On a gray Wednesday morning in London, LRN, a provider of ethics, compliance and corporate culture solutions, gathered a group of women from across the compliance profession for a breakfast discussion on the evolving role of women in the field.
The conversation moved easily between themes of authenticity and intelligent risk-taking, anchored by a fireside chat between Rebecca Mayfield, chief compliance officer and senior counsel at Merlin Entertainments, and Patsy Doerr, chief people and culture officer at LRN.
It was during that discussion that the conversation turned, inevitably, to AI. As organizations continue to experiment with the technology, questions around culture, governance and board-level understanding are becoming harder to ignore. After the session, I caught up with Doerr to explore those questions in more depth.
Doerr, who joined LRN in February 2026, brings experience from senior roles at UDR, Credit Suisse and Thomson Reuters. Her perspective aligns closely with the findings of LRN’s 2026 Program Effectiveness Report, which draws on insights from more than 2,500 ethics and compliance professionals globally.
The data suggests that while 39 percent of organizations are now using AI in at least one compliance function, many still struggle to explain how it improves outcomes. High-impact programs are pulling ahead, with 42 percent deploying AI-enhanced training compared to 30 percent of their peers. Yet governance remains underdeveloped, with limited documentation around model purpose, data lineage and validation. The result is a widening gap between adoption and effective oversight.
It is within this context that culture, as Doerr explains, becomes the defining factor. As the adoption of AI continues at pace, she believes the biggest challenge is not technological in nature, but cultural. ‘In my experience, this is the case with anything that has such a big impact,’ she says. ‘There are many things companies need to do, but a few key cultural considerations stand out.’
At the top of the list is resistance to change. ‘Generally speaking, everybody has a different capacity for change, so understanding that capacity as it relates to AI is critical,’ she explains, pointing to concerns about job loss and a lack of capability as key drivers. ‘That makes resistance to change even more important.’
Patsy Doerr, chief people and culture officer at LRN
Closely tied to this are persistent skill gaps. ‘I’m a firm believer that education is at the heart of everything,’ Doerr says. ‘But there are skill gaps because this is a new technology.’ Some of these gaps are generational, while others reflect uneven exposure depending on role. This helps explain why, despite 39 percent of organizations now using AI in at least one compliance function – according to the 2026 Program Effectiveness Report – many still struggle to articulate its impact.
For Doerr, alignment is equally important. ‘Companies also need to think about alignment with their values and how AI fits within those values,’ she says. ‘Organizations need clarity on how AI will be used and how that aligns with what they stand for.’ Without that clarity, adoption risks becoming fragmented, a challenge reflected in the report’s finding that many organizations lack a coherent approach to AI governance.
Underlying all of this is a more practical issue with cultural implications. ‘There’s also a big issue with data quality,’ she notes. ‘In most organizations, data sits in different departments without strong processes to integrate it.’ While often framed as a technical problem, the impact is cultural, affecting collaboration and engagement across functions.
Direction from the top of an organization, however, remains the defining factor. ‘Leadership buy-in is always critical,’ Doerr says. ‘Not just talking the talk but actually walking the walk.’
That leadership intent does not always translate cleanly through the organization. Middle management, she argues, often becomes the point where transformation efforts lose momentum. ‘What often happens is that leadership buys into a strategy, but as it flows through the organization, the message evolves or changes,’ she says. ‘It’s like the game of telephone we played as kids, in which the message shifts along the way.’
With AI, that distortion is amplified by uncertainty. ‘There are added challenges around understanding, fear and change,’ she says. To counter this, managers need to take a more active role in interpreting and translating strategy. ‘Middle managers need to educate themselves to understand how AI can support them and help alleviate fears about job loss,’ she says, adding that the focus should shift toward opportunity. ‘They should focus more on the opportunities AI presents, especially in terms of data analysis and spending more time on insights and actions rather than just gathering data.’
At the same time, the human element cannot be overlooked. ‘They also need to stay focused on their people, helping them navigate the change,’ she says. ‘Managers should keep communication open, transparent and candid as things continue to evolve.’
If there is a single capability that underpins successful adoption, it is learning agility. ‘I have a very strong view, based on research, that learning agility is one of the top predictors, if not the top predictor, of success,’ Doerr says. In practice, this means more than training programs. ‘It’s critical that leaders, managers and employees practice learning agility,’ she explains, emphasizing the need to ‘learn from every experience, accept feedback and adapt behavior intentionally.’
This requires deliberate cultural design. ‘We need to build systems that actively create an environment where learning agility is embraced company wide,’ she says. That environment depends on transparency, inclusion and psychological safety. ‘People need to feel safe to challenge the norm, admit what they don’t know about AI and ask questions without fear of repercussions.’
These principles extend to the boardroom. ‘The governance piece is critical,’ Doerr says. ‘Setting clear guidelines for how the technology will be used, ensuring it operates within a framework of ethics, compliance, transparency and positive intent.’ Yet many organizations are still catching up in this area, with limited documentation around model purpose and data lineage continuing to hinder effective oversight.
For learning to stick, it must move beyond one-off interventions. ‘One-off training doesn’t work, particularly with something evolving as quickly as AI,’ she says. ‘Learning has to be ongoing and it has to be embedded into how work actually happens.’ That means integrating tools into daily workflows and reinforcing behaviors over time. ‘Organizations need to build systems and processes where learning is part of the day-to-day, not something separate from it.’
Equally important is how that learning is measured and rewarded. ‘If we’re not reinforcing and recognizing learning behaviors, they won’t stick,’ she says. ‘Organizations need to think about how they reward curiosity, adaptability and growth not just outcomes.’
As companies continue to experiment with AI, striking the right balance between innovation and control becomes critical. ‘This is where psychological safety and accountability have to go hand in hand,’ Doerr says. ‘You want to encourage experimentation, because that’s how organizations learn and innovate but it has to happen within a clear framework.’
That framework must be clearly communicated. ‘Employees need to understand what responsible use looks like,’ she says, pointing to the importance of guidelines around ethics, data use and compliance. Within those boundaries, there should be real encouragement to test, learn and iterate. ‘Leaders have to model that it’s okay to try things, learn from mistakes and improve,’ she says, while reinforcing that ‘experimentation doesn’t mean a lack of discipline.’
This balance is also shaping how organizations approach broader cultural priorities such as ESG or DEI. ‘What we’re seeing is a move toward embedding these concepts into how organizations operate, rather than treating them as standalone initiatives,’ Doerr says. In the context of AI, that shift becomes more pronounced. ‘Questions of fairness, bias, transparency and inclusion are central,’ she explains. ‘The focus should be on how those principles show up in the design, deployment and governance of AI.’
AI, in this sense, is accelerating integration. ‘It forces organizations to be more explicit about their values and how they are put into practice,’ she says, particularly by requiring diverse perspectives in decision-making and a more intentional approach to identifying bias.
Ultimately, an AI-ready culture is defined by integration and intent. ‘An AI-ready culture is one where technology is fully integrated into how the organization thinks, operates and makes decisions not something sitting on the side,’ Doerr says. Hallmarks include ‘a high degree of learning agility,’ ‘transparency in how AI is being used’ and ‘a strong foundation of trust because employees understand both the opportunities and the boundaries.’
The distinction between meaningful adoption and superficial uptake is increasingly clear. ‘Organizations that are getting it right tend to have very open communication, strong cross-functional collaboration and leaders who are actively engaged,’ she says. By contrast, weaker approaches are easy to spot. ‘You see a lot of disconnected pilots, limited understanding among employees and a lack of clarity around purpose,’ she notes. ‘That’s usually a sign that the cultural foundations haven’t been fully established yet.’