{"stable_id":"8574a74693efd62c","slug":"td-mortgage-pre-adjudication","company":"TD Bank Group","workflow_name":"Mortgage and HELOC pre-adjudication and application memo preparation","function_code":"credit","pattern_codes":["step_before_decision","deterministic_control"],"changed_assumption":"That preparing a mortgage or HELOC file for a credit decision requires a human to read the document package end to end - classifying documents, extracting figures, calculating income, checking consents and writing the summary - because borrower documents are too variable for a rules engine to handle reliably.","evidence_strength":"mixed","publication_tier":"showcase","freshness":"current","reviewed_at":"2026-08-23","updated_at":"2026-08-22","source_quality_summary":"1 independent, 1 primary; publication outcomes are verified and reported.","caveat_summary":"TD explicitly labels these 'Early results', and that hedge must travel with the figure. Both the baseline and the result are TD's own averages; TD does not publish the measurement definition, the sample size, or the variance. American Banker restates the figures rather than measuring them independently, though it adds independent expert commentary. The treated population size is undisclosed, so the aggregate hours saved cannot be derived. The metric covers pre-adjudication only - not application-to-funding cycle time - and TD states this is 'an impactful but small chunk of the overall journey towards funding a mortgage'. Directional and entirely unquantified. TD asserts improved accuracy and reduced unit cost without publishing a baseline error rate, a post-deployment error rate, or any co","freshness_caveat":null,"workflow_summary":{"before":[{"actor":"Borrower","action":"Submits the purchase agreement, ID, account statements, and proof of income.","actor_type":"control"},{"actor":"Human reviewer / mortgage operations analyst","action":"Classifies documents, finds relevant figures, verifies details, calculates income, and writes the summary.","actor_type":"human"},{"actor":"Credit adjudicator","action":"Reviews the human-prepared summary and makes the lending decision.","actor_type":"control"},{"actor":"Borrower","action":"Waits for the pre-adjudication result, which took an average of 15 hours of processing.","actor_type":"control"}],"hinge":"Preparing a decision file does not require a human to read every page.","after":[{"actor":"Borrower","action":"Submits the same document package.","actor_type":"system"},{"actor":"Autonomous AI agents built by TD's Layer 6","action":"Classifies documents, extracts key data, calculates income, checks consent, finds discrepancies, and drafts a summary.","actor_type":"ai"},{"actor":"Credit adjudicator","action":"Receives the AI-generated summary memo and makes the lending decision. The decision right is unchanged.","actor_type":"system"},{"actor":"Trustworthy AI team (led from Layer 6)","action":"Tests privacy, security, fairness, accountability, explainability; monitors performance, accuracy, answer relevance.","actor_type":"human"}],"decision_mode":"retained","decision_marker":"AI prepares; the human adjudicator decides"},"before":[{"order":1,"actor":"Borrower","action":"Submits the purchase agreement, ID, account statements, and proof of income.","handoff_to":"Human document reviewer","control":"Document checklist; incomplete packages returned."},{"order":2,"actor":"Human reviewer / mortgage operations analyst","action":"Classifies documents, finds relevant figures, verifies details, calculates income, and writes the summary.","handoff_to":"Credit adjudicator","control":"Manual verification against policy requirements; consent checks performed by hand."},{"order":3,"actor":"Credit adjudicator","action":"Reviews the human-prepared summary and makes the lending decision.","handoff_to":"Borrower / funding process","control":"Credit policy and adjudication authority."},{"order":4,"actor":"Borrower","action":"Waits for the pre-adjudication result, which took an average of 15 hours of processing.","handoff_to":"","control":"Service-level expectations only."}],"after":[{"order":1,"actor":"Borrower","action":"Submits the same document package.","handoff_to":"Autonomous AI agents","control":"Document checklist unchanged."},{"order":2,"actor":"Autonomous AI agents built by TD's Layer 6","action":"Classifies documents, extracts key data, calculates income, checks consent, finds discrepancies, and drafts a summary.","handoff_to":"Credit adjudicator","control":"Deterministic tools (calculator or spreadsheet) are mandated for arithmetic rather than the language model; TD-specific term and acronym definitions supplied; guardrails restrict the model to an approved tool library."},{"order":3,"actor":"Credit adjudicator","action":"Receives the AI-generated summary memo and makes the lending decision. The decision right is unchanged.","handoff_to":"Borrower / funding process","control":"Credit policy and adjudication authority retained by the human; an independent commentator describes the design as one where 'a human still needs to be accountable for the final decision'."},{"order":4,"actor":"Trustworthy AI team (led from Layer 6)","action":"Tests privacy, security, fairness, accountability, explainability; monitors performance, accuracy, answer relevance.","handoff_to":"Model release / remediation backlog","control":"Rigorous pre-deployment check series; post-deployment monitoring with defects routed to future releases."