Business Processes Redefined LLC
A published position

AI in collections, measured and named.

The accounts receivable industry is about to make a lot of irresponsible AI claims. We'd rather tell you exactly where AI is and isn't used in our operation, and the regulatory lines we hold. This page is for the buyer whose legal team is nervous about every "AI-powered" sales deck on the market.

The position, in one paragraph

What we believe about AI in accounts receivable.

AI is useful for back-office work in our operation and unsuitable for unsupervised customer-facing decisions in this regulatory environment. We use it where the work is internal, the output is reviewed, and the regulatory exposure is governed. We do not use it to make collection decisions, generate customer-facing communications autonomously, or replace the human judgment that sits behind every consumer-facing action. When that calculus shifts, whether because the regulation evolves or the technology matures, we will say so publicly before we say so commercially.

At a glance

Where AI is, and where it isn't.

Where AI is used

Internal, reviewed, governed.

  • Operational analysis on our own data
  • Document drafting and internal tooling
  • Pattern detection in operational cycles
Where it isn't

Lines we hold.

  • Customer-facing communications
  • Collection decisions on individual accounts
  • AI voice agents on customer calls
  • Regulated data outside controlled environments
Where we use it

Internal applications, with named human review.

01

Operational analysis on our own data.

Portfolio segmentation, reconciliation discrepancy review, placement quality checks, and internal reporting summaries draw on AI-assisted analysis. Every output is reviewed by a named human accountable for the underlying work before it informs a placement or client-facing decision.

02

Document drafting and operational tooling.

Internal standard operating procedures (SOPs), training materials, agency-facing instructions, and reporting templates are drafted with AI assistance and reviewed by the operations lead accountable for the document. The artifact is a BPR document; the AI is a tool.

03

Pattern detection in operational cycles.

Reconciliation review, dispute log review, and complaint pattern analysis benefit from AI-assisted pattern detection. The detection is input to a human decision, not a decision itself.

Where we do not

Lines we hold, until the regulation and technology say otherwise.

A

No AI-generated customer-facing communications.

Every customer-facing communication (call script, email, letter, response to a dispute) is authored by a person and reviewed before deployment. AI drafting is not used to generate consumer-facing language autonomously. The regulatory exposure under UDAAP and FDCPA is too sharp, and the buyer's brand risk is too real.

B

No AI-driven collection decisions on individual accounts.

AI is not used to decide which accounts to call, when to call them, what to offer, when to escalate, or when to close. Those are human decisions governed by the placement instructions and the documented operating procedures. The audit trail says "human, then process," not "model output."

C

No AI voice agents on customer calls.

We do not deploy AI voice agents that take inbound or outbound consumer calls without human disclosure and consent under applicable state law. The TCPA and state-level consent regimes are sharp, and even where deployment would be technically permissible, it conflicts with brand-protective recovery as a principle.

D

No AI handling of regulated data outside controlled environments.

Customer personally identifiable information (PII), financial data covered by the Gramm-Leach-Bliley Act (GLBA), and HIPAA-adjacent information (where applicable) are not processed by AI systems outside controlled environments with documented data-handling protocols. When in doubt, the answer is no.

How we'll communicate change

This position is dated. When it moves, we'll tell you.

The regulatory perimeter on AI in consumer finance is changing. The CFPB has issued guidance on AI in credit and collections; states are moving on disclosure requirements for AI-generated communications; the FTC has signaled scrutiny of "AI washing." When our operating posture shifts in response to a regulatory development or a maturing technology, the change will be published on this page with an effective date and a description of what changed. Existing clients will be notified before the change reaches material engagements.

Current version: v1.0 · effective May 2026

The accounts receivable industry is about to make a lot of irresponsible AI claims.

Get in touch

If your legal team has questions, ask them.

We'd rather have the conversation in writing, with the people accountable for the answer in the room.

Start a conversation →