--- title: LLM and AI agent applications in finance · 2026-05 application surface overview aliases: - llm-agent-finance-application-overview-2026 - LLM finance applications 2026 - AI agent finance use cases 2026 - where LLMs touch finance 2026 - LLM banking insurance trading underwriting 2026 domain: agent-economy created: 2026-05-25 last_updated: 2026-07-29 last_tended: 2026-07-29 review_by: 2026-10-27 confidence: likely tags: [agent-economy, llm, ai-finance, overview, 2026-event, banking, insurance, trading, kyc-aml, fraud, robo-advisor, underwriting] status: active sources: - https://www.fsb.org/wp-content/uploads/P11112024.pdf - https://www.bis.org/publ/work1194.htm - https://www.imf.org/en/Publications/fintech-notes - https://www.federalreserve.gov/publications/financial-stability-report.htm - https://www.fsa.go.jp/en/news/2024/index.html - https://www.sec.gov/news/speech - https://www.fca.org.uk/publications - https://www.mas.gov.sg/news - https://www.bloomberg.com/professional/blog/bloomberggpt/ - https://www.jpmorgan.com/technology/news/indexgpt - https://www.goldmansachs.com/marquee/ - https://www.morganstanley.com/press-releases - https://www.anthropic.com/customers - https://openai.com/index/finance - https://docs.anthropic.com/ - https://platform.openai.com/docs/models - https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models - https://www.fsa.go.jp/en/news/2025/20250304/aidp.html - https://www.sec.gov/rules-regulations/2025/06/s7-12-23 --- # LLM and AI agent applications in finance · 2026-05 application surface overview ## TL;DR As of mid-2026, the LLM / AI-agent footprint in finance has bifurcated into **shipped production surfaces** (customer-facing chatbots, back-office automation, fraud / AML triage, developer copilots) and **constrained-pilot surfaces** (trading signal generation, credit underwriting decision support, advisory-grade recommendations). The first cluster has crossed the "default tooling" threshold at G-SIBs — Morgan Stanley AI @ Morgan Stanley, JPM IndexGPT / SpectrumGPT, Goldman GS AI Platform, BBVA / ING Anthropic deployments, Mizuho / SMBC / MUFG internal copilots. The second cluster remains gated by **FSB / BIS / IMF supervisory caution**, the **SEC's predictive-data-analytics rule trajectory**, **FCA AI-in-financial-services papers**, and Japan's **FSA AI principles 2024-2026** — all of which keep human-in-the-loop requirements on any decision that materially affects a customer or market. The maturity map by category (production / pilot / research) is the load-bearing routing map this entry encodes. See [[agent-economy/agent-actorship-debate|agent actorship debate]] for the personhood framing and [[agent-economy/agent-legal-tax-liability-framework|agent legal and tax liability framework]] for the deployer-liability default. ## Wiki route This entry sits under [[agent-economy/INDEX|agent-economy index]]. Read it against [[agent-economy/agent-legal-tax-liability-framework|agent legal and tax liability framework]] for the liability waterfall, [[agent-economy/agent-protocol-mainnet-adoption-2026|agent protocol mainnet adoption 2026]] for the underlying payment-rail readiness, and [[agent-economy/ai-driven-trading-regulation-japan-2026|AI-driven trading regulation Japan 2026]] for the trading-specific regulator stance. For developer tooling see [[agent-economy/claude-code-extension-architecture|Claude Code extension architecture]] and [[agent-economy/stripe-agent-toolkit-position|Stripe agent toolkit position]]. For the data-pipeline / signal angle see [[agent-economy/agent-driven-market-data-interpretation-pipeline|agent-driven market data interpretation pipeline]]. For custody and authorization composition see [[agent-economy/agent-custody-and-authorization-framework-2026|agent custody and authorization framework 2026]]. For identity see [[agent-economy/agent-identity-defi-and-traditional-finance-bridge|agent identity DeFi and traditional finance bridge]]. ## Seven application surfaces · maturity by category The 2026-05 maturity map across seven categories. **PROD** = at least one G-SIB / top-10-by-AUM operator running in real customer / regulatory