Frozen compliance budgets, multiplying AI risk: the CCO job becomes triage
An American Banker column urges compliance chiefs to swap constraint for strategy. With budgets flat, the real decision is which controls a bank agrees to run thinner.
The Investor · Invest desk
What happened
- Most compliance leaders report frozen or reduced budgets and headcount, according to an American Banker opinion column.
- Around 85% of banks report some level of AI deployment.
- Only about 15% of surveyed executives place themselves at the leading edge of AI use.
- The U.S. saw more than 30 filings in 2025 from fintechs and other nontraditional applicants seeking charters, acquisitions or conversions.
- The column's prescription is that CCOs abandon a constraint posture, raise their risk tolerance and act as strategic leaders.
Why it matters
- decision With headcount fixed, taking on AI and digital-asset oversight means naming which existing controls get thinner coverage, and that naming lands on the CCO rather than the budget holder.
- constraint Region-by-region control retooling draws on the same analyst and engineering hours the new technology work needs, which caps how much else compliance can absorb regardless of ambition.
- contradiction The same column asks for greater risk tolerance and describes hazards that are harder to detect and interpret; tolerance without detection capacity is not a posture, it is unmeasured exposure.
- precedent If mandates get rewritten while supervision is permissive, the rewrite becomes the baseline for the next tightening cycle, and reclaiming control resources will require an enforcement event to...
A flat budget is a prioritization engine whether or not anyone calls it that. Anything added under one gets funded by letting something else run thinner, so the practical content of the American Banker column's advice [2] is a list of controls an institution is prepared to see degrade. The column does not supply the list.
Its own figures show where the strain sits. Seventy percentage points separate the banks running AI somewhere in the business from the executives who describe themselves as leading-edge users of it [6]. Presence in the room is not command of the subject: a further 35 points separate the risk and compliance leaders who say they sit in enterprise AI decisions from that leading-edge group [7]. That leaves a large population of institutions where models are in production and the people accountable for the control environment rate their own fluency as unremarkable.
Worth noting what is not disclosed. These arrive as 2025 survey data with no named sponsor, instrument or sample size [3], which makes them usable as direction and useless as a benchmark. No CCO is walking an unattributed 15% into a board meeting and coming out with headcount.
The origin story matters to the arithmetic. AI entered compliance through narrow work where the payoff could be measured: anti-money-laundering alerts, sanctions screening, transaction monitoring [4]. Enterprise-wide deployment reverses the direction of the review. Compliance stops assessing tooling it commissioned for itself and starts assessing decisions made elsewhere, which is a larger job performed by the same or fewer people [12].
The timing argument is that regulatory easing opens a rare window to redefine the mandate [11]. Windows close. A function that converts control capacity into strategic influence during a permissive stretch does not get the analysts back when supervision tightens, and the charter, acquisition and conversion filings from nontraditional applicants [13] mean the supervised field is getting more crowded while the compliance bench is not.
The column dates the current shape of the role to post-crisis reform two decades ago [5]. That architecture was paid for by enforcement pressure, and enforcement pressure is exactly what the present environment withholds. Which is why "reinvention" here reads less as ambition than as the only available answer to a budget that has stopped growing while the surface area under review has not [14].
What to watch
- Whether named survey data with a disclosed sample and sponsor confirms the 85% deployment and 15% leading-edge figures, or revises them.
- How many of the 2025 nontraditional charter, acquisition and conversion filings are actually approved, and on what conditions.
- The first supervisory findings that fault AI governance staffing rather than AI model output, which would reprice the reallocation the column recommends.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence22
- Adoption38
- Hype gap+36
- Incentives66
- Confidence30
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Global banks face more fragmented regulatory frameworks across jurisdictions, which can force CCOs to retool controls to match a particular region's regime, a process the column calls costly and destabilizing.
- [2]
The column argues the CCO role must move from a posture of constraint to one of enablement, with CCOs working smarter and with more tolerance for risk even as their resources are cut.
- [3]
The column presents the 85%, roughly-half and 15% figures as 2025 survey data without naming the survey, its sponsor or its sample size.
- [4]
For most financial institutions AI adoption began in targeted use cases with clear benefits, improving accuracy and reducing noise in anti-money-laundering alerts, sanctions screening and transaction monitoring.
- [5]
The column says the foundational dimensions of the CCO role still hark back to changes spurred by the financial meltdown two decades ago.
- [6]
There is a gap of about 70 percentage points between the share of banks reporting some AI deployment and the share of executives who consider themselves at the leading edge of AI use.
- [7]
Taking 'approximately half' as 50%, there is a gap of about 35 percentage points between risk and compliance leaders involved in enterprise AI decisions and those who describe themselves as leading-edge AI users.
- [8]
Roughly 85% of banks report some level of AI deployment.
- [9]
Approximately half of U.S. risk and compliance leaders surveyed in 2025 indicated involvement in enterprise AI decision making.
- [10]
Only about 15% of the executives surveyed said they considered themselves to be at the leading edge of AI use.
- [11]
The column says current regulatory easing provides a rare window of opportunity for CCOs to redefine their mandate and reposition compliance as a partner in business strategy.
- [12]
Most compliance leaders report reduced or frozen compliance budgets and headcounts, even as banking risk becomes more complex.
- [13]
In the U.S. in 2025 there were more than 30 filings from fintechs and other nontraditional applicants for de novo charters, bank acquisitions or conversions.
- [14]
The column says AI adoption, growth of digital assets and competition from nontraditional players are introducing new potential hazards that are harder to detect, interpret and control.
Sources
1 independent publisher whose own reporting we read for this story.
- americanbanker.comBank chief compliance officers need to adjust their risk tolerance
1 article · August 21, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.
Topics
- Bank Regulation and SupervisionFollow
- AI in Financial Crime ControlsFollow
- Fintech Bank ChartersFollow
- AI GovernanceFollow
- Compliance OperationsFollow
Entities
- American BankerFollow
- Chief Compliance Officer (CCO)Follow