For all the planning and predicting that goes on everywhere, there are events and developments that occur that we absolutely didn’t expect and catch us offguard. At the start of 2025, I would have bet all the money in my pocket against all the money in the pockets of everyone else in the world that there was no way I would end up with a dog in my house. Yet, as 2025 ends, there is now a dog in my house. AccountsRecovery asked a handful of professionals from across the credit and collection industry to share what surprised them the most about 2025.
Carl Briganti, CSS Impact
The biggest surprise of 2025 was not that AI gained traction in the collections industry, but how quickly it moved from experimentation to mission-critical infrastructure. Agencies have begun deploying AI at a deeply technical, use-case-driven level, using it to scale operations, lower cost structures, and protect margins in ways that weren’t previously possible. What exceeded expectations was the downstream impact on human performance, with AI materially elevating agent consistency, decision-making, and productivity. Just as significant was the improvement in legal and regulatory compliance, as AI enforced process discipline and reduced risk at scale. Together, these shifts redefined how agencies think about growth, efficiency, and operational safety moving forward.
Todd Wyatt, Scott & Associates
I have been genuinely surprised by how rapidly AI technology has advanced within the business landscape. What once appeared experimental has quickly become a practical tool driving efficiency and better decision-making. In the ARM industry specifically, I believe AI is poised to fundamentally reshape the collections function by enhancing quality control, tailoring consumer engagement, and optimizing process workflows. My business is no longer asking if we should adopt AI, but how fast we can do so. Building a Data Science team wasn’t in the playbook until 2025 decided to call an “audible”
Dale Watts, Armstrong & Associates
For me the biggest surprise was not any single technology, but just how fast companies had to adapt to change and how that pace keeps accelerating. Long-term planning doesn’t mean what it used to, time frames that once felt safe no longer give us a clear picture of where things are headed. These days, success looks more like the ability to adapt and execute than having everything perfectly mapped out.
Leslie Bender, Eversheds Sutherland
The most surprising development in consumer finance in 2025 was not the adoption of artificial intelligence itself—technological advancement had been anticipated for years—but rather the velocity and scope of deregulation that enabled the aggressive deployment of AI across underwriting, credit scoring, fraud detection, and debt collection. The Trump administration’s pro-innovation, anti-regulatory posture created an environment in which consumer-facing financial services companies embraced automation with minimal governmental oversight, fundamentally transforming the operational landscape of consumer credit and collections while introducing unprecedented legal and ethical risks.
The Deregulatory Framework and Industry Response
The administration’s 2025 AI policy prioritized market-driven innovation over prescriptive regulation, effectively rolling back dozens of federal regulatory guidance and changing enforcement priorities established during prior administrations. This shift removed regulatory friction that had previously constrained algorithmic decision-making in consumer finance, particularly in areas governed by the Fair Credit Reporting Act (FCRA), the Fair Debt Collection Practices Act (FDCPA), and prohibitions against Unfair, Deceptive, or Abusive Acts or Practices (UDAAP).
Financial services companies and their mission-critical vendors responded with rapid deployment of AI systems that automated decisions historically requiring human judgment. Underwriting models incorporated machine learning algorithms capable of processing thousands of non-traditional data points, from social media activity to utility payment histories. Credit scoring evolved beyond FICO-based assessments to incorporate behavioral analytics and predictive modeling. Debt collection operations adopted AI-powered communication systems that personalized outreach timing, messaging, and channel selection based on debtor profiles.
The efficiency gains are proving substantial, helping financial services companies with shrinking margins do more with less. Institutions reported reduced operational costs in collections, more efficient consumer self-service tools and platforms, and improved accuracy in fraud detection. These improvements promise to position early adopters as market leaders, creating competitive pressure across the industry to deploy comparable AI capabilities.
Emerging Legal Risks and Compliance Challenges
The surprise of 2025 lay not in AI’s capabilities but in the potential legal vulnerabilities created by deploying these systems in a deregulated environment. Three statutory frameworks may present immediate compliance challenges.
