Some law and advisory firms have taken it upon themselves to investigate this question: How can AI support the compliance function?

Surveyed compliance and investigations professionals want to deploy AI to improve internal efficiencies and transform the way they work, agreeing that the investment of money in the AI technology space is a worthy one. And compliance wants to be involved in the process, advocating for roles to be clearly spelled out.

Many of the professionals polled are not clear where to start, thanks to their worries about unstructured datasets, Intellectual property issues, and a desire for customization, never mind tracking a wide array of rules from different jurisdictions. (And did I mention those rules keep changing?)

Let’s look at two reports based on survey results – one from September 2025 and one from 2024. Not too much has changed; compliance professionals are still saying their use of AI could be described as early stage. What will move them to more fulsome adoption?

Use of AI in compliance and investigations (2025)

Survey results compiled by the law firm White & Case reveal a period of transition, one in which early adopters are realizing tangible benefits while running up against growing pains such as implementation challenges, gaps in policy development, and the inherent risks of deploying this transformative technology.

Here are some key findings:

• Overall, 36% of respondents report using AI in both their compliance and investigations processes, with a further 26% using it for compliance tasks only.
• Respondents that are publicly listed companies are almost twice as likely (44%) to use AI for both compliance and investigations compared with their private sector counterparts (23%). White & Case observes that this difference likely is due to the larger data volumes and potentially higher investment capacity that public entities often exhibit, and possibly the accordingly higher expectations from regulators around the use and deployment of data analytics in underlying compliance programs.
• Corporates show significantly higher adoption of AI (43%) compared with private equity firms (10%). Six out of every ten (59%) of the highest-revenue-generating respondents already leverage AI for both compliance and investigations, with just a 14% adoption rate among the lowest-revenue-generating respondents. This, of course, points to the resources needed to provide the financial means and technical expertise required of businesses investing in and deploying these tools.
• For respondents currently using AI, the largest cohort (36%) has been using it for one to two years, closely followed by those implementing it for one year or less (34%), with almost half of the highest-revenue organizations using the tools for two to five years.
• The predominant rationale behind implementing AI for compliance and investigations are two elements: added efficiency (73% of respondents) and cost savings (71%). White & Case quotes one member of the ethics and compliance function of a US company: “We use AI for compliance and investigations to lower the amount of manual work. Manual work has become time consuming due to the changing regulations and the complexity of the process. So, the use of AI became inevitable at a certain point.”
• Relatedly, the top use cases are summarizing documents (88%) and reviewing documents during investigations (85%). One respondent specifically mentioned how well the contextual information is being captured and processed using AI, making it incredibly helpful for compliance and investigations teams. “There is a better understanding of everyday and uncommon risks in our activities.”
• Key concerns in this area center on data security and reliability. Data protection is the top issue (64%), which demonstrates respondents’ anxieties about managing sensitive personal and corporate data as well as ensuring compliance with privacy regulations in a variety of jurisdictions. Inaccuracy (57%) is the next major concern, underscoring the risks linked with potential biases in algorithms and hallucinations from generative AI outputs.

Risks & rewards of AI in compliance and investigations (2024)

A September 2024 DLA Piper AI Governance Report offers cross-sector data from 600 respondents, and asked companies about deployment of AI in terms of its compliance and business value and the policies developed around its use.

Several companies said they are contemplating what to do next with AI, with 71% in the “explorer” phase.

A live poll of compliance and investigations professionals following the law firm’s recent event for said professionals revealed that the uptake of AI in the compliance function is much lower than the use of AI by their organizations as a whole.

The poll also suggested that data protection, data quality, and hallucination are the top three concerns for compliance and investigations professionals.

“Bias and accountability plus fragmented global regulation are also key risk areas, and the knowledge and skills gap is the greatest barrier that prevents companies from leveraging AI in investigations. Interestingly, data suggests that cost is not the number one constraint. It is a concern, but not the determinative factor,” DLA Piper noted.

Some boards of directors can be roadblocks, too. Some question whether the tools are capable enough (remember, legal folks are trained to be highly risk-averse), and they and compliance professionals are also worried about reputational risk if anything goes wrong.

And that concern is not unfounded, considering that regulators will be monitoring how well businesses oversee their use of AI tools, even as they encourage innovation in the space and increasingly use the tools themselves.

All of this leads DLA Piper to wonder whether there is an “inherent tension between the use of disruptive technology and accepted compliance practice,” again, given compliance’s key concern over reputational risk.

But DLA Piper reminds compliance teams that there are some clear rewards and opportunities here: AI offers the capacity to identify high-risk behavior, evaluate suspicious trading patterns, observe red flags for money-laundering risks, and manage existing systems and controls.

The firm lists some of the key AI compliance products available, including the following: 

• Predictive compliance analytics; 
• AI-enhanced whistleblower systems, including anonymization of reporters (although they queried whether this is truly AI); 
• Compliance automation software; 
• AI tools for risk assessment & management; 
• AI-powered document analysis and review; and 
• Regulatory change management. 

Cultural barriers

Like DLA Piper, White & Case hinted at the inherent conflict that compliance professionals feel about AI tools, given their natural concerns for technology malfunctioning, reputational risk, and data security.

White & Case even said this, referencing personal liability concerns: “Some legal and compliance teams remain skeptical of AI, fearing that automation could either dilute their influence or introduce errors for which they will be held responsible.”

One factor that could help mitigate any hesitations, besides everyone just getting more comfortable with the tech over time and more educational opportunities helping everyone leverage the tools more effectively, is the reality that you have to use these tools to keep up with your competitors.

And regulatory bodies are developing their own tools for investigative and enforcement purposes.

Some countries (like Canada – see our All In AI conference coverage) are well along in developing AI-specific audit and ethical frameworks around the tools, and are offering training and innovation incentives to help organizations gain more confidence.

Finally, it might be worth seeking out vendors offering targeted training, customization features, and business-specific solutions, plus reinforce the importance for organizations to augment the use of AI with human judgment and whose tools specifically enable this.