a hybrid human-AI workforce, your role is to orchestrate performance, safety, and alignment with business goals. To do it well, you need six critical capabilities. Build AI operational literacy. Understand how agents function, how prompts shape outcomes, and how to diagnose failures. Develop deep functional expertise. Know the business process your agent supports. You can’t train or tune an agent effectively unless you understand what great performance looks like in your domain. Practice systems thinking. Map how agents interact across teams and workflows. Design coordination across multiple agents and ensure smooth handoffs to humans when complexity increases. Strengthen change resilience. Work in rapid test-deploy-learn cycles. Adjust logic frequently as models evolve and business priorities shift. Refine your prompt craftsmanship. Treat prompts as training tools. Continuously improve language and structure to drive accuracy, tone, and consistency. Design hybrid work intentionally. Define clear escalation paths. Shift human KPIs toward outcomes, not activity. Help your team focus on high-value judgment, persuasion, and relationship work while agents handle scale. This tip is adapted from “ To Thrive in the AI Era, Companies Need Agent Managers ,” by Suraj Srinivasan and Vivienne Wei. . . . When Integrating AI, Focus on Psychological Safety Integrating AI changes how your team works together. While leaders often focus on tools and automation, the real impact often shows up as declining interpersonal trust, coordination, and decision-making. To avoid this kind of dysfunction, treat AI adoption as a team development effort—not just a tech upgrade—by focusing on psychological safety. Reframe AI adoption as a learning process. AI tools should be introduced with the expectation of discovery, not perfection. Signal that early missteps are valuable learning moments, not failures. Make questioning AI a norm, not a red flag. Model fallibility and curiosity. Openly admit when AI outputs confuse you or when you’ve made mistakes using it. Encourage open reflection through rituals like after-action reviews that explore where AI helped (or didn’t). Curiosity about AI behavior helps build psychological safety. Create smart failure protocols. Differentiate between smart risks and sloppy mistakes. Test AI in low-risk settings first, gather feedback, and adapt. Share what’s learned visibly so others grow from it. Emphasize human connection. As AI takes on more tasks, your team’s interpersonal collaboration becomes even more critical. Hold space for conversations about concerns, uncertainties, and roles—and empower your team to trust their own judgment to override AI’s recommendations. This tip is adapted from “ How to Foster Psychological Safety When AI Erodes Trust on Your Team ,” by Jayshree Seth and Amy C. Edmondson. . . . Unblock Your AI Transformation Most large-scale companies have initiated AI pilots and provided widespread access to AI tools. But while an AI pilot may succeed individually, saving time and money and making processes more efficient, those gains often don’t scale across the organization. To move beyond isolated, short-term wins, you need to rethink processes, roles, and systems as a whole. Here’s how. Start with a clean-sheet redesign. Don’t layer AI onto broken workflows. Step back and rebuild processes from scratch. Map how work would flow if agents and humans collaborated from day one. Define where AI acts and where humans orchestrate, then design seamless handoffs before writing any code. Capture knowledge before it disappears. Treat organizational expertise as a core asset. Identify where critical knowledge lives and make it explicit. Pair experienced operators with builders to document rules, judgment, and context. Position this as legacy-building so people engage instead of resisting. Manage AI like a workforce. Shift from tool governance to workforce management. Define who can create agents, what they can do, and how performance is tracked. Build systems to monitor accuracy, permissions, and accountability—the same way you would manage employees. Redesign roles around higher-value work. Don’t let productivity gains disappear. Reallocate time toward design, oversight, and decision-making. Clarify new responsibilities so people guide and validate AI outputs instead of doing manual tasks. This tip is adapted from “ The ‘Last Mile’ Problem Slowing AI Transformation ,” by Karim R. Lakhani et al. . . . Protect Your Team from AI-Fueled Overload AI promises productivity gains, but without guardrails it can quietly intensify work instead of reducing it. As tasks accelerate, expectations rise. People attempt more, juggle more—and struggle to stop. If you don’t deliberately shape how AI is used, you risk increasing your team’s workload, straining their attention, and exhausting them. Here’s how to protect them from AI overload. Build intentional pauses. As speed increases, create structured moments to slow down. Before major decisions, require a brief pause to test assumptions, surface one counterargument, and reconnect to core goals. Protect short intervals for reflection between work cycles. These pauses regulate tempo and prevent overload from accumulating. Sequence the work. AI enables constant background activity. Don’t let that dictate your pace. Batch non-urgent notifications together, hold updates until natural breakpoints, and protect focus windows from interruptions. Encourage teams to move work forward in defined phases instead of reacting to every output immediately. Sequencing reduces fragmentation and protects decision quality. Re-anchor work in human connection. As AI makes solo work easier, intentionally create space for dialogue and collaboration. Schedule brief check-ins and shared reflection moments. Encourage discussion before finalizing major outputs. Multiple human perspectives sharpen thinking and restore energy that continuous AI interaction can drain. This tip is adapted from “ AI Doesn’t Reduce Work—It Intensifies It ,” by Aruna Ranganathan and Xingqi Maggie Ye.

Our Favorite Management Tips on Leading with AI

Summary.

Partner Center