Leadership Authenticity and Responsible AI

Lisa Davis discusses leadership, responsible AI, and her book on navigating a male-dominated corporate infrastructure.

In this AI Marketers Guild community conversation, host David Berkowitz and co-host Kathleen Murphy interview Lisa Davis, author of The Only Woman in the Room: How to Win in a Workplace That Is Still Built for Men. Drawing on her 40-year career as a technology executive—including Chief Information Officer (CIO) roles at Intel, Blue Shield of California, and within the Department of Defense—Davis shares critical leadership insights on navigating career advancement, driving organizational change, and building authentic relationships.

The discussion bridges the gap between organizational diversity and the modern challenges of AI transformation. Davis, who also serves as the Chief Strategy Officer for the Global Council of Responsible AI, details the cultural, structural, and technical barriers to successful AI integration. The conversation outlines how leaders can implement responsible AI governance, establish effective employee incentives, and use emerging technologies to democratize innovation across industries.

Key moments

[3:34] Why Is the Modern Workplace System Still Biased Against Women and Minorities?

Lisa Davis explains that the modern workplace system remains structurally biased because corporate frameworks were originally designed after World War II around a household model where men went to work and women stayed home.

Although more than half of women are currently in the workforce—with 60% of those working women acting as the primary breadwinners—workplace structures have not evolved to support the modern family. Davis outlines that "the system" encompasses infrastructure issues like flexible work policies, paid family leave, and affordable childcare. Without federal, state, and corporate policy shifts in these areas, women and diverse leaders frequently drop out of the workforce because balancing their careers and personal lives becomes structurally unsustainable.

[5:30] Can AI Technology Help Fix Structural Workplace Inequality?

Lisa Davis suggests that while AI democratizes access to data and changes daily workflows, it cannot solve core structural workplace inequalities without proactive policy changes from human leaders and governments.

Addressing systemic challenges like paid family leave or affordable childcare requires active leadership, policy interventions, and administrative accountability rather than automated technological solutions. However, Davis notes that those who reach executive roles can leverage their positions to influence and implement better internal policies. Ultimately, solving infrastructure issues requires a call to action at the federal, state, and corporate levels to support both partners in a modern household.

[8:59] Why Do Most Enterprise AI Transformations Fail to Deliver an ROI?

Lisa Davis attributes the widespread failure of enterprise AI transformations to organizations treating AI as an isolated technology project rather than aligning business workflows, governance, and organizational accountability.

Citing an MIT report showing that 90% of enterprises fail to see a return on investment (ROI) from AI, Davis emphasizes that successful digital transformation must start at the top and align all the way down. Too often, executive boards and CEOs push for AI adoption but scapegoat digital or IT teams when projects fail, leaving business units with zero accountability. To drive actual business outcomes, organizations must establish robust governance frameworks, align funding, prioritize projects based on measurable returns, and build joint accountability across all business units.

[10:39] How Should Leaders Structure a Successful AI Transformation Workflow?

Lisa Davis advises that leaders should never start an AI transformation with the technology itself, but should instead begin by analyzing business workflows to find where automation can drive clear ROI and business outcomes.

Successful technology integration requires mapping out existing processes, identifying friction points, and prioritizing use cases that yield meaningful business outcomes. Leaders must secure cross-company alignment, build accountability directly into performance plans and incentives, and actively monitor progress via operational dashboards. Without structured governance and top-down alignment, business units will implement disjointed point solutions or localized bots, resulting in high maintenance costs, architectural chaos, and severe cybersecurity risks.

[15:18] Why Is Authenticity a Critical Leadership Trait in the Era of Generative AI?

Lisa Davis asserts that while generative AI provides instant access to virtually any content, it cannot replicate a leader's unique human wisdom, lived experiences, and authentic perspective.

As AI-generated content floods the workplace, genuine human authenticity becomes a key differentiator for successful leaders. True authenticity manifests through core values, honesty, decision-making transparency, and alignment between a leader's words and actions. Davis counters the myth that authentic leaders are "soft" or low-performing, explaining that they are highly focused on bottom-line results but prioritize the growth, mentorship, and empowerment of their teams.

[23:11] How Can Leaders Build a Strong Personal Brand and Develop Effective Allies?

Lisa Davis explains that while strong performance is the baseline, building a powerful personal brand and developing allies requires deliberate exposure, active networking, and cultivating supporters who advocate for you when you are not in the room.

