Credit Analysts
13-2041.00Median wage $83,510/yr64,390 employed (US)Rank #36 of 923 scored · top 4% by substitution
Analyze credit data and financial statements of individuals or firms to determine the degree of risk involved in extending credit or lending money. Prepare reports with credit information for use in decisionmaking.
Sub-scores
0–100 · band = confidence interval from rater disagreement
Substitution61
Exposure61
Augmentation86
Substitution — the headline: capability discounted by cost, barriers and adoption.
Exposure — technical capability alone, regardless of whether anyone deploys it.
Augmentation — how much AI assists without replacing. High here + moderate substitution = a changing job, not a disappearing one.
Tasks on the substitution scale
11 rated tasks, binned by substitution score.
Position among all scored occupations
Distribution of 923 occupation scores; the marker is this occupation.
Tasks with substitution ≥ 70
18%
Run 1.0.0-draft.1 · computed 2026-08-05 · rater panel: claude-sonnet-5, claude-haiku-4-5-20251001 · intervals span rater disagreement.
Why this score
The five weighted dimensions of the composite, averaged across this occupation's tasks (importance-weighted, panel mean). Exact weights and formulas: /api/v1/methodology.
Task automatabilityw 35%62
panel mean rating 3.5/5 → substitution pressure 62/100
Technical feasibility todayw 20%58
panel mean rating 3.3/5 → substitution pressure 58/100
Cost vs. human wagew 15%73
panel mean rating 3.9/5 → substitution pressure 73/100
Adoption barriersw 20%inverted — strong barriers lower the score49
panel mean rating 3.1/5 (barrier strength) → substitution pressure 49/100
Sector adoption velocityw 10%69
panel mean rating 3.8/5 → substitution pressure 69/100
Task breakdown (11 tasks)
Substitution pressure per task, weighted by O*NET importance in the composite. Expand a task for the full rater audit trail — every rating, every model, every rationale.
Generate financial ratios, using computer programs, to evaluate customers' financial status.
85CI 75–95 · exposure 87 · augmentation 88 · importance 4.5/5 · click for rater detail
Generate financial ratios, using computer programs, to evaluate customers' financial status.
85| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Banking and lending sectors have already adopted automated ratio generation and financial analytics platforms at scale for decades; this is a standard, normalized practice across consumer and commercial credit. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Placeholder |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI tools assist credit analysts by automating ratio computation, freeing time for interpretation, judgment, and deeper analysis; the human analyst remains in the loop for evaluating what the ratios mean for creditworthiness and risk. |
| Augmentation potential | claude-sonnet-5 | 5/5 | Placeholder |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | Financial ratio calculation from structured data is purely computational and rule-based; modern AI systems and existing financial software can generate standard ratios (liquidity, profitability, leverage, efficiency) end-to-end with 100% accuracy and far >50% time savings versus manual calculation. |
| Task automatability | claude-sonnet-5 | 4/5 | Ratio computation from financial statements is formulaic and easily performed by software or AI agents that parse data and compute standard metrics, saving most of the manual time., |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | While credit decisions must be documented and defensible, the ratio generation step itself has minimal regulatory mandate that a human perform it; oversight and interpretation of ratios remain human responsibilities, but the generation task faces no hard licensing barrier. |
| Adoption barriers | claude-sonnet-5 | 2/5 | Placeholder |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference cost for ratio extraction and calculation is negligible (fractions of a cent per analysis), while a credit analyst's loaded wage for this work is substantial; the cost ratio favors automation by multiple orders of magnitude. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Placeholder |
| Technical feasibility today | claude-haiku-4-5-20251001 | 5/5 | Mature products (accounting software, Bloomberg terminals, specialized credit analysis platforms) already perform this reliably at scale in production; ratio generation is a solved, widely deployed task in banking and lending organizations. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Placeholder |
Review individual or commercial customer files to identify and select delinquent accounts for collection.
77CI 75–79 · exposure 75 · augmentation 88 · importance 3.3/5 · click for rater detail
Review individual or commercial customer files to identify and select delinquent accounts for collection.
77| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Financial services, banking, and consumer lending sectors have adopted automated delinquency detection and account selection widely; major banks and collection agencies routinely use rule-based and ML-driven systems for this task at scale. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Financial services is a fast-adopting sector for automation in credit risk and collections workflows, with many institutions already using automated delinquency scoring in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI significantly assists human analysts by pre-filtering and prioritizing delinquent accounts, generating risk scores, and flagging priority cases, allowing analysts to focus on exceptions, customer contact strategy, and dispute resolution rather than file review. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI-driven prioritization and flagging significantly speeds up analysts' ability to identify and triage delinquent accounts, letting them focus on judgment-intensive cases. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can reliably identify delinquent accounts by analyzing payment history, account status, and predefined delinquency criteria with high accuracy and significant time savings over manual file review. The task is primarily pattern-matching against structured data (payment dates, amounts, account flags), which modern ML excels at. |
| Task automatability | claude-sonnet-5 | 4/5 | Screening customer files for delinquency indicators against defined criteria (days past due, balance thresholds) is a structured, rules-based data task that current AI/automation systems handle well, though final selection judgment may still need review. |
| Adoption barriers | claude-haiku-4-5-20251001 | 2/5 | Minimal legal barriers exist to automating the *selection* of delinquent accounts; however, some regulatory oversight of collection practices and organizational preference to have humans review edge cases or customer disputes provide modest friction to full substitution. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human perform this specific screening step, though collections practices are regulated (FDCPA) so oversight and audit trails are expected. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI-driven account flagging costs pennies per account once deployed, versus hours of human analyst time at loaded wages of $50–80/hour; the cost difference is at least an order of magnitude in favor of automation. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated rule engines and ML scoring can process thousands of accounts at a fraction of the cost of manual file review, though initial integration with legacy core banking systems adds cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products from fintech and collections platforms already perform automated delinquency identification and account selection in production environments; systems routinely flag accounts meeting delinquency criteria with low error rates in banking and lending sectors. |
| Technical feasibility today | claude-sonnet-5 | 4/5 | Collections and risk management software already automates delinquency flagging and account triage in production at banks and lenders, though edge cases and dispute handling still require analyst oversight. |
Analyze credit data and financial statements to determine the degree of risk involved in extending credit or lending money.
65CI 56–74 · exposure 62 · augmentation 100 · importance 4.9/5 · click for rater detail
Analyze credit data and financial statements to determine the degree of risk involved in extending credit or lending money.
65| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Banking, credit unions, and fintech companies are actively deploying AI-driven credit scoring and risk assessment tools; major lenders use algorithmic systems to pre-screen applications and flag risks. Adoption is measurable and deepening in the financial services sector. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Financial services is a fast-adopting sector for AI/ML in credit scoring and underwriting, with widespread production deployment particularly in retail and SME lending. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI excels at augmenting credit analysts by automating data extraction, calculating comprehensive risk metrics, and highlighting anomalies, allowing analysts to focus on judgment and relationship context. Human-AI collaboration is the dominant production model in credit analysis today. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI substantially augments credit analysts by automating data aggregation, ratio calculation, and preliminary risk flagging, letting analysts focus on judgment-intensive aspects of the decision. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI can extract and analyze financial metrics, calculate risk ratios, and flag anomalies from statements with high consistency, achieving substantial time savings on data processing and preliminary risk scoring. However, judgment on context-specific risk factors (relationship history, industry outlook, management quality) often requires human discretion, preventing full end-to-end automation at equal quality. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can extract and analyze financial statement data and flag risk factors, but final risk judgment involving qualitative context, negotiation, and nonstandard data still requires human synthesis, so only partial time savings are achievable end-to-end. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Credit decisions carry regulatory oversight and liability exposure (Fair Lending Act, Truth in Lending Act compliance), and many institutions maintain human review and sign-off requirements for high-stakes loans. However, no hard legal mandate requires a licensed analyst to perform the entire task—AI-assisted workflows are already adopted. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Lending decisions face regulatory scrutiny (fair lending laws, model risk management, ECOA) requiring explainability and human oversight, creating moderate but not absolute barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference costs for document parsing, ratio calculation, and risk scoring are orders of magnitude cheaper than the fully-loaded cost of a credit analyst reviewing statements and generating analysis, while reducing human review time significantly. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated data extraction and statistical risk models are far cheaper per application than analyst hours, especially for high-volume standardized credit decisions, though complex corporate credit still needs costlier human review. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products (e.g., credit scoring engines, financial analysis platforms) reliably extract and quantify financial metrics, analyze historical patterns, and generate risk assessments used in production by lenders. Some gaps remain in qualitative judgment and novel scenarios, but the core analytical workflow is demonstrably operational at scale. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Deployed credit-scoring and underwriting AI tools are used in production at banks and fintechs, but they typically handle standardized consumer/small-business credit rather than full commercial credit analysis with judgment calls. |
Prepare reports that include the degree of risk involved in extending credit or lending money.
