AI is creating new opportunities for RegTech firms, but investment alone does not guarantee a competitive edge. As financial institutions accelerate AI adoption, vendors face growing pressure to demonstrate how their technology can deliver measurable improvements in compliance, strengthen risk management and reduce operational costs.
RegTech Analyst and Parker & Lawrence Research recently released a major industry report looking at the global state of RegTech in 2026. As part of their findings, they looked at the return on investment on the commercial benefits of AI, and how this differed based on company side.
FinTech Global recently sat down with several leading players in RegTech to get their thoughts on these findings in a first part of a two-part series.
In the view of Remonda Kirketerp-Møller, CEO of Muinmos, when considering why some firms are generating significantly higher returns than others, this is often due to the fact they ask AI a different question.
She said, “Most firms start with “how much work can this take off our desks?” The firms pulling ahead ask “how much more business can we take on, and how quickly?”
The research that is behind this question, she said, makes the point clearly. Reduced manual effort is reported by roughly half of firms in every return group, from laggards (47.5%) to leaders (50.6%).
She added, “Efficiency has become the entry ticket, not the differentiator. Leaders are far more likely to report faster onboarding and approvals (43.4% against 27.5% for laggards) and faster product launches or market expansion (18.1% against 2.5%).”
Laggards, on the other hand, are more likely to count headcount savings as a main gain. In other words, laggards point AI at cost.
Kirketerp-Møller added, “Leaders point it at the moment where compliance meets revenue. Onboarding is the clearest example: it is the first real experience a client has of a firm, and every day it takes is a day that client is not trading, investing or transacting. A firm that shortens it has not just saved effort. It has brought revenue forward.”
Kara Modica, Managing Director at ACA Group, believes on this point that the differentiator is no longer access to AI.
She said, “The technology is widely available. The firms generating the strongest returns are those combining AI with deep domain expertise and a clear understanding of the business outcomes they are trying to achieve. AI is most effective when it is applied to well-defined industry problems, informed by institutional knowledge and high-quality data.”
In many ways, Modica added, AI amplifies expertise. Organizations that know their customers, workflows, risks, and opportunities can translate that knowledge into AI-driven decisions and insights that create measurable value.
Another key opinion on this was shared by Victor Chauhan, co-founder at Argus Pro. He stressed that the research showed the highest-return group leading in categories such as faster onboarding, approvals and reviews; faster product launches and market expansion; and lower technology costs. And not by small margins.
“They were around 1.6 times as likely to report faster onboarding, nearly twice as likely to report lower technology costs, and more than seven times as likely to report faster product launches,” he said.
In addition, the same group lagged in categories such as improved risk, fraud and control failures detection, as well as reduced headcount and outsourcing costs.
“Here the lowest-return group led, being around 1.5 times as likely to cite better detection and more than twice as likely to cite headcount reduction,” he remarked. These he said are two of the reasons most firms give for buying AI in the first place.
That headcount reduction finding, he said, was ‘quite interesting’. The group reporting the weakest returns is the group most likely to claim it. “When the whole strategy is to reduce cost, the savings are banked, but capabilities don’t improve, and growth never arrives,” said Chauhan.
Chauhan stressed some caveats, such as the fact these were self-reported returns, so the table shows what high-return firms attribute their returns to, not what caused them.
So, why are some companies generating substantially higher returns? This comes down, Chauhan states, to where the value lands.
He said, “Detection improvements (i.e., losses avoided) are real, but they’re hard to show on a P&L statement. They also mean a potential increase in workload, i.e., more to review. But speed is different. Faster onboarding and faster product launches mean revenue arrives faster; revenue gets reported in the accounts.”
According to RelyComply, when it comes to this topic, the firms getting most from AI treat it as a business transformation, not a standalone technology projects.
For the South African firm, three habits set them apart.
“They start with the commercial problem. They don’t adopt AI just because it’s available. First, they ask where better intelligence would improve decisions, reduce risk or give customers a real advantage. Then they choose the solution,”
In addition, those most successful get their data right. “AI can’t produce reliable insight from information scattered across disconnected systems, inconsistent records or outdated processes. Connecting the right data and building intelligence into existing workflows is what makes AI useful day to day.”
Another area is that they measure before and after. For example, a clear baseline shows whether AI is delivering a return and where further investment is justified.
The company said, “This discipline also helps firms back the use cases with the most potential, instead of spreading resources across too many disconnected experiments.”
Moving from efficiency to growth
How does AI move from operational efficiency to commercial growth?
For Chauhan, beyond automation, commercial value appears to have been earned by those firms that changed a process – whilst those who struggled appear to have added a tool to existing ones.
He said, “Automating a poor process just produces a faster poor process. What this means for RegTech firms is that our customers are buying a judgment, not a throughput. We’ve long argued that it’s less about ‘human in the loop’ and more about ‘removing the human from the hamster wheel’: automation that frees up time and lets compliance professionals focus on higher-value work. That’s AI worth buying.”
Meanwhile, Kirketerp-Møller stressed that such a move can be driven by measuring ourselves in our clients’ commercial terms rather than our own.
