Attending INSOL London 2026 was a reminder of how quickly our industry is evolving. The panel on international asset tracing and recovery underscored a reality we see every day at Kroll: The tools, techniques and expectations around enforcement are changing at unprecedented speed. What once required months of painstaking manual work can now be accelerated through artificial intelligence (AI), but the implications go far beyond efficiency. They touch on how we enforce, how we safeguard integrity and how we prepare for a future where fraud actors are just as adept at using technology as we are.
From Fragmentation to Clarity
For decades, asset tracing meant piecing together fragments: PDFs typed into spreadsheets, bank account statements converted to pivot tables and registries searched one jurisdiction at a time. Today, AI platforms can replicate some of the analyst’s manual workflow in minutes, synthesizing multilingual news, corporate filings and digital footprints into a coherent case file. The difference is not just speed, it is the ability to see the full picture rather than a partial sample. While this is true, the analyst’s role is still crucial in checking the underlying data and challenging the output.
This matters for enforcement. When office holders need to lock down assets quickly, the ability to map ownership structures, identify side businesses and trace connections across borders is critical. AI can provide that clarity, but as several panelists emphasized, it cannot replace human judgment. It can point us toward red flags, but it is the practitioner who decides what is actionable.
Making Music Out of the Noise
Another demonstration that resonated deeply was how forensic AI tools can contextualize communications across platforms such as WhatsApp and email into chronological narratives. Instead of drowning in noise, investigators can focus on conversations that matter, such as those involving money flows. For enforcement professionals, this is transformative. It means faster identification of relevant evidence, quicker progression from suspicion to proof and ultimately stronger recovery outcomes for creditors.
Yet the lesson is not that AI does the work for us. It is that AI helps us get to the work that matters sooner, demonstrating “time to value.” That resonates with our experience at Kroll: Creditors do not benefit from months lost in administrative processes. They benefit when intelligence is converted into enforceable action —quickly and reliably.
Legal Lens: Reliability, Authenticity, Confidentiality
The panel also highlighted the legal challenges of AI adoption. Courts worldwide are grappling with reliability, deepfakes and confidentiality. Guidance varies by jurisdiction, but common themes are emerging: Lawyers remain accountable for submissions, experts must disclose and verify their use of AI and factual evidence requires heightened scrutiny.
For enforcement, this is a double-edged sword. On one hand, AI can accelerate the initial stages of instruction, helping lawyers and office holders make sense of vast data sets under urgent conditions. On the other, we are already seeing how easy it has become for bad actors to use AI tools to produce forged documents and amended data sets.
Human in the Loop
A recurring theme was the human in the loop. AI can synthesize, grade and contextualize, but judgment remains ours. That is not a limitation, it is a safeguard. Enforcement is not about delegating responsibility to machines, it is about equipping professionals with tools that expand their reach. As fraudsters adopt AI to obscure assets, our agility in applying these tools will determine whether we stay ahead.
Bias is another concern. If AI directs us to “look here,” we risk missing evidence elsewhere. The solution is not to abandon AI, but to cultivate inquisitiveness. Practitioners must interrogate outputs, ask what has been missed and return to primary sources. At Kroll, we emphasize that AI is a research assistant, not a decision maker. It can guide, but it cannot conclude.
Looking Ahead: The Next 12 Months
Where does this leave us? AI will not replace professionals, but it will expand the scope of work. Cases once deemed too small or too complex may now be viable. Efficiency gains will lower thresholds to entry, opening enforcement to more creditors and more jurisdictions. Fraudsters will use AI, but so will we, and the race will be won by those who combine speed with judgment.
In the next 12 months, expect AI to become more accurate, balancing stochastic creativity with deterministic reliability. Expect more litigation, not less, as technology both exposes fraud and enables new forms of concealment. And expect enforcement professionals to remain indispensable—not despite AI, but because of it. The integrity of our profession lies in human accountability, and AI, properly harnessed, strengthens rather than weakens that foundation.