}],"decision_rights":"The lending decision right remains with the human credit adjudicator. American Banker states plainly that the agent 'handles these steps and presents the results to a credit adjudicator to make a decision'. The agent's own authority is bounded in two published ways: it must route arithmetic to deterministic tools rather than compute it, and it may only use an approved library of tools ('here is the library of things that you can use, here is what they are for'). Validation is against 'select policy requirements' - TD does not disclose which policy checks are in scope, so the boundary of the agent's validation authority is only partly public.","exception_path":"Not published as a discrete exception workflow. The structural exception path is the human adjudicator, who receives the memo and decides; TD chose the use case specifically because it has 'clear inputs, clear outputs, and a human in the loop'. Post-deployment, the Trustworthy AI team monitors models and routes observed mistakes to future releases. TD does not disclose what happens to an individual file when the agent cannot classify a document, cannot compute income, or flags a discrepancy it cannot resolve.","removed_work":["Manual classification of the borrower's submitted document package","Reading and searching documents of up to 100 pages to locate specific figures","Manual extraction and transcription of key application data","Manual income calculation and annualisation from pay stubs and rental income","Manual consent checks","Manual inconsistency and discrepancy searching across the package","Hand-writing the application summary memo for the adjudicator"],"outcomes":[{"metric":"Pre-adjudication processing time per mortgage or HELOC application (document review through summary memo generation)","baseline":"An average of 15 hours","result":"An average of less than three minutes","period":"Early results from production deployment in January 2026, disclosed 2026-05-21 and reported 2026-05-28 - an observation window of approximately four months. TD does not state the measurement window explicitly.","scale":"TD Canada Real Estate Secured Lending mortgage and HELOC pre-adjudication, in production. The treated application count is not disclosed. For context on the workflow's population, TD states it underwrites hundreds of thousands of mortgages annually and recorded $826 million in US residential mortgage origination volume in 2025 - both are book-size figures, not the treated cohort.","attribution_caveat":"TD explicitly labels these 'Early results', and that hedge must travel with the figure. Both the baseline and the result are TD's own averages; TD does not publish the measurement definition, the sample size, or the variance. American Banker restates the figures rather than measuring them independently, though it adds independent expert commentary. The treated population size is undisclosed, so the aggregate hours saved cannot be derived. The metric covers pre-adjudication only - not application-to-funding cycle time - and TD states this is 'an impactful but small chunk of the overall journey towards funding a mortgage'.","evidence_label":"reported"},{"metric":"Accuracy and cost of the pre-adjudication output","baseline":"Accuracy and unit cost of the previous manual human review, neither quantified by TD","result":"TD states the agent provides 'more accurate results and thereby reducing both risk and the unit cost on adjudication'; the mortgage officers 'get a more complete package, they get a more accurate summary at the very end'","period":"Early results from the January 2026 production deployment, disclosed 2026-05-21","scale":"TD Canada RESL mortgage and HELOC pre-adjudication; no quantified accuracy delta, error rate or unit-cost figure disclosed","attribution_caveat":"Directional and entirely unquantified. TD asserts improved accuracy and reduced unit cost without publishing a baseline error rate, a post-deployment error rate, or any cost figure. This outcome is recorded because the accuracy claim is central to TD's own framing, but it cannot support a numeric claim and should not be presented as a measured result. An independent expert quoted by American Banker separately notes that human verification of AI output carries its own burden and that reviewers 'can get bored and fail to check everything carefully' - a risk TD does not address publicly.","evidence_label":"reported"}],"executive_lesson":"TD picked the step before the decision, not the decision. The agent does the document work - classify, extract, calculate, check consents, find discrepancies, draft the memo - and the credit adjudicator still owns the lending call. That boundary is why this shipped in a regulated process. Two design choices are worth copying regardless of industry. First, TD refused to let the language model do arithmetic and forced it to call a deterministic calculator, turning a well-known failure mode into an explicit control rather than a residual risk. Second, TD chose a first use case with clear inputs, clear outputs and a human already in the loop, and said so - the value was real but the point was to build an operating model it could reuse. Note what TD has not published: the number of applications treated, any accuracy figure, and where the freed reviewer capacity went. The 15-hour-to-3-minute figure is a per-application average labelled 'early results', and it cannot be multiplied into a programme benefit.","anti_pattern":"Multiplying a per-application average by the institution's total book. TD discloses 15 hours to under three minutes per application and, separately, that it underwrites hundreds of thousands of mortgages annually - but the treated application count is not published, and those two numbers must not be combined into an aggregate saving. The second anti-pattern is treating the human-in-the-loop as a control without funding it: an independent expert quoted by American Banker warns that verification can take longer than expected and that reviewers 'can get bored and fail to check everything carefully'. An organisation that automates preparation, keeps a nominal human approver, and then measures only cycle time has moved the risk rather than managed it.","questions_for_leaders":["In our highest-volume document workflow, which step precedes the decision - and could we automate only that step without touching the decision right?","Where are we letting a language model perform precision-critical operations it is structurally bad at, and could we force a deterministic tool call instead?","If we published a per-unit improvement, would we also be able to publish the treated population, or would the number be unusable?","Who verifies the agent's output, how long does that verification actually take, and have we measured whether verification quality degrades over time?","Which policy checks are we comfortable letting an agent validate, and which must stay human - and is that boundary written down?","Is our first agentic use case chosen to prove an operating model we can reuse, or to chase the largest available benefit?"],"collections":["human-still-decides","regulated-autonomy","embodied-work","negative-results"],"bundle_version":"1.0.0","bundle_fingerprint":"sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93","canonical_url":"https://brianletort.ai/transformations/td-mortgage-pre-adjudication"}