Jennie Glazer is the Chief Executive Officer of Coqual, a global think tank helping companies redesign how they work and lead. On June 16, Jennie and Elysee Consulting Founder and CEO Abenaa Hayes are leading Chief Members through a hands-on workshop focused on what’s happening inside organizations as AI reshapes work and how leaders are engaging that shift in practice. Attendees will work through real scenarios like the ones below and leave with language, prompts, and tools they can use with their teams immediately. Members: RSVP via the Chief app.

Nadia, CHRO for a global retail company, came to me after a nose-dive in trust scores in a recent employee survey.

She was doing the dance many leaders are doing right now inside large, complex companies: facing a tired workforce, pressure to grow, tighter budgets, political and economic volatility, and a board asking when AI investments would start to pay off.

AI transformation was her company’s top focus and budget investment. They brought in a big consulting firm that created a careful, polished rollout plan. Big considerations in the plan included “licenses are expensive,” and “risk needs to be managed.” Because some teams had clearer use cases, the rollout began with selected groups to drive early adoption, learning, and expansion.

It all sounded sensible.

Then I sat down with people within the same company to understand the trust deficit beneath the surface, and I uncovered a deep disconnect between leaders and their teams. From where employees sat, the rollout carried another message:

  • They got the tool. (We did not.)
  • They got the training. (We are waiting.)
  • They must be the future of this company.

That is the part of AI transformation that more leaders need to name. AI is changing how work gets done. It is also changing how people read opportunity and their future at your company.

Access becomes a signal. Training becomes a signal. Permission to experiment becomes a signal.

Coqual and Catalyst’s new research shows why this matters. Only 34% of employees said their organization makes AI training and resources available to all. Only 35% said their organization helps them build AI skills relevant to their role. And only 37% of leaders demonstrated the mix of AI decision skills, inclusive behaviors, and flexible mindset we call Convergent Leadership. The payoff is measurable: Convergent Leaders were far more likely than non-Convergent Leaders to deliver higher team productivity, 93% vs. 34%; increased revenue or profit margin, 53% vs. 26%; and increased customer loyalty or satisfaction, 58% vs. 22%.

To bring these results to life in your own organization, consider this: The real opportunity for human-first AI is in how it drives stronger decisions, teams, and trust. Leading organizations are anchoring transformation with three leadership principles.

1. Access: Who Is Being Equipped to Make Better Decisions With AI?

Access especially matters inside talent systems. Krishna, head of internal mobility at a big tech firm with more than 200,000 employees, designed an AI-enabled internal mobility platform. Employees upload their resumes. AI identifies skills. The system suggests roles. Hiring managers can then see internal candidates whose skills appear to align with open jobs. 

The promise was real: better mobility, stronger retention, and a clearer view of talent already inside the organization. Before launching the new process, Krishna asked me to facilitate a conversation with cross-functional executives around a deceptively simple question:

What if the system starts with self-description?

It was the right question. Self-assessment is useful, but imperfect. Tasha Eurich’s work on self-awareness found that while most people believe they are self-aware, only 10% to 15% of those studied met the criteria for self-awareness. And of course, people (especially across identities) do not describe themselves equally. Some are comfortable claiming a skill after trying it once. Others need years of proof before they will say they are good at something. Some have sponsors who translate their value. Others are doing excellent work that remains under-described and under-seen.

Krishna’s team realized, as we dug into the questions, that an AI-generated talent match could overvalue the people most fluent in being visible. So they built better questions into the process:

  • Who may be qualified and under-described?
  • Whose strengths are least likely to appear in this data?
  • What evidence beyond self-description should shape the recommendation?

Capability does not always arrive in polished language. Human-first AI requires leaders to check whether new systems can see the talent that old systems already missed.

2. Quality: Is AI Improving Decision Quality and Effectiveness?

Sharon flew her leadership team from 13 time zones to the Midwest for a laptop-open AI working session. Her team, focused on operations inside a global beauty company, wanted to drive AI adoption. Before they could begin, Sam raised his hand and asked:

Could we refine the goal from adopting AI to improving quality and effectiveness through AI?

To say the room went from “Minnesota nice” to an eruption of emotions is an understatement. People had tea to spill. They vented about dashboards tracking usage and the mixed effects they were seeing on their work. Some teams were using AI more often. The output moved faster. The real question was whether decisions were improving. Sharon listened and let the room change and own this refined goal. That moment mattered. It made space for a different conversation, one focused on customer satisfaction, employee learning, and innovation.

This is what Convergent Leadership looks like in action. AI decision skills help leaders ask whether the tool is improving the work. Inclusive behaviors make it safe for someone like Sam to challenge the goal in front of the rest of the class. A flexible mindset allows the leader to adapt when a better question appears. By the end of the session, the team had built a simple quality check leaders could use with their teams:

I want to see the judgment you applied, the tradeoffs you considered, and the questions that still need human review.

Leaders can also ask:

  • What decision should this improve?
  • What assumptions need to be tested?
  • Whose perspective is missing?
  • What still requires human review?

The strongest teams will use AI to improve how they think together.

3. Time: What Should AI Make Possible?

At the end of almost every session I run on human-first AI, I ask one question:

If AI helps us save time, what should you do with that time?

I love that question because it changes the emotional meaning of AI. One pharmaceuticals team decided to use some of the time AI saved for learning and experimentation. They created space to try tools together, compare what they were learning, and make sense of AI as a team. That may sound small. It is not.

If saved time is lost to more meetings, requests, and pressure, employees will quickly understand the bargain. If saved time becomes space for better thinking, stronger coaching, deeper learning, and work the team has not had time to do well, AI begins to feel like an investment. Leaders can say:

If AI helps us save time, we are going to decide intentionally how to use that time. Some of it may go to speed. Some of it needs to go to learning, better judgment, stronger collaboration, and work we have not had time to do well.

Then ask:

  • What should AI help us do better?
  • What decisions should be improved because of it?
  • What should we stop doing altogether? And my favorite question:
What human work deserves more time now?

That last question is the one I hope all leaders sit with.

Human-first AI means telling people how access decisions are made. It means setting a higher bar for judgment. It means checking whether new systems can see overlooked talent. It means deciding, out loud, what saved time is for.

Because the promise of AI should not be a faster treadmill. It should be better work, better decisions, and more space for people to contribute at their best.