64CI 57–70 · exposure 70 · augmentation 88 · importance 4.5/5 · click for rater detail
Prepare reports that include the degree of risk involved in extending credit or lending money.
64| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Financial services sector has rapidly adopted AI/ML for credit risk assessment and reporting over the past 5–7 years, with major lenders deploying algorithmic risk models in production and using automated reporting. Adoption is faster and deeper in large institutions than regional banks, but trend is clearly accelerating. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Banking and financial services are adopting AI/ML tools for credit analysis and report generation at a moderate pace, with pilots and partial production use common but full-scale autonomous deployment still limited. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI substantially assists credit analysts by automating data gathering, calculating risk scores, flagging anomalies, and drafting report sections, allowing analysts to focus on judgment calls and complex cases. This augmentation demonstrably raises analyst productivity and report quality when humans remain in the loop. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI substantially augments credit analysts by rapidly drafting risk narratives, flagging anomalies, and summarizing financial data, letting analysts focus on judgment and final sign-off. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can now extract financial data, calculate risk metrics, and generate draft risk assessment reports with minimal human intervention, meeting or exceeding 50% time savings at comparable quality. However, final judgment on credit risk typically requires human review of nuanced factors, preventing full end-to-end automation. |
| Task automatability | claude-sonnet-5 | 4/5 | Generating credit risk reports involves synthesizing financial statements, ratios, and market data into structured narratives, which LLMs can draft effectively when given structured inputs, saving significant analyst time on the writing/synthesis portion.“}, |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory oversight and compliance requirements (Basel III, Dodd-Frank, fair lending laws) mandate that credit decisions and risk assessments be traceable and defensible, requiring human sign-off and explainability. Liability asymmetry—errors in lending decisions carry high legal and financial costs—creates friction against full substitution. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Credit decisions often require documented human sign-off for regulatory and fair-lending compliance (e.g., ECOA, SR 11-7 model risk guidance), creating moderate but not absolute barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-driven credit risk analysis and report generation costs a fraction of a full-time analyst's loaded wage, particularly when processing high volumes. Inference and integration costs are low relative to the analytical labor replaced, yielding significant cost advantage. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Once integrated with data feeds, AI-generated risk narratives cost a small fraction of analyst hourly rates for the drafting portion, though data integration and oversight add some cost. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Multiple deployed products (e.g., credit scoring platforms, risk analytics software, LLM-powered report generation) reliably perform substantial portions of risk report preparation in production banking environments. Some edge cases and complex scenarios still require human review, but core functionality is mature and widely deployed. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Deployed fintech and banking AI tools (e.g., in credit scoring and report drafting platforms) exist and are used in production, but human review and judgment calls on risk classification remain standard, so full autonomous reliability is not yet achieved. |
Compare liquidity, profitability, and credit histories of establishments being evaluated with those of similar establishments in the same industries and geographic locations.
62CI 50–75 · exposure 62 · augmentation 88 · importance 4.3/5 · click for rater detail
Compare liquidity, profitability, and credit histories of establishments being evaluated with those of similar establishments in the same industries and geographic locations.
62| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Financial services, banking, and credit assessment are digitized sectors with strong AI adoption. Many lenders and credit bureaus are deploying AI-assisted comparative analysis; adoption is moving from pilots to production, particularly in retail and SME lending. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Financial services broadly show strong AI adoption, but credit analysis specifically remains at the pilot-to-partial-production stage, with many institutions still using traditional spreadsheet-based comparative methods. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI excels at augmenting credit analysts by rapidly surfacing peer comparisons, flagging outliers, and organizing disparate data sources; analysts retain judgment on weighting, context, and exceptions. This is a high-value augmentation scenario where AI handles data work while the human focuses on interpretation and risk judgment. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools significantly speed up data gathering, ratio calculations, and drafting of comparative narratives, letting analysts focus judgment on interpreting results and edge cases, meaningfully boosting productivity while keeping humans in the loop. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI can efficiently gather, organize, and compare financial metrics (liquidity ratios, profitability measures, credit histories) across comparable firms using structured data sources. While some judgment about industry comparables and contextual adjustments remains, the core comparative analysis can be performed with >50% time savings using current financial data APIs and LLM-based summarization tools. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can readily pull financial data and generate comparative ratio analyses, but sourcing accurate comparable industry/geographic benchmarks and contextualizing anomalies still often requires human judgment and verification, limiting full end-to-end automation today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Credit decisions carry regulatory and reputational stakes, and most institutions require human sign-off on credit recommendations. However, the comparative analysis itself (rather than the final decision) faces modest friction; compliance and audit oversight are common but not absolute legal requirements for the analysis step. |
| Adoption barriers | claude-sonnet-5 | 2/5 | No licensing requirement mandates a human specifically perform comparative benchmarking, though credit decisions relying on this analysis may carry regulatory documentation and fair-lending scrutiny that create some friction. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI inference costs for data aggregation, ratio calculation, and comparative analysis are a small fraction of a credit analyst's loaded cost. Setup and oversight require human time, but per-analysis AI cost is substantially lower than hourly analyst wage for equivalent analytical output. |
| Cost vs. human wage | claude-sonnet-5 | 3/5 | Automated data aggregation and ratio computation is cheap, but curating reliable comparable-firm datasets and validating outputs still requires analyst oversight, keeping all-in costs only moderately below human-only costs. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed products in fintech and credit intelligence (e.g., Bloomberg terminals with AI features, Dun & Bradstreet's AI-enhanced reports, specialized credit analysis platforms) reliably perform comparative financial analysis at scale. Some edge cases and custom adjustments require human review, but the baseline comparative task is mature and production-proven. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Financial analysis and benchmarking tools (e.g., Moody's, S&P platforms, fintech underwriting software) exist and are used in production, but they still require analyst review and struggle with nonstandard or private company data, so reliability is mixed rather than uniformly high. |
Evaluate customer records and recommend payment plans, based on earnings, savings data, payment history, and purchase activity.
62CI 49–76 · exposure 62 · augmentation 75 · importance 3.7/5 · click for rater detail
Evaluate customer records and recommend payment plans, based on earnings, savings data, payment history, and purchase activity.
62| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Financial services sectors are leading adopters of AI for decisioning; automated credit evaluation is already mainstream in banking and fintech, with widespread displacement of routine analyst tasks documented in industry reports. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Financial services broadly show strong AI adoption, but credit risk and collections functions tend to move more cautiously due to regulatory scrutiny, placing this in the middle range. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI augmentation is already significant in practice: systems flag anomalies, surface comparable accounts, and generate preliminary recommendations that analysts refine, substantially raising throughput and consistency of human review. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools significantly speed up data aggregation, pattern detection in payment history, and draft plan generation, meaningfully boosting analyst productivity while they retain final judgment. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI systems can analyze structured financial data (earnings, savings, payment history, purchase activity) and apply rule-based or ML scoring models to recommend payment plans with high consistency. However, edge cases requiring judgment about personal circumstances or regulatory exceptions may still require human review, preventing a full 5. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can analyze structured financial data and generate payment plan recommendations, but integrating disparate data sources and handling edge cases still requires human judgment and setup, so only partial end-to-end automation meets the 50% bar today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Credit decisions face substantial regulatory barriers (Fair Lending Act, FCRA, ECOA compliance) and audit/liability requirements that typically mandate human review, documentation, and sign-off on material decisions, creating mandatory human-in-the-loop friction. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Lending decisions are subject to fair lending regulations (e.g., ECOA, FCRA) requiring explainability and human oversight, creating moderate compliance friction though not requiring a licensed professional sign-off in all cases. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | AI inference on structured financial records is extremely cheap per evaluation—often fractions of a cent—compared to a human credit analyst's loaded cost of $40–80+ per hour. The ratio easily favors AI by an order of magnitude for high-volume operations. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated data analysis and rule/ML-based recommendation engines are substantially cheaper per case than analyst hours once built, though initial integration and compliance costs are non-trivial. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Credit scoring and decisioning systems are deployed at scale in financial institutions today, with demonstrated performance on payment plan recommendations. Most major banks use automated systems for routine cases, though some regulatory oversight and manual review remain standard practice. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Credit scoring and underwriting products (e.g., automated decisioning platforms) exist and are used in production, but recommending nuanced payment plans still often involves human review, especially for non-standard cases. |
Analyze financial data, such as income growth, quality of management, and market share to determine expected profitability of loans.