traffic, **PILOT** = consortium or single-firm regulated pilot with public disclosure, **RESEARCH** = pre-pilot prototype with published papers but no production traffic. | Category | Publicly observed use class | Minimum evidence for a maturity claim | Core control questions | |---|---|---|---| | **(a) Customer-facing chat / assistance** | Retrieval, drafting, service triage and staff-assisted responses | Named institution, dated first-party disclosure, affected users, production / pilot label and human-handoff design | disclosure, hallucination handling, authentication, complaint / escalation, record retention | | **(b) Back-office automation** | Document extraction, KYC / AML support, compliance search and workflow drafting | Deployed workflow, decision boundary, data source, reviewer role and measured error / override data | access control, data lineage, model validation, reviewer independence and audit trail | | **(c) Trading / execution support** | Research summarization, event extraction, signal generation and execution support | Exact role in the order lifecycle, venue / asset scope, pre-trade limits and production evidence | market-abuse controls, latency, kill switch, model / code change control and replay | | **(d) Credit / underwriting support** | Document analysis, feature generation, decision support and explanation drafting | Whether AI affects approval / pricing, validation results, adverse-action process and applicable law | bias / fairness, explainability, data quality, override and appeal | | **(e) Fraud / financial-crime detection** | Risk scoring, alert triage, anomaly detection and investigation support | Production scope, false-positive / false-negative evaluation, customer-impact boundary and review process | drift, feedback loops, appeal, watchlist / data quality and model governance | | **(f) Advice / suitability support** | Research, portfolio explanation, adviser copilot and recommendation drafting | Licensed-entity disclosure, final decision maker, suitability / fiduciary controls and monitored outcomes | conflicts, suitability, disclosure, human approval and communication records | | **(g) Developer tooling** | Code completion, testing, review, documentation and operations support | Named tool, approved repositories / data classes, security review and production SDLC boundary | secrets / code leakage, dependency risk, review gates, testing and change management | Sources: ^[https://www.fsb.org/2024/11/the-financial-stability-implications-of-artificial-intelligence/] ^[https://www.bis.org/publ/work1194.htm] ^[https://www.fsa.go.jp/en/news/2025/20250304/aidp.html] **Reading the map**: the gap between PROD and PILOT correlates almost perfectly with whether the AI output is a **customer-facing financial decision** (PILOT) versus a **support / drafting / triage output reviewed by a licensed human** (PROD). Regulators have not blocked AI in finance — they have blocked AI from making the **final** customer-impacting decision without human sign-off. ## Per-category deep dive ### (a) Customer-facing chatbots **Production reality 2026-05**: Morgan Stanley's "AI @ Morgan Stanley Assistant" went live in 2023 for FA-facing use, expanded in 2024-2025 to FA-with-client surfaces, and by 2026-Q1 supports drafting client communications subject to FA review. BBVA + OpenAI partnership (announced 2025) covers customer-service triage in Spain and Mexico. ING + Anthropic deployed Claude for internal-facing knowledge retrieval that surfaces to call-center agents in real time. Mizuho's "M-AI Insight" and SMBC's "SMBC GAI" are deployed at scale internally but customer-facing surfaces remain gated through human agent. **What "PROD" actually means here**: the LLM drafts; a licensed human approves; the draft becomes the customer communication. The LLM does not directly answer the customer end-to-end for material questions (mortgage rates, account balances, advice). The exceptions are scripted FAQ (which has been ML / NLP-driven for a decade) — those have been quietly upgraded to LLM in 2024-2026 without re-architecting the consumer-protection envelope. **Regulator stance**: FCA's 2024 "AI in