Fair Credit Reporting Act (FCRA) Implications
FCRA mandates accuracy and fairness in consumer reporting and requires institutions to provide adverse action notices when automated decisions negatively impact credit eligibility. AI-driven credit scoring models, however, introduced opacity, complicating compliance. Algorithmic bias—the systematic, repeatable errors in computer systems that lead to unfair outcomes favoring certain groups over others—emerged as a critical risk. Models trained on historical data perpetuated discriminatory patterns embedded in legacy lending practices, potentially violating FCRA’s fairness requirements.
Institutions remain strictly liable for inaccurate adverse action notices and incomplete investigations, yet many lack the technical capacity to audit complex machine learning models for bias. The absence of regulatory guidance on algorithmic transparency standards created uncertainty regarding the sufficiency of existing compliance protocols.
Fair Debt Collection Practices Act (FDCPA) Considerations
FDCPA prohibits harassment, false representations, and deceptive practices in debt collection. AI-enabled collections systems, designed to optimize contact frequency and messaging intensity, pose risks of statutory violations. Automated outreach platforms lacking human oversight may generate communication patterns that, while technically efficient, approach or exceed FDCPA’s boundaries on contact frequency and permissible communication times.
Transparency requirements prove particularly challenging. AI-generated messaging that personalized language based on debtor psychology raises questions regarding whether such communications constitute deceptive practices, especially when consumers were unaware they were interacting with automated systems rather than human collectors. State regulators have made it clear that they will hold AI agents, chatbots and other resources responsible for compliance (just as they would humans).
UDAAP Enforcement Risks
The prohibition against Unfair, Deceptive, or Abusive Acts or Practices has provided regulators with broad authority to challenge AI deployments that produce consumer harm. Institutions that fail to disclose AI-driven decision-making processes may face claims alleging that such opacity constitutes abusive practices. Models that generated unfair outcomes—defined as substantial consumer injury not reasonably avoidable and not outweighed by countervailing benefits—created regulatory vulnerability even in the absence of intentional discrimination.
Explainability emerged as a critical compliance need. The “black box” nature of advanced machine learning models made it difficult for institutions to provide the human-readable explanations increasingly demanded by state regulators and consumer advocates, exposing them to UDAAP scrutiny.
Strategic and Practical Implications
The surprising lesson of 2025 was that technological capability divorced from regulatory guardrails created compliance risk rather than competitive advantage. Institutions that rush to deploy AI without robust governance frameworks may find themselves exposed to enforcement actions, reputational damage, and class action litigation.
Effective risk management in this environment may require five critical measures. First, regular algorithmic audits to detect and remediate bias in credit scoring and underwriting models ensure ongoing FCRA compliance. Second, implementation of explainable AI frameworks—technical approaches that make algorithmic decision-making interpretable to non-specialists—enhances transparency and mitigates UDAAP exposure. Third, maintaining human-in-the-loop oversight for high-risk credit decisions and collections activities provides a compliance safeguard against automated errors and abusive practices. Fourth, updated compliance training that incorporates AI governance and ethical considerations may equip personnel to identify and address algorithmic risks. Fifth, proactive engagement with regulators to align AI practices with evolving legal standards positions institutions to anticipate rather than react to enforcement priorities.
The Path Forward
As 2025 concluded, the consumer finance, credit and collections industries confronted a paradox: deregulation had enabled technological advancement that improved operational efficiency while simultaneously creating legal vulnerabilities that threatened institutional stability. The administration’s light-touch regulatory approach accelerated AI adoption and discouraged states from creating bottlenecks, leaving open critical questions regarding fairness, transparency, and consumer protection unanswered.
The defining surprise of 2025 was the recognition that in our industry, innovation without governance generates risk. Institutions that treat AI deployment as purely a technological initiative, rather than a compliance and ethical imperative, may discover that algorithmic efficiency cannot compensate for regulatory exposure. The year demonstrated that, in a deregulated environment, competitive advantage belongs not to the fastest AI adopters but to organizations that embed legal and ethical safeguards into algorithmic systems from inception. The 2025 experience established that sustainable AI transformation in consumer finance requires balancing innovation with accountability—a lesson that may shape industry practices and regulatory responses for years to come.