61CI 51–70 · exposure 62 · augmentation 100 · importance 4.5/5 · click for rater detail
Analyze financial data, such as income growth, quality of management, and market share to determine expected profitability of loans.
61| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Banks and fintech lenders have rapidly deployed automated underwriting and risk-scoring systems; major institutions use AI-driven pipelines for initial screening and decisioning, making this a high-adoption sector with measurable displacement of routine analysis. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Banking and financial services are among the faster-adopting sectors for AI-driven credit risk and underwriting tools, with substantial production deployment of automated scoring and analytics. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI excels at surfacing patterns in financial data, generating comparative analyses, and flagging anomalies, allowing analysts to focus on judgment calls and complex cases. This is a textbook augmentation scenario where AI productivity gains are demonstrable while humans retain decision authority. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI substantially accelerates data gathering, ratio calculation, trend analysis, and report drafting for credit analysts, letting them focus on judgment calls while dramatically speeding the overall workflow. |
| Task automatability | claude-haiku-4-5-20251001 | 4/5 | AI can extract and analyze structured financial data (income, market share, ratios) from documents and databases, and apply predictive models to estimate loan profitability with high consistency. However, qualitative assessment of management quality and subtle contextual factors still require human judgment, preventing full end-to-end automation at the ≥50% threshold. |
| Task automatability | claude-sonnet-5 | 3/5 | AI can process financial statements and generate profitability projections and ratio analysis quickly, but 'quality of management' assessment and nuanced judgment on expected profitability still require human synthesis and accountability, so only partial automation meets the 50% bar today. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory frameworks (Basel III, fair lending rules, consumer protection) mandate documented underwriting decisions and human accountability; many institutions require a licensed analyst or compliance review before loan approval, creating legal and organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Lending decisions face regulatory scrutiny (fair lending laws, model risk management, SR 11-7 guidance) requiring human sign-off and explainability, creating moderate but not absolute barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | Inference costs for financial data extraction, ratio computation, and risk modeling are minimal compared to analyst salaries; integration and oversight add overhead, but the ratio still favors AI significantly—likely 5–10× cheaper per analysis when scaled. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated data extraction and financial modeling via AI is dramatically cheaper per data point than analyst hours, though oversight and validation costs reduce the gap somewhat from the theoretical maximum. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Mature ML systems and credit-scoring platforms (e.g., automated underwriting engines from major lenders) reliably analyze financial metrics and generate risk scores in production. Deployed products handle quantitative analysis well, though human review remains standard for complex cases. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Deployed fintech and banking tools (e.g., automated underwriting and credit scoring platforms) reliably compute financial ratios and risk metrics, but comprehensive analysis incorporating qualitative factors like management quality is still narrow and often requires human review in production. |
Consult with customers to resolve complaints and verify financial and credit transactions.
59CI 32–85 · exposure 58 · augmentation 75 · importance 3.1/5 · click for rater detail
Consult with customers to resolve complaints and verify financial and credit transactions.
59| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 5/5 | Financial services and banking sectors are among the fastest adopters of AI-driven customer service automation; major institutions have deployed chatbots and automated transaction verification systems at scale, with clear market momentum and measurable displacement of routine analyst interactions. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Financial services is a fast-adopting sector for AI tools generally, but complaint resolution specifically still sees mostly pilot-stage deployment with humans retained for judgment-heavy interactions. |
| Augmentation potential | claude-haiku-4-5-20251001 | 4/5 | AI assistants can substantially augment credit analysts by surfacing relevant transaction history, flagging suspicious patterns, and drafting complaint responses, allowing humans to focus on complex disputes and relationship management while staying in the loop. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools can quickly surface transaction histories, flag discrepancies, and draft responses, meaningfully speeding up the analyst's verification and complaint-handling workflow while the analyst still manages the customer relationship. |
| Task automatability | claude-haiku-4-5-20251001 | 5/5 | AI-powered chatbots and automated systems can now handle complaint resolution workflows, verify transactions via database queries, and respond to common credit inquiries with sufficient speed and accuracy to exceed the 50% time-saving threshold for many routine cases, especially when integrated with backend systems. |
| Task automatability | claude-sonnet-5 | 2/5 | This task combines relational conflict resolution with verification work; current AI can assist verification but cannot fully own customer complaint resolution requiring empathy, negotiation, and judgment calls end-to-end.' |
| Adoption barriers | claude-haiku-4-5-20251001 | 3/5 | Regulatory requirements around financial advice, customer contact preferences, and liability for complaint resolution create meaningful friction; however, these barriers are surmountable through proper AI oversight frameworks and regulatory compliance automation rather than legal prohibitions on automation itself. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Financial services carry regulatory disclosure and dispute-resolution obligations (e.g., FCRA-related processes) plus customer preference for human contact in disputes, creating moderate friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 5/5 | Inference and integration costs for chatbot-based complaint resolution and transaction verification are orders of magnitude cheaper than analyst labor, with minimal ongoing oversight needed for routine inquiries. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | AI can cut cost for routine verification steps, but complaint handling often requires escalation and human oversight, keeping blended cost closer to human levels once error correction and compliance review are factored in. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 4/5 | Deployed conversational AI and transaction verification systems are production-ready in banking; while some complex disputes still require human escalation, automated complaint triage and transaction verification are reliably performing at scale in major financial institutions today. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | Chatbots and verification tools exist in production for simple account queries, but complex complaint resolution involving credit disputes still routes to human analysts in most deployed systems. |
Complete loan applications, including credit analyses and summaries of loan requests, and submit to loan committees for approval.
57CI 49–65 · exposure 58 · augmentation 100 · importance 4.8/5 · click for rater detail
Complete loan applications, including credit analyses and summaries of loan requests, and submit to loan committees for approval.
57| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 4/5 | Financial services and banking sectors have rapidly deployed AI for document automation and risk scoring; major lenders and fintech platforms are actively adopting these workflows in production, with visible displacement of routine data-entry and initial-screening roles. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Banking and financial services are moderately fast adopters of AI for underwriting support, with many pilots and some production use, but heavily regulated core credit decisioning adoption remains cautious. |
| Augmentation potential | claude-haiku-4-5-20251001 | 5/5 | AI-generated summaries, risk flags, score predictions, and financial statement analysis materially accelerate analyst productivity while they retain final judgment; this is one of the clearest cases of human-in-the-loop augmentation in financial services. |
| Augmentation potential | claude-sonnet-5 | 5/5 | AI tools significantly speed up data gathering, financial statement analysis, and drafting of loan summaries, letting analysts focus on judgment and risk assessment while staying in the approval loop. |
| Task automatability | claude-haiku-4-5-20251001 | 3/5 | Current AI can automate significant portions of loan processing—data extraction, initial credit scoring, document verification, and summaries of financial statements—but loan committee submission and final approval recommendations require human judgment on borderline cases and retain meaningful oversight requirements that prevent full end-to-end automation at the 50% time-saving threshold. |