financial services" paper and the 2025-2026 consultation arc explicitly accept LLM-drafted customer comms with human review. Japan's FSA AI principles (2021, updated 2024) treat customer chatbot as a "support tool" with the financial institution retaining full Article 35 FIEA suitability obligation. The EU AI Act treats consumer-facing financial chatbot as "limited risk" (transparency-only) unless used for creditworthiness assessment (high-risk, see (d)). **Leading vendors**: Anthropic Claude (BBVA, ING, Mizuho), OpenAI GPT-4o / GPT-5 (Morgan Stanley, Bank of America initial pilot), Google Gemini (Citi pilot disclosed 2025), Cohere (BlackRock Aladdin Copilot adjacent), domestic Japan (NEC cotomi, NTT tsuzumi, PFN PLaMo). ### (b) Back-office automation · KYC / AML / compliance review **Production reality**: JPM's SpectrumGPT (compliance document review), HSBC's AI compliance assistant, Citi's compliance copilot, Nomura's ops AI, Mizuho's RM-AI for relationship-manager note drafting. The function set: KYC document extraction (passport / utility-bill / corporate-registry OCR + structured-data extraction), AML transaction-monitoring alert triage (LLM summarizes the alert context for human investigator), Suspicious Activity Report (SAR) drafting (LLM drafts, human compliance officer approves), sanctions-list adjacency review (LLM scores name matches against OFAC / EU sanctions list). **Why this category is PROD-mature**: KYC / AML output is consumed by **internal investigators**, not by customers. The deployer's compliance officer retains full regulatory accountability under [[fintech/aml-cft-fatf-grey-list-overview|FATF Recommendation 20]] and Japan's Act on Prevention of Transfer of Criminal Proceeds. The LLM accelerates the human's throughput rather than replacing the regulated decision. **Audit trail requirement**: every major LLM-augmented compliance system in production keeps a **per-decision prompt log** and **model-version stamp** so post-hoc review can reproduce the LLM's reasoning trace. See [[agent-economy/agent-driven-market-data-interpretation-pipeline|agent-driven market data interpretation pipeline]] for the analogous trail in trading. ### (c) Trading and execution · NLU news → trading signal / agent-driven hedging **Pilot status 2026-05**: Goldman Marquee + Marquee AI (Goldman's institutional analytics + AI overlay), JPM IndexGPT (LLM-driven thematic basket construction launched 2024-2025), BlackRock Aladdin Copilot (portfolio-manager-facing LLM, not customer-facing), Renaissance / Two Sigma / Citadel internal AI research (not publicly disclosed in detail). Bloomberg's BloombergGPT (2023 publication) and Bloomberg AI (productized in Terminal 2024-2026) provide finance-tuned LLM surfaces that buy-side firms layer their own logic on. **Why this category is PILOT not PROD**: market-impact risk. An LLM-driven trade that fat-fingers a $500M order can move a market. Regulators (SEC under Reg SCI / Reg SHO / new SAB 122, FCA under MAR / MIFID-II algo controls, FSA under FIEA Article 38-2 algo-trading rules, ESMA under MAR / MIFID-II RTS 6) require **pre-trade controls** (price collars, size limits) and **kill-switches** that no current LLM can self-attest to. The compromise: LLM generates **signal**; deterministic execution algo enforces the **risk controls**. The signal can be LLM-derived but the trade itself goes through the same algo control framework as a human-derived signal. **See [[agent-economy/ai-driven-trading-regulation-japan-2026|AI-driven trading regulation Japan 2026]]** for the regulatory deep dive. ### (d) Credit underwriting · LLM-augmented **Pilot status**: Upstart (FICO-supplement ML, now LLM-augmented document parsing 2024-2026), Pagaya (consumer-credit AI underwriting), Klarna (BNPL AI underwriting), Affirm (similar), Rakuten Card (Japan internal AI underwriting pilot), Mercari Credit (BNPL with AI). Use case: LLM reads unstructured documents (pay stubs, bank statements, gig-economy income records) and produces structured income / cash-flow features that feed the underwriting model. **Why PILOT not PROD**: regulators require **adverse-action explainability**. US ECOA / Regulation B and CFPB Circular 2022-03 require that when credit is denied, the lender must give specific reasons that the applicant can act on. An LLM "this applicant looks risky" is not an acceptable adverse-action reason. The 2026 compromise: LLM produces structured features, traditional scorecard model produces the decision and the reason codes, the lender ships the reason codes. EU AI Act (2024/1689) lists creditworthiness assessment as **high-risk** (Annex III §5(b)) — requiring full Article 9-15 compliance (risk management, data governance, technical documentation, human oversight). Japan's FSA + METI AI principles preserve full lender responsibility under the Banking Act and Money Lending Business Act. ### (e) Fraud detection **Production reality**: Visa Risk Manager + AI overlay, Mastercard Decision Intelligence, Stripe Radar (ML core, LLM augmentation for merchant-comm drafting), JCB Smart Code AI, Suica fraud monitoring, JP banking-association AI fraud pilots. The function set: real-time card-not-present scoring, account-takeover detection, merchant-onboarding fraud (synthetic-identity detection), wire-fraud / business-email-compromise prevention. **Why PROD**: fraud is **risk-scoring then deterministic action** (block / step-up / allow). LLM augments the scoring; deterministic rules handle the action. No regulator requires "explainability" for fraud blocks at the same level as adverse-credit decisions, though the EU's PSD2 strong-customer-authentication and Japan's FSA fraud prevention guidelines require basic transparency on why a transaction was blocked. ### (f) Advisory · robo-advisor evolution **Pilot status**: WealthNavi AI assistant (Japan robo-advisor adding LLM-driven conversational interface, 2024-2026), Schwab Intelligent Portfolios + AI, Vanguard Personal Advisor + AI, Mizuho M-AI Insight, SMBC Trust AI wealth-management pilot. The wedge: existing robo-advisors (Betterment, Wealthfront, WealthNavi, THEO) had **mostly-static** UX with rule-based rebalancing; LLM adds conversational interface, scenario simulation, and personalized commentary. **Why PILOT not PROD**: suitability and fiduciary. SEC Reg BI requires broker-dealers to act in the customer's best interest with a written rationale; Japan FIEA Article 38-2 + 40 requires Type-1 financial-instruments business operators to assess customer attributes before recommending products; EU MIFID-II Article 25 requires suitability assessment with documentation. An LLM that says "you should rebalance into emerging-market bonds" without documented suitability evaluation creates regulator risk. The 2026 compromise: LLM produces commentary marked "for information only, not advice"; the actual rebalancing recommendation comes from the existing rule-based engine with full suitability documentation. See [[payment-firms/wealthnavi|WealthNavi]] for the canonical Japan robo-advisor footprint. ### (g) Developer tooling **Production reality 2026-05**: Anthropic Claude Code adopted by BBVA, Mizuho, Goldman, Morgan Stanley internal dev orgs; GitHub Copilot (OpenAI Codex / GPT-4 backbone) deployed at JPM, Citi, BofA; Bloomberg internal AI dev tools; domestic Japan (Mizuho internal codegen, NTT Data internal). Use case: internal-facing developer productivity — code review, test generation, infrastructure-as-code drafting, SQL-from-natural-language, regulatory-document-to-code translation. **Why PROD with no regulator friction**: developer tooling is **internal-use**, not customer-facing or market-facing. The deployer retains code-review and CI/CD gates as before. Most major financial-services AI rules carve out internal-use developer tooling. The risk is **code-supply-chain** rather than financial-decision risk — see [[security/module-path-confusion-supply-chain-attack|module-path-confusion supply chain attack]] and [[security/fork-and-rebrand-5-layer-audit-framework|fork and rebrand audit framework]] for the underlying threat models. ## Vendor landscape · 2026-05 leaders by category | Category | General model API role | Financial-data / domain-system role | In-house layer | Evidence required before naming a “leader” | |---|---|---|---|---| | Customer