She explained, “A RegTech provider that sells “hours saved” is selling the one benefit every competitor can also claim. The more valuable conversation is about what clients can now do that they could not do before: open accounts the same day, launch a product without rebuilding the compliance process behind it, or enter a new jurisdiction through configuration rather than a year-long project.”
The Muinmos CEO stressed that we talk about completion rates rather than pass rates, due to the fact no tech providers should decide who gets approved.
“That is a regulatory and risk decision for the firm. What we should be accountable for is how many genuine applicants complete the process, how quickly, and with a file that stands up to scrutiny,” she said.
The same research, Kirketerp-Møller highlighted, points to a second, less obvious lever. “Leaders are also more likely to report lower technology costs. That is what happens when firms consolidate onto fewer, better connected systems instead of adding another tool for every new rule.”
Modica stated that the next evolution of RegTech is moving beyond task automation and toward intelligent decision support.
She said, “Rather than simply automating individual processes, firms should leverage AI to connect disparate data sources, normalize information, and generate actionable insights. Advances in agent-based and multi-step reasoning systems enable AI to analyze information iteratively, much like human experts do: gathering context, evaluating multiple inputs, and refining conclusions.”
This, Modica outlined, creates new opportunities not only for operational efficiency but also for proactive risk identification, client advisory services, and entirely new revenue streams.
RelyComply stated their view that AI becomes a competitive advantage when it gives customers something they value beyond lower costs.
The firm remarked, “Take AML and KYC solutions. Better risk detection, fewer false positives and faster customer reviews improve compliance operations. They also make the customer experience smoother.That strengthens a provider’s market position. It helps financial institutions scale with confidence, respond to changing risks and manage their compliance obligations more effectively.”
AI, the company stressed, also makes room for more distinctive services. These include richer risk assessments, more responsive monitoring and compliance workflows that adapt.
“When these capabilities are built into a broader product, they deepen customer relationships and help providers stand out in a competitive market. The shift is from selling automation to delivering better business outcomes,” said the company.
Building trust
How do you build trust into every stage of AI adoption? For RelyComply, in a highly regulated industry, commercial value cannot come at the cost of accountability.
The company said, “AI systems need proper governance: clear accountability, effective monitoring and appropriate human oversight. Decisions that affect customer access, risk classification or a financial crime investigation must be explainable and open to review.”
RelyComply added that model performance needs continuous checking also, stating that criminal behaviour evolves, data changes and models lose accuracy over time.
The firm said, “Ongoing validation, performance monitoring and documented controls catch weaknesses early. Left unchecked, those weaknesses erode customer confidence and fall short of regulatory expectations.”
Trust isn’t only a safeguard, the firm added, it’s also a commercial asset. “Financial institutions are more likely to adopt and expand AI-enabled solutions when they understand how the technology supports decisions and how its risks are managed.”
What makes a scaler successful
What separates firms that scale AI successfully from those that don’t? From the stance of RelyComply, the final difference between AI leaders and laggards is whether they can scale what works across the organisation.
The South African company remarked, “That takes more than technical infrastructure. Teams need the skills to interpret AI outputs, challenge recommendations and build insights into established workflows. Leaders need to align investment with business priorities. Technology and compliance teams need to work together on implementation, governance and performance measurement.”
The company concluded by stressing businesses don’t need a wholesale transformation on day one. “Start with one focused use case, set clear success criteria, test the outcomes and expand what works. What you learn from each deployment then shapes the next investment. Scaling is also an ongoing process. Regular reviews keep firms ready for new risks, changing customer expectations and emerging opportunities.”
For Kirketerp-Møller, a key separator comes from three things.
“First, they fix the process before they automate it. AI applied to a fragmented process scales the fragmentation: you get faster silos, and more alerts landing on the same people. The firms that scale run one orchestrated flow, where each automated step hands a complete case to the next.”
She went on, “Second, they can explain every decision. I trained as a lawyer, and accountability is where AI projects most often stall. A pilot can run on enthusiasm; scaling needs the head of compliance, and eventually the regulator, to accept the outcome.”
She raised the example of the EBA, who has found that over half of the serious compliance failures reported to its EuReCA database involved the improper use of RegTech tools.
“The firms that understand, and can evidence, what their systems do are the ones that earn the confidence to scale them,” she said.
On the third point, they measure what matters commercially. “If success is defined as fewer hours, the project ends when the hours are saved. If it is defined as faster time to revenue, better conversion and room to grow into new markets, AI keeps earning its place,” she finished.
The last word came from Modica, who made clear her view that successful AI adoption is just as much a data and operating model challenge, as a technology challenge.
She remarked, “Organizations that scale AI effectively invest in data quality, governance, and employee enablement. The principle remains the same as with previous technology waves: poor inputs produce poor outputs.
“Firms must ensure AI has access to accurate, relevant, and contextual information while also educating employees on when AI is the right solution and when traditional approaches are more appropriate. Without that discipline, costs can increase rapidly while business value remains elusive.”
Copyright © 2026 RegTech Analyst
Copyright © 2026 RegTech Analyst