| Task automatability | claude-sonnet-5 | 4/5 | Extracting financial data, computing ratios, and drafting standardized credit summaries is largely templatable and AI can already do most of the analytical drafting; however, edge-case judgment and committee-ready synthesis still need human review, so it falls short of full automation. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Regulatory frameworks (Truth in Lending Act, Fair Credit Reporting Act, Basel III) impose accountability and audit trails that implicitly require human sign-off; many jurisdictions and lending institutions legally or contractually require a licensed or qualified human analyst to attest to credit analyses before committee submission. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Lending decisions face regulatory scrutiny (fair lending laws, model risk management) and require documented human accountability for approval decisions, creating moderate but not absolute barriers to full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 4/5 | AI-powered document processing and initial analysis costs are substantially lower than a credit analyst's loaded wage, though ongoing oversight and quality checks moderate the savings; all-in cost is likely 30–50% of human equivalent. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated data extraction, ratio calculation, and draft summary generation are far cheaper per unit than analyst hours, though integration and oversight costs prevent a full order-of-magnitude advantage in all cases. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 3/5 | Commercial loan processing platforms with AI components exist in production (document parsing, risk flagging), but reliable end-to-end credit analysis and committee-ready summaries require human review and correction; no fully autonomous system submits applications without material human verification. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Fintechs and banks deploy AI-assisted underwriting and document summarization tools in production, but full end-to-end loan packaging with credit judgment still typically requires human analyst sign-off, limiting reliability at scale. |
Contact customers to collect payments on delinquent accounts.
44CI 31–56 · exposure 38 · augmentation 63 · importance 4.0/5 · click for rater detail
Contact customers to collect payments on delinquent accounts.
44| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 3/5 | Financial services and debt-collection firms have piloted AI outreach (reminders, pre-collection) at scale, but human agents still dominate for actual collections. Adoption is uneven: some firms deploy chatbots for early-stage reminders, but production replacement of collectors remains limited. |
| Sector adoption velocity | claude-sonnet-5 | 4/5 | Financial services and collections are among the more digitized sectors with substantial investment in automated dunning, chatbots, and predictive contact strategies already in production. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist collectors by preparing account summaries, suggesting talking points, and handling low-risk reminders, improving agent productivity; however, the judgmental and relational core of collection work limits how much AI can transform the task while keeping humans in the loop. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI tools help prioritize accounts, draft communications, predict best contact times/channels, and flag risk, meaningfully boosting analyst productivity while humans handle sensitive negotiations. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | While AI can draft collection messages and schedule outreach, the task involves negotiation, empathy, and judgment about hardship that require human interaction. Current systems cannot reliably handle the full conversational, legal, and ethical complexity of collections work to achieve 50% time savings at equal quality. |
| Task automatability | claude-sonnet-5 | 3/5 | AI voice/chat agents and automated dunning systems can handle routine outreach and reminders, but escalation, negotiation, and judgment calls on delinquent accounts still often require human involvement, especially for larger or complex accounts. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Collections work is heavily regulated (FDCPA, TCPA, GDPR) with strict licensing, recording, and consent requirements; liability exposure for improper contact is high; and many customers prefer human interaction for dispute resolution or hardship claims, creating legal and organizational friction against full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | Debt collection is subject to regulations (e.g., FDCPA in the US, TCPA) governing communication methods and disclosures, creating compliance friction, though not requiring a licensed professional to perform the contact itself. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 3/5 | AI-driven outreach (chatbots, automated calling) has lower marginal cost per contact than human collectors, but integration, compliance monitoring, and human escalation overhead make the all-in cost roughly comparable to live collection agents. |
| Cost vs. human wage | claude-sonnet-5 | 4/5 | Automated outreach (SMS, email, voice bots) is dramatically cheaper per contact than human collectors, though oversight and escalation paths add some cost back in for complex cases. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 2/5 | Chatbots and IVR systems exist for basic payment reminders, but deployed systems struggle with nuanced collection scenarios, dispute handling, and regulatory compliance. Production deployments remain narrow and often require human escalation, falling short of reliable end-to-end task performance. |
| Technical feasibility today | claude-sonnet-5 | 3/5 | Collections automation platforms and AI-driven dialers/chatbots are deployed in production at scale in consumer collections, but reliability drops for negotiated settlements, disputes, and compliance-sensitive interactions. |
Confer with credit association and other business representatives to exchange credit information.
