Key takeaways from my conversation with Erica Anderson Rooney, author of The AI Gap

This blog post provides a summarized version of my conversation with Erica Anderson Rooney, featuring key highlights with slight alterations for readability. To listen to the full discussion, check out the latest episode of She Builds with AI on Spotify, Amazon Music, or your preferred podcast platform.

In this episode of She Builds with AI, I sat down with Erica Anderson Rooney, executive coach, keynote speaker, founder of Her Collective, former Chief People Officer, and author of Glass Ceilings and Sticky Floors and The AI Gap: Women, AI, and the Next Great Leap Forward.

Our conversation explored AI through a lens that often receives too little attention: power. We talked about women in leadership, sticky-floor mindsets, AI bias, confidence, visibility, responsible AI, and why the conversation about “learning AI” needs to go far beyond tools, prompts, and productivity hacks.

Because AI is already shaping who gets hired, who gets promoted, who gets seen, and who gets invited into the rooms where decisions are made. That means the AI gap is not only a technical issue. It is a leadership issue, a systems issue, and in many ways, a deeply personal one.

AI Is Not Just a Skills Gap. It Is a Power Gap.

One of the strongest ideas Erica shared is that the AI gap should not only be understood as a skills gap. It is also a power gap.

That distinction matters.

When organizations talk about AI adoption, the conversation often starts with training, tooling, experimentation, or productivity. Those things are important, but they are not the full picture. AI fluency is quickly becoming connected to influence. The people who learn how to use AI early, who experiment with it, who build confidence with it, and who become visible around it are often the same people who get invited into the next set of strategic conversations.

They are the ones asked to help shape workflows, evaluate tools, define governance, influence leadership decisions, and decide what “good” looks like in an AI-enabled organization.

If women are not equally encouraged, supported, and rewarded for experimenting with AI, the gap does not stay small. It compounds.

That is why the question is not simply whether women use AI. The bigger question is whether women are present where AI is being shaped, implemented, governed, and trusted.

The Sticky Floors That Keep Women Playing Small

Before we talked about AI, Erica brought the conversation back to the internal patterns she explored in her first book, Glass Ceilings and Sticky Floors.

The phrase “sticky floors” describes the beliefs, behaviors, and internalized patterns that keep women stuck, even when they are capable, qualified, and already doing the work.

These patterns can look subtle from the outside.

Waiting until something is perfect before sharing it.
Assuming hard work will automatically be noticed.
Feeling the need to earn your worth again and again.
Questioning whether you belong in the room, even after you have already earned your seat.
Preparing longer, polishing more, and holding back while others move forward with more confidence.

What stayed with me is that Erica does not frame this as blaming women. Quite the opposite. She is very clear that women are operating inside systems that were not designed neutrally. But she also points to the beliefs many women have internalized inside those systems.

That becomes especially relevant in the age of AI.

If a woman already believes she is “not technical enough,” she may delay experimentation. If she delays experimentation, she may build fluency more slowly. If she builds fluency more slowly, she may be perceived as less ready for AI-related opportunities, even if she has the judgment, context, and leadership capability the organization actually needs.

In other words, sticky floors can become AI barriers.

AI Does Not Create Inequality From Scratch

A central theme in our conversation was Erica’s point that AI does not create inequality from scratch. But it can accelerate the patterns that already exist.

That is one of the most important things leaders need to understand.

AI systems are trained on existing data, language, patterns, decisions, and narratives. They often reflect what has been common in the past, not necessarily what is fair, capable, inclusive, or right for the future.

If leadership has historically skewed male, AI may learn a narrow pattern of what leadership looks like. If most CEOs or CFOs in the data are men, AI-generated assumptions about a “top candidate” can quietly reinforce that image. If hiring, promotion, performance management, or leadership assessment tools are built on biased historical data, organizations may end up scaling the very patterns they say they want to change.

This is where responsible AI becomes very practical.

It is not enough to ask whether a tool is fast, impressive, or efficient. Leaders need to ask what it is optimizing for, whose patterns it reflects, which candidates it surfaces, which signals it values, and what it may be missing.

AI can sound confident. That does not mean it is right.

The Risk of Automating Inequality

One of the most important leadership risks Erica highlighted is the possibility of automating inequality without realizing it.

This can happen when organizations use AI in hiring, promotion, performance reviews, leadership decisions, or talent identification without questioning the outputs deeply enough.

If an AI-supported system recommends one candidate over another, leaders are still accountable for that decision. If a tool surfaces a narrow group of people as “high potential,” someone still needs to ask why. If an applicant tracking system repeatedly prioritizes similar profiles, organizations need to understand whether it is truly identifying capability or simply reproducing familiar patterns.

This is where leaders need to build the habit of interrogation.

What blind spots might this output have?
Which groups may be underrepresented in the result?
What criteria were actually used?
Can we defend this recommendation?
What would change if we asked for a more diverse comparison set?
Are we optimizing for familiarity or for future potential?

The point is not to reject AI in people decisions entirely. The point is to stay responsible for the judgment that surrounds it.

AI can support decision-making. It should not replace accountability.

AI Fluency Does Not Require Becoming an Engineer

Another part of the conversation that I found especially important was Erica’s perspective on AI fluency.

Many women still say, “I am not technical.” And because of that, they assume AI is not for them, or that they need to become engineers before they can use it meaningfully.

Erica challenges that idea directly.