Delivering flawless results is rarely enough on its own to secure a promotion. True professional progression relies heavily on exposure, strong relationship management, and consistency of actions that align with your core values. Davis stresses the importance of finding allies—particularly male allies in male-dominated industries—to champion diverse leaders. Furthermore, she emphasizes that leaders who reach executive status have a responsibility to mentor and "reach behind" to pull other diverse candidates up to the leadership table.

[24:47] How Do You Incentivize Employees to Adopt AI and Increase Productivity?

Lisa Davis recommends establishing clear organizational alignment by embedding specific AI adoption goals and bonus incentives directly into employee annual performance plans.

Since culture often eats strategy for breakfast, driving AI adoption is primarily a change management and cultural shift rather than a technological challenge. Because workers juggle competing priorities, they will focus on what leadership actively measures and rewards. By tying bonus structures and performance reviews to collaborative AI milestones, executives can create the necessary corporate infrastructure to sustain technological transformation and ensure team-wide buy-in.

[27:37] How Can Enterprises Address Algorithmic Bias and Liability in AI Recruitment Tools?

Lisa Davis notes that businesses are increasingly being held legally liable for biased decisions made by AI tools, making human oversight essential for high-stakes workflows like recruitment.

Using automated AI software to screen job applicants has introduced significant algorithmic bias, leading to recent lawsuits where companies were held liable for discriminatory rejection patterns. To mitigate these legal and ethical risks, Davis stresses that enterprises must keep "humans in the loop," especially for sensitive workflows in HR, healthcare, and finance. Organizations must recognize that AI models can sometimes return inconsistent or degrading quality of information, meaning automated workflows require continuous evaluation and human verification.

[32:58] Why Does the Career Pipeline Gap Persist for Women Seeking Executive Roles?

Lisa Davis explains that the career pipeline gap persists because women frequently face structural penalties when taking time off to start a family, causing them to fall behind male peers in career progression and wealth accumulation.

Even as women earn more undergraduate and advanced degrees than ever before, they remain highly underrepresented in leadership, representing only 11% of U.S. corporate CEOs. This drop-off occurs primarily in mid-level roles when women start families and find themselves without systemic support, such as paid family leave or affordable childcare. Davis notes that because the U.S. is the only modern industrialized nation without federally mandated paid family leave, many women are forced to permanently sacrifice career velocity, making it critical for companies to redesign internal support policies.

[38:51] How Does Generative AI Democratize Technical Innovation for Non-Technical Professionals?

Lisa Davis highlights that generative AI democratizes technology because, unlike older cloud or digital platforms, users do not need a background in engineering to leverage it and contribute to its governance.

AI interfaces allow professionals from diverse fields—such as sociology, law, anthropology, and marketing—to directly engage with and direct high-level technical tasks. Davis stresses that this accessibility makes it more critical than ever for diverse voices to participate in the conversation around AI development. With women currently representing less than 20% of technical engineering roles (down from 34% in the mid-1980s), broad societal participation is necessary to prevent a small, non-representative group of developers from dictating the future trajectory of human-centric technology.

[42:24] What Is the Grace AI Framework and How Does It Support Enterprise AI Governance?

Lisa Davis explains that the Grace AI Framework, developed by the Global Council for Responsible AI, is designed to help enterprises establish operational controls and safety guardrails for adopting artificial intelligence.

Operating in partnership with established cybersecurity frameworks like the federal NIST (National Institute of Standards and Technology) framework, Grace AI assists companies in measuring and auditing their AI systems' safety. The Global Council for Responsible AI aims to drive the adoption of this framework at federal, state, and county levels worldwide. Davis urges technology and marketing professionals to get involved with the council to help model, test, and advocate for standardizing AI governance controls across different business sectors.

[44:36] What Examples of Governance Guardrails Are Necessary for Enterprise AI?

Lisa Davis emphasizes that critical AI governance guardrails must include comprehensive auditability, decision-making traceability, and clear data lineage.

In standard IT security, any corporate system must offer complete transparency regarding where its data originates and how its processes operate. However, many current AI models operate as "black boxes," leaving creators and users unable to explain how specific algorithmic outputs are generated. To manage this liability, Davis and Berkowitz suggest using Retrieval-Augmented Generation (RAG) architectures in enterprise settings. RAG solutions ensure that the data feeding the AI is human-curated and fully traceable, ensuring compliance, safety, and verifiable decision pathways.