If there was one clear message coming out of Compliance Week’s January 2026 AI conference, The Leading Edge: Applying AI and Data Analytics in E&C, it was not about tools, vendors, or futuristic promises. It was about discipline. More specifically, it was about something compliance professionals have preached for decades and are now being pressured to skip: the “why.”
In a recent episode of the podcast From the Editor’s Desk, I sat down with Compliance Week Editor in Chief Aaron Nicodemus to gather his reflections on the conference and its implications for compliance leaders. What emerged was not a story about artificial intelligence replacing compliance, but about AI exposing weaknesses in how organizations make decisions, manage pressure from the top, and integrate ethics into innovation. For compliance professionals, the discussion was a reminder that AI is not a technology problem. It is a governance problem.
The Step Everyone Is Skipping: Why Before What
One of the most striking takeaways from the conference came from Jen Gennai, former AI Ethics and Compliance Advisor at Google. Her message was deceptively simple: companies are skipping the “why.” Organizations are rushing to implement AI tools without first articulating what problem they are trying to solve or why AI is the appropriate solution. Instead of defining the use case and then selecting the right tool, teams are buying technology first and hoping value emerges later.
For compliance professionals, this should sound uncomfortably familiar. Risk management, third-party due diligence, investigations; every mature compliance process begins with a defined purpose. There is a reason the first step in the third-party risk management process is the Business Rationale. This is the ‘why’, requiring a business sponsor to explain why your organization needs a new or different business partner. Yet when AI enters the picture, that discipline often evaporates. The result is experimentation without accountability and pilots without strategy.
The irony is that compliance already knows how to do this. The failure is not a lack of knowledge; it is pressure.
Tone at the Top, Revisited: Pressure Without Direction
According to a recent Compliance Week and konaAI study released at the conference, more than 60 percent of compliance officers feel pressure from the board or C-suite to “use AI.” Not to use it in a specific way. Not to achieve a defined outcome. To use it. This top-down mandate creates a new kind of compliance risk. When leadership demands adoption without guidance, teams feel compelled to move quickly, sometimes cutting corners they would never cut in other risk domains.
This is not inherently nefarious. Boards are doing what they believe is necessary to keep their organizations competitive. But pressure without clarity creates the conditions for poor governance. Compliance leaders must recognize this moment not as a threat, but as an opening. Because when leadership says “use AI,” compliance has an opportunity to respond with structure: identify manual pain points, define defensible use cases, and align AI deployment with existing policies and ethical standards. The mandate may be broad, but the implementation can and should be deliberate.
Humans in the Loop: Why Oversight Is Not Optional
Another recurring theme from the conference was the danger of letting AI evaluate AI. Scaling tools without human oversight compounds error. One flawed assumption becomes many. Bias multiplies. Outputs drift. The lesson here is not anti-technology; it is pro-governance. AI works best when humans remain embedded throughout the lifecycle: selecting tools, defining scope, reviewing outputs, and deciding whether the system is working at all.
This aligns squarely with long-standing compliance principles. Judgment-heavy decisions, investigations, escalations, and remediations must remain human. Attempting to automate them introduces fairness and defensibility risks that no compliance program can explain away after the fact. AI should accelerate compliance work, not absolve responsibility for it.
Trust and Integrity: The Core Compliance Tension with AI
The most profound tension discussed at the conference was philosophical. Compliance programs are built on trust and integrity. AI, by contrast, is often perceived as opaque, untrustworthy, and occasionally wrong. This creates a credibility problem.
Why would a compliance function that spends years telling employees to act ethically, verify sources, and question assumptions deploy a tool that fabricates answers or cannot explain its reasoning? If compliance cannot articulate why an AI system aligns with the organization’s ethical standards, it should not be deployed, no matter how efficient it appears to be. Trust is not just about outputs. It extends to inputs, data quality, and understanding how systems interact with information. AI amplifies what it is given. Bad data does not improve through automation; it spreads faster.
Iteration Over Perfection: Learning Is Part of the Process
A healthy counterpoint emerged as well: AI is not a one-shot deployment. It requires iteration. Early failures are not proof that AI does not work; they are evidence that learning has begun. Several speakers emphasized that AI improves through feedback. Teams must be willing to correct it, teach it, and refine its outputs over time. Compliance professionals who abandon tools after one or two imperfect attempts misunderstand how the technology functions.
That said, iteration does not excuse carelessness. Learning must occur within guardrails: governance frameworks, usage boundaries, and documentation matter more, not less, when tools evolve.
Compliance as Value Creator, Not Speed Bump
One of the most encouraging insights from the conference was how AI is reshaping compliance’s role inside organizations. When compliance is involved early, before tools are rolled out, it becomes a partner in innovation rather than an obstacle.
Nicodemus pointed out companies like Robinhood, and Hemma Lomax, Deputy General Counsel, Vice President, and Head of Business Integrity at DocuSign, illustrated this point clearly. Compliance teams that embed themselves in product development and operational change help shape tools that work within ethical and regulatory boundaries from the start. That credibility compounds.
Lomax noted that at DocuSign, she and her compliance teams have gone further, creating AI agents that perform defined tasks continuously, with built-in ethical guardrails. When these tools are handed to new users, the hard questions have already been answered. This is how compliance becomes a competitive advantage; not by saying no, but by helping the business say yes safely.
No Experts, Only Practitioners
Another refreshing theme from the conference was humility. No one claimed to be an AI expert. Especially not in compliance. That matters. When technologies move quickly, false certainty is dangerous. Compliance professionals should not be intimidated by those who claim mastery. Instead, they should lean into their strengths: skepticism, documentation, and principled decision-making. AI does not require omniscience. It requires informed judgment.
The Vibe Shift: From Fear to Engagement
Perhaps the most telling insight came not from the stage, but from the hallways. Compared to earlier events, the mood around AI has shifted. Compliance professionals are no longer crossing their arms in resistance. They recognize the benefits and risks and want to engage. No one believes AI will disappear. The debate is no longer whether to use it, but how. Some organizations will lean in aggressively. Others will move cautiously. All will need compliance to guide those choices. The most effective analogy offered was this: AI is like a very confident intern. Smart. Fast. Occasionally wrong. Useful, but never in charge.
Conclusion: AI Is a Compliance Opportunity, If Compliance Leads
The Compliance Week AI conference made one thing clear: AI is not undermining compliance. It is testing it. Programs that lack clarity, governance, or confidence will struggle. Programs that know who they are, what they stand for, and how they make decisions will thrive. For compliance professionals, the question is not whether AI belongs at the table. It already sits there. The real question is whether compliance will claim its seat, not as a roadblock, but as the function that ensures innovation aligns with integrity. That is not a burden. It is an opportunity.