You do not need to be an electrician to turn on the lights. You do not need to be a mechanic to drive a car. And you do not need to be a technologist to use AI well in your role.

That does not mean technical expertise is no longer valuable. Of course it is. But AI fluency is not the same as technical mastery.

For many professionals, AI fluency starts much more simply. It starts with curiosity. It starts with taking a real pain point from your work or life and asking an AI system how it might help. It starts with learning how to give context, ask better questions, challenge the output, and refine the result.

Erica described AI not as a tool to be passively used, but as something closer to a specialized subject matter expert. Not an intern in the background doing only basic tasks, but a thought partner that can help you research, synthesize, reframe, pressure-test, and surface blind spots.

That mindset shift matters.

When women realize they do not need permission, perfection, or a technical identity to begin, AI becomes less intimidating. It becomes something they can explore, shape, and use on their own terms.

AI as Leverage, Not Replacement

A recurring theme in our conversation was agency.

Erica does not talk about AI as something that should replace human judgment, voice, or presence. She talks about it as leverage.

Used well, AI can help women protect time, reduce friction, build confidence, prepare more effectively, and create capacity for the work that matters most. It can help with research, presentation development, business building, content creation, strategic thinking, and making sense of complex information.

But the human remains responsible for the output.

This is where the conversation becomes more nuanced than “AI saves time.” Saving time is valuable, but the deeper question is what we do with the time and energy AI gives back.

Do we simply fill that space with more work?
Or do we use it to think more clearly, recover more intentionally, lead more thoughtfully, and show up with more confidence?

That is why this conversation goes beyond the workplace.

AI affects how women make decisions, run businesses, lead teams, care for others, manage energy, communicate their value, and structure their lives. For many women, the opportunity is not only to become more productive. It is to become less trapped in the invisible overwork that has often been normalized.

What Leaders Should Audit First

For organizations, Erica emphasized that responsible AI requires the right people in the room from the beginning.

If AI is being used in hiring, promotion, performance management, or leadership decisions, HR cannot be brought in at the end as a compliance checkpoint. HR needs to be involved early, alongside IT, business leaders, legal, and the people who will actually live with the systems being introduced.

This is not only a technology rollout. It is a change management process.

Organizations need to map the outcome they want, understand where AI is influencing decisions, define what accountability looks like, and make sure the governance process is understandable enough for people to actually follow.

They also need to look around the table and ask who is missing.

If everyone involved in selecting, implementing, or governing AI systems thinks similarly, has similar experiences, and represents similar perspectives, important risks will be missed. Sometimes the most valuable insight comes from someone closer to the work, someone who sees the friction points, the unintended consequences, or the human impact more clearly than senior leaders do.

Responsible AI is not only about policies. It is about participation.

Knowing What Not To Outsource

One of the most thoughtful parts of the conversation was the reminder that using AI well also means knowing where not to use it.

Not every process should be automated simply because it can be.

Erica shared the example of a leader in life insurance who did not want to replace call center interactions with AI because customers were often calling during some of the hardest moments of their lives. In that context, efficiency was not the highest value. Empathy, trust, and human presence mattered more.

That example stayed with me because it captures something essential about human-centered AI.

The goal is not to put AI everywhere. The goal is to understand where AI creates value, where it reduces unnecessary friction, and where it would remove something deeply human that should remain protected.

This applies in business, leadership, and private life.

There are moments where AI can help us prepare, organize, summarize, draft, and think. And there are moments where what is needed is not more speed, but more humanity.

The future of AI will not only be shaped by what we automate. It will also be shaped by what we consciously choose not to automate.

Closing The AI Gap Means Getting Hands-On

Toward the end of the episode, Erica shared a clear vision for what better could look like.

More women in positions of power.
More women building, leading, governing, and shaping AI.
More women challenging the defaults instead of adapting themselves to systems that were never neutral.
More organizations using AI to close gaps rather than quietly scale them.

But that future does not happen automatically.

It requires women to get hands-on, stay curious, and build fluency before they feel fully ready. It requires leaders to question confident outputs, involve diverse voices, and take responsibility for the systems they implement. It requires organizations to understand that AI adoption is not just about speed, but about what kind of future gets built through that speed.

For me, this conversation was a strong reminder that women do not need to wait until they feel technical enough, senior enough, or ready enough to enter the AI conversation.

The room is already being built.

The systems are already being shaped.

And we need more women inside that process now, not later.

Listen to the Full Episode

This conversation with Erica Rooney is full of thoughtful insights on women, power, AI fluency, leadership, responsible AI, sticky floors, and the future of work and life.

If you are a woman exploring how to build confidence with AI, a leader responsible for AI adoption, or someone thinking about how technology can shape opportunity more fairly, this episode offers a grounded and deeply relevant perspective.

Listen to the full episode of She Builds with AI and share it with someone who should be part of the AI conversation now.

Because closing the AI gap is not only about learning new tools.

It is about making sure women help shape what comes next.

Stay Connected with Erica Rooney:

🔗 Website: https://www.ericaandersonrooney.com
🔗 LinkedIn: https://www.linkedin.com/in/ericarooney
🔗 Instagram: https://www.youtube.com/@EricaAndersonRooney
🔗 YouTube: https://www.youtube.com/@EricaAndersonRooney

---

About She Builds with AI

She Builds with AI is a podcast spotlighting women who are shaping the future with AI and emerging technology across industries and around the world.