For all the planning and predicting that goes on everywhere, there are events and developments that occur that we absolutely didn’t expect and catch us offguard. At the start of 2025, I would have bet all the money in my pocket against all the money in the pockets of everyone else in the world that there was no way I would end up with a dog in my house. Yet, as 2025 ends, there is now a dog in my house. AccountsRecovery asked a handful of professionals from across the credit and collection industry to share what surprised them the most about 2025.
Carl Briganti, CSS Impact
The biggest surprise of 2025 was not that AI gained traction in the collections industry, but how quickly it moved from experimentation to mission-critical infrastructure. Agencies have begun deploying AI at a deeply technical, use-case-driven level, using it to scale operations, lower cost structures, and protect margins in ways that weren’t previously possible. What exceeded expectations was the downstream impact on human performance, with AI materially elevating agent consistency, decision-making, and productivity. Just as significant was the improvement in legal and regulatory compliance, as AI enforced process discipline and reduced risk at scale. Together, these shifts redefined how agencies think about growth, efficiency, and operational safety moving forward.
Todd Wyatt, Scott & Associates
I have been genuinely surprised by how rapidly AI technology has advanced within the business landscape. What once appeared experimental has quickly become a practical tool driving efficiency and better decision-making. In the ARM industry specifically, I believe AI is poised to fundamentally reshape the collections function by enhancing quality control, tailoring consumer engagement, and optimizing process workflows. My business is no longer asking if we should adopt AI, but how fast we can do so. Building a Data Science team wasn’t in the playbook until 2025 decided to call an “audible”
Dale Watts, Armstrong & Associates
For me the biggest surprise was not any single technology, but just how fast companies had to adapt to change and how that pace keeps accelerating. Long-term planning doesn’t mean what it used to, time frames that once felt safe no longer give us a clear picture of where things are headed. These days, success looks more like the ability to adapt and execute than having everything perfectly mapped out.
Leslie Bender, Eversheds Sutherland
The most surprising development in consumer finance in 2025 was not the adoption of artificial intelligence itself—technological advancement had been anticipated for years—but rather the velocity and scope of deregulation that enabled the aggressive deployment of AI across underwriting, credit scoring, fraud detection, and debt collection. The Trump administration’s pro-innovation, anti-regulatory posture created an environment in which consumer-facing financial services companies embraced automation with minimal governmental oversight, fundamentally transforming the operational landscape of consumer credit and collections while introducing unprecedented legal and ethical risks.
The Deregulatory Framework and Industry Response
The administration’s 2025 AI policy prioritized market-driven innovation over prescriptive regulation, effectively rolling back dozens of federal regulatory guidance and changing enforcement priorities established during prior administrations. This shift removed regulatory friction that had previously constrained algorithmic decision-making in consumer finance, particularly in areas governed by the Fair Credit Reporting Act (FCRA), the Fair Debt Collection Practices Act (FDCPA), and prohibitions against Unfair, Deceptive, or Abusive Acts or Practices (UDAAP).
Financial services companies and their mission-critical vendors responded with rapid deployment of AI systems that automated decisions historically requiring human judgment. Underwriting models incorporated machine learning algorithms capable of processing thousands of non-traditional data points, from social media activity to utility payment histories. Credit scoring evolved beyond FICO-based assessments to incorporate behavioral analytics and predictive modeling. Debt collection operations adopted AI-powered communication systems that personalized outreach timing, messaging, and channel selection based on debtor profiles.
The efficiency gains are proving substantial, helping financial services companies with shrinking margins do more with less. Institutions reported reduced operational costs in collections, more efficient consumer self-service tools and platforms, and improved accuracy in fraud detection. These improvements promise to position early adopters as market leaders, creating competitive pressure across the industry to deploy comparable AI capabilities.
Emerging Legal Risks and Compliance Challenges
The surprise of 2025 lay not in AI’s capabilities but in the potential legal vulnerabilities created by deploying these systems in a deregulated environment. Three statutory frameworks may present immediate compliance challenges.
