The latest enforcement actions reveal why the financial industry must move beyond compliance processes and toward more effective financial crime detection.
In the span of five days, two of America’s largest financial institutions faced significant regulatory enforcement actions involving failures in their anti-money laundering programs.
On October 8, the Office of the Comptroller of the Currency (OCC) imposed a $350 million penalty against American Express National Bank, citing systemic weaknesses in its Bank Secrecy Act and anti-money laundering (BSA/AML) compliance program. Regulators identified failures to adequately detect, evaluate, and report approximately $13 billion in suspected trade-based money laundering activity over the past decade.
On October 13, Bank of America faced a $1.2 billion FinCEN penalty for BSA/AML violations, alongside an $11.6 million OFAC sanctions settlement. According to the Treasury, the bank failed to timely report more than $5 billion in suspicious transactions, including over $1 billion associated with priority illicit finance risks.
These are extraordinary numbers. But the penalties themselves are not the most important takeaway.
The larger issue is what these enforcement actions reveal about the effectiveness of financial crime compliance.
For decades, financial institutions have invested billions in monitoring systems, compliance personnel, and investigative processes. Yet critical risks can still go undetected, remain uninvestigated, or fail to reach law enforcement when the information matters most.
The question is no longer simply whether institutions have compliance programs in place. It is whether those programs are actually effective at identifying and disrupting financial crime.
The gap between compliance activity and effectiveness
The findings against American Express and Bank of America point to different institutional failures, but a common underlying challenge.
At American Express, regulators found that the bank’s risk assessment was insufficiently aligned with its business model. The institution focused too narrowly on deposit products and failed to adequately account for risks associated with its much larger credit and charge card business.
At Bank of America, regulators identified longstanding monitoring and escalation deficiencies, including suspicious activity that was not investigated or reported in a timely manner. The OCC also highlighted the institution’s failure to adequately address weaknesses identified in a December 2024 enforcement order.
Neither case can be reduced to a single technology failure. Both involved broader deficiencies in governance, risk assessment, controls, staffing, or oversight.
But together, they illustrate an important distinction.
Operating a compliance program is not the same as operating an effective financial crime prevention program.
Institutions can generate alerts, complete investigations, maintain extensive documentation, and still miss meaningful threats.
Compliance activity is measurable. Effectiveness is harder.
And ultimately, effectiveness is what matters.
Financial crime is evolving faster than legacy approaches
The risks highlighted in these actions are not theoretical.
Treasury specifically identified suspected activity involving Chinese money laundering networks, potential trade-based money laundering, and illicit finance connected to the U.S. Southwest border. The Bank of America sanctions settlement also involved apparent violations across multiple sanctions programs.
These threats increasingly exploit the complexity of the global financial system.
Criminal networks operate across borders, institutions, intermediaries, and products. Individual transactions may appear unremarkable when viewed in isolation. The underlying risk often becomes clearer only when institutions connect information about entities, relationships, behavior, and broader intelligence.
This creates a fundamental challenge for traditional compliance architectures.
Systems built around static rules, fragmented data, periodic reviews, and manual investigations are often poorly positioned to recognize complex, evolving patterns of illicit activity.
And when investigators are overwhelmed by large volumes of low-value alerts, the consequences extend beyond operational inefficiency.
Important signals risk being buried.
We should be clear about the distinction: reducing false positives is valuable, but it is not the ultimate objective.
The real objective is to identify more meaningful risk, earlier, with greater accuracy and actionable context.
AI creates an opportunity to change the equation
The next generation of financial crime compliance should not simply automate existing processes.
It should materially improve what institutions can detect, understand, and act upon.
Advances in artificial intelligence, entity resolution, network intelligence, and workflow automation now make it possible to approach longstanding compliance challenges differently.
Rather than relying exclusively on isolated alerts, institutions can increasingly combine structured and unstructured intelligence to better understand relationships, identify emerging risks, and prioritize activity that warrants investigation.
AI can also help investigators assess information more quickly, reduce repetitive manual work, and produce better-supported decisions.
But deploying AI is not, by itself, a solution.
Effective compliance requires high-quality risk intelligence, comprehensive screening coverage, explainable decisions, strong governance, and continuous testing.
A faster system that misses the same risks is not progress.
And an automated system without appropriate oversight simply creates a different category of risk.
The opportunity is to combine modern AI capabilities with stronger underlying intelligence and disciplined risk management. This allows financial institutions to improve both operational efficiency and the effectiveness of their financial crime defenses.
These two outcomes should reinforce one another.
A leadership challenge, not simply a compliance challenge
There is another lesson that deserves attention.
The effectiveness of an AML program cannot rest solely with the compliance department.
It requires alignment between executive leadership, business operations, risk management, technology, and the board.
Institutions need to understand where their greatest exposures exist, whether their monitoring systems adequately cover those risks, and whether material weaknesses are being addressed before they become regulatory findings.
That means asking more difficult questions.
Are we measuring the volume of compliance work being completed, or the quality of the risks being identified?
Are our systems designed around the actual risks in our business?
Can our investigators access the intelligence needed to make informed decisions?
And when deficiencies are identified, can we demonstrate meaningful improvement?
These are strategic questions with implications well beyond regulatory compliance.
The future of compliance must be measured by outcomes
The American Express and Bank of America enforcement actions should prompt serious reflection across the financial services industry.
Not because every institution faces the same challenges, or because technology alone could have prevented these specific failures.
But because they highlight the consequences of allowing the gap between compliance obligations and actual risk detection to persist.
At Sigma360, we have long believed that the industry must move beyond treating financial crime compliance primarily as a regulatory exercise.
The opportunity is larger than improving productivity or lowering the cost of investigations.
It is about giving financial institutions the intelligence and capabilities they need to identify illicit activity more effectively, support sound decisions, and protect the integrity of the financial system.
Modern technology makes substantial progress possible. But realizing that potential requires institutions to rethink the way compliance programs are designed, evaluated, and managed.
The next chapter of financial crime compliance cannot be defined by how many alerts institutions process.
It must be defined by how effectively they identify and confront the risks that matter.
That is the standard the industry should be working toward.