26CI 20–32 · exposure 20 · augmentation 63 · importance 3.1/5 · click for rater detail
Confer with credit association and other business representatives to exchange credit information.
26| Dimension | Model | Rating | Rationale |
|---|---|---|---|
| Sector adoption velocity | claude-haiku-4-5-20251001 | 2/5 | While financial services are digitizing, credit conferencing remains a high-touch, relationship-dependent activity with limited automation adoption in practice. Most implementations remain assistive rather than substitutive. |
| Sector adoption velocity | claude-sonnet-5 | 3/5 | Financial services broadly show fast AI adoption for analytics, but the specific interpersonal conferring aspect is still handled by humans, giving moderate overall velocity for this particular task. |
| Augmentation potential | claude-haiku-4-5-20251001 | 3/5 | AI can assist by preparing briefing materials, summarizing prior credit information, drafting follow-up emails, and flagging key risk factors, meaningfully improving analyst productivity without replacing human participation in the actual conversation. |
| Augmentation potential | claude-sonnet-5 | 4/5 | AI can help credit analysts prepare talking points, summarize prior credit histories, and draft correspondence, meaningfully boosting productivity even though the human still leads the exchange. |
| Task automatability | claude-haiku-4-5-20251001 | 2/5 | This task involves real-time negotiation and relationship management with external parties, requiring nuanced understanding of context, trust-building, and collaborative problem-solving that current AI cannot reliably do end-to-end. While AI can draft communications and summarize information, the actual conferencing and information exchange still depends on human judgment and interpersonal dynamics. |
| Task automatability | claude-sonnet-5 | 2/5 | This task involves interpersonal negotiation, relationship management, and real-time exchange of nuanced credit information with external parties, which current AI cannot fully replace though it can support parts like data lookup or summarization. |
| Adoption barriers | claude-haiku-4-5-20251001 | 4/5 | Credit conferencing typically requires a licensed or credentialed human representative (credit analyst or manager) to legally represent the organization and make binding commitments. Counterparties expect to negotiate with an authorized human, creating contractual and reputational barriers to full automation. |
| Adoption barriers | claude-sonnet-5 | 3/5 | While not strictly licensed, credit information exchange often involves confidentiality agreements, institutional trust, and compliance considerations that create meaningful friction against full automation. |
| Cost vs. human wage | claude-haiku-4-5-20251001 | 2/5 | AI tools can assist with preparation and follow-up documentation at low cost, but cannot replace the human credit analyst conducting the actual conference, so the cost of human labor remains dominant in the workflow. |
| Cost vs. human wage | claude-sonnet-5 | 2/5 | Because a human must still conduct the actual conferring and relationship-based exchange, AI mainly reduces prep time rather than replacing the labor, so cost savings are moderate at best. |
| Technical feasibility today | claude-haiku-4-5-20251001 | 1/5 | No deployed product reliably conducts independent credit conferences or negotiates with external business representatives at scale. AI can support communication drafting but cannot autonomously participate in or lead multi-party credit discussions requiring real-time judgment. |
| Technical feasibility today | claude-sonnet-5 | 2/5 | No deployed product autonomously confers with external business representatives to exchange credit information; existing tools support underlying data retrieval but not the interpersonal exchange itself. |
Related occupations — Business & Financial Operations
How to read this
A high substitution score does not mean this job disappears — it means a large share of its current tasks face replacement pressure, so the mix of tasks is likely to change. High augmentation alongside substitution typically means the occupation reorganizes around the protected tasks. Wide confidence intervals mean the rater panel disagreed: treat those scores as open questions, not verdicts.
What would change this score
New model capabilities (automatability, feasibility), falling inference costs (cost ratio), regulation and licensing shifts (barriers), and measured sector adoption (velocity) all re-enter at every index release. Each release is recomputed, versioned and kept queryable — scores are claims with a date on them, not permanent labels.