Fair Credit Reporting Act (FCRA) Implications
FCRA mandates accuracy and fairness in consumer reporting and requires institutions to provide adverse action notices when automated decisions negatively impact credit eligibility. AI-driven credit scoring models, however, introduced opacity, complicating compliance. Algorithmic bias—the systematic, repeatable errors in computer systems that lead to unfair outcomes favoring certain groups over others—emerged as a critical risk. Models trained on historical data perpetuated discriminatory patterns embedded in legacy lending practices, potentially violating FCRA’s fairness requirements.
Institutions remain strictly liable for inaccurate adverse action notices and incomplete investigations, yet many lack the technical capacity to audit complex machine learning models for bias. The absence of regulatory guidance on algorithmic transparency standards created uncertainty regarding the sufficiency of existing compliance protocols.
Fair Debt Collection Practices Act (FDCPA) Considerations
FDCPA prohibits harassment, false representations, and deceptive practices in debt collection. AI-enabled collections systems, designed to optimize contact frequency and messaging intensity, pose risks of statutory violations. Automated outreach platforms lacking human oversight may generate communication patterns that, while technically efficient, approach or exceed FDCPA’s boundaries on contact frequency and permissible communication times.
Transparency requirements prove particularly challenging. AI-generated messaging that personalized language based on debtor psychology raises questions regarding whether such communications constitute deceptive practices, especially when consumers were unaware they were interacting with automated systems rather than human collectors. State regulators have made it clear that they will hold AI agents, chatbots and other resources responsible for compliance (just as they would humans).
UDAAP Enforcement Risks
The prohibition against Unfair, Deceptive, or Abusive Acts or Practices has provided regulators with broad authority to challenge AI deployments that produce consumer harm. Institutions that fail to disclose AI-driven decision-making processes may face claims alleging that such opacity constitutes abusive practices. Models that generated unfair outcomes—defined as substantial consumer injury not reasonably avoidable and not outweighed by countervailing benefits—created regulatory vulnerability even in the absence of intentional discrimination.
Explainability emerged as a critical compliance need. The “black box” nature of advanced machine learning models made it difficult for institutions to provide the human-readable explanations increasingly demanded by state regulators and consumer advocates, exposing them to UDAAP scrutiny.
Strategic and Practical Implications
The surprising lesson of 2025 was that technological capability divorced from regulatory guardrails created compliance risk rather than competitive advantage. Institutions that rush to deploy AI without robust governance frameworks may find themselves exposed to enforcement actions, reputational damage, and class action litigation.
Effective risk management in this environment may require five critical measures. First, regular algorithmic audits to detect and remediate bias in credit scoring and underwriting models ensure ongoing FCRA compliance. Second, implementation of explainable AI frameworks—technical approaches that make algorithmic decision-making interpretable to non-specialists—enhances transparency and mitigates UDAAP exposure. Third, maintaining human-in-the-loop oversight for high-risk credit decisions and collections activities provides a compliance safeguard against automated errors and abusive practices. Fourth, updated compliance training that incorporates AI governance and ethical considerations may equip personnel to identify and address algorithmic risks. Fifth, proactive engagement with regulators to align AI practices with evolving legal standards positions institutions to anticipate rather than react to enforcement priorities.
The Path Forward
As 2025 concluded, the consumer finance, credit and collections industries confronted a paradox: deregulation had enabled technological advancement that improved operational efficiency while simultaneously creating legal vulnerabilities that threatened institutional stability. The administration’s light-touch regulatory approach accelerated AI adoption and discouraged states from creating bottlenecks, leaving open critical questions regarding fairness, transparency, and consumer protection unanswered.
The defining surprise of 2025 was the recognition that in our industry, innovation without governance generates risk. Institutions that treat AI deployment as purely a technological initiative, rather than a compliance and ethical imperative, may discover that algorithmic efficiency cannot compensate for regulatory exposure. The year demonstrated that, in a deregulated environment, competitive advantage belongs not to the fastest AI adopters but to organizations that embed legal and ethical safeguards into algorithmic systems from inception. The 2025 experience established that sustainable AI transformation in consumer finance requires balancing innovation with accountability—a lesson that may shape industry practices and regulatory responses for years to come.