chat | language generation, classification and tool use | authenticated customer / product knowledge and policy retrieval | conversation policy, escalation and channel integration | production scope, active users, quality / complaint data and first-party disclosure | | Back-office / compliance | extraction, summarization and agent workflow | KYC / AML data, case system and policy corpus | access, reviewer workflow and audit logging | measured throughput / error change and decision boundary | | Trading signal | text / event interpretation and research tooling | licensed market data, analytics and order / risk systems | strategy code, pre-trade controls and surveillance | live-vs-research status, asset / venue scope and monitored outcomes | | Credit underwriting | document / feature assistance and explanation support | bureau / cash-flow / application data and decision engine | policy, validation, adverse action and appeal | approval / pricing role, fairness / performance validation and legal review | | Fraud | anomaly explanation, entity resolution and alert support | network / transaction data and case management | deterministic controls, investigation and feedback | false-positive / loss metrics, production scope and customer-action boundary | | Advisory | research synthesis and communication drafting | portfolio, risk, suitability and product data | licensed adviser workflow and conflict controls | client-facing scope, adviser approval and suitability monitoring | | Developer tooling | code generation, review and tests | internal repositories, package / vulnerability systems | SDLC, CI, secrets and release controls | approved use, adoption metric, defect / security outcomes and source disclosure | Sources: ^[https://docs.anthropic.com/] ^[https://platform.openai.com/docs/models] ^[https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models] ^[https://www.bloomberg.com/professional/products/ai/] ## Composition with agent payment stack This entry maps the **application surfaces**; the underlying **transaction infrastructure** when those applications act autonomously is covered by the agent-economy protocol stack. The composition: - **Application surface** (this entry) — "what does the LLM do for the bank?" - **Identity layer** — [[agent-economy/agent-identity-defi-and-traditional-finance-bridge|agent identity DeFi and traditional finance bridge]] - **Custody and authorization** — [[agent-economy/agent-custody-and-authorization-framework-2026|agent custody and authorization framework 2026]] - **Payment protocol** — [[agent-economy/agent-payment-protocol-four-way-comparison-2026|agent payment protocol four-way comparison 2026]] - **Data interpretation pipeline** — [[agent-economy/agent-driven-market-data-interpretation-pipeline|agent-driven market data interpretation pipeline]] - **Regulator framework (trading-specific)** — [[agent-economy/ai-driven-trading-regulation-japan-2026|AI-driven trading regulation Japan 2026]] - **Mainnet readiness** — [[agent-economy/agent-protocol-mainnet-adoption-2026|agent protocol mainnet adoption 2026]] Most 2026 production deployments compose **3-4 of these layers** rather than treating "AI in finance" as a single monolith. The categorization here is meant to clarify which application category actually needs which infrastructure layer. ## Regulator stance summary · 2026-05 | Authority / source | Confirmed public material | Confirmed focus | Do not overread as | |---|---|---|---| | **FSB** | 2024 report on financial-stability implications of AI | third-party dependencies, market correlations, cyber / model risk and information gaps | A binding firm-level AI rule | | **BIS** | Working papers and research on AI in finance / central banking | use cases, governance, economics and supervisory questions | A regulation issued to private firms | | **IMF** | Fintech Notes and policy research | macro-financial, consumer-protection and capacity considerations | Directly enforceable national law | | **US Federal Reserve** | SR 11-7 and financial-stability publications | model-risk governance and system-wide vulnerabilities according to applicable scope | A statement that every LLM use is automatically an SR 11-7 model | | **US SEC** | Speeches / roundtables and the June 2025 withdrawal of the predictive-data-analytics proposal | existing securities duties, conflicts, disclosure and oversight questions | A current final predictive-data-analytics rule | | **UK FCA** | AI / financial-services publications and innovation work | safe adoption, accountability and existing regulatory outcomes | Blanket approval or prohibition of a particular model | | **European Union** | Regulation (EU) 2024/1689 plus sectoral financial law | role- and use-case-specific AI Act duties and existing financial obligations | A rule that every finance LLM is automatically high-risk | | **Japan FSA** | AI Discussion Paper v1.1 and AI public-private forum | use cases, governance, risk and regulatory-clarity dialogue | A final AI-specific trading or banking rule; the paper describes preliminary analysis | | **Singapore MAS** | Public AI / data-governance, FEAT / Veritas and consultation materials | fairness, ethics, accountability, transparency and sectoral supervision | A universal certification or safe harbor | Sources: ^[https://www.fsb.org/2024/11/the-financial-stability-implications-of-artificial-intelligence/] ^[https://www.bis.org/publ/work1194.htm] ^[https://www.imf.org/en/Publications/fintech-notes] ^[https://www.federalreserve.gov/supervisionreg/srletters/sr1107.htm] ^[https://www.sec.gov/rules-regulations/2025/06/s7-12-23] ^[https://www.fca.org.uk/publications] ^[https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32024R1689] ^[https://www.fsa.go.jp/en/news/2025/20250304/aidp.html] ^[https://www.mas.gov.sg/news] The cross-jurisdictional convergence: **no jurisdiction is granting AI agent personhood**; **all major jurisdictions retain deployer accountability**; **EU AI Act sets the high-water mark for ex-ante regulation**; **US / UK / JP / SG lean toward principle-based supervision with existing financial-services rules carrying most of the weight**. ## Sources - FSB AI/ML report (fsb.org) - BIS Working Paper 1194 (bis.org) - IMF Fintech Notes (imf.org) - US Federal Reserve Financial Stability Report (federalreserve.gov) - US SEC speeches and PDA rule trajectory (sec.gov) - UK FCA AI in financial services publications (fca.org.uk) - EU AI Act Regulation 2024/1689 (eur-lex.europa.eu) - Japan FSA news 2024 (fsa.go.jp) - Singapore MAS FEAT principles (mas.gov.sg) - Bloomberg BloombergGPT publications and Terminal AI announcements (bloomberg.com) - JPMorgan IndexGPT / SpectrumGPT public press (jpmorgan.com) - Goldman Marquee + Marquee AI public press (goldmansachs.com) - Morgan Stanley AI @ Morgan Stanley press releases (morganstanley.com) - Anthropic customer pages (anthropic.com/customers) - OpenAI finance customer pages (openai.com/index/finance) ## Related - [[INDEX|Wiki Index]] - [[agent-economy/INDEX|agent-economy index]] - [[agent-economy/agent-actorship-debate|agent actorship debate]] - [[agent-economy/agent-legal-tax-liability-framework|agent legal and tax liability framework]] - [[agent-economy/agent-payment-protocol-four-way-comparison-2026|agent payment protocol four-way comparison]] - [[agent-economy/agent-protocol-mainnet-adoption-2026|agent protocol mainnet adoption 2026]] - [[agent-economy/ai-driven-trading-regulation-japan-2026|AI-driven trading regulation Japan 2026]] - [[agent-economy/agent-driven-market-data-interpretation-pipeline|agent-driven market data interpretation pipeline]] - [[agent-economy/agent-custody-and-authorization-framework-2026|agent custody and authorization framework]] - [[agent-economy/agent-identity-defi-and-traditional-finance-bridge|agent identity DeFi and traditional finance bridge]] - [[agent-economy/claude-code-extension-architecture|Claude Code extension architecture]] - [[agent-economy/stripe-agent-toolkit-position|Stripe agent toolkit position]] - [[fintech/INDEX|fintech index]] - [[payment-firms/wealthnavi|WealthNavi]] - [[security/module-path-confusion-supply-chain-attack|module path confusion supply chain attack]] - [[security/fork-and-rebrand-5-layer-audit-framework|fork and rebrand 5-layer audit framework]]