Built for FCA-regulated firms
Ask a compliance question. One model generates the answer. Ten agents try to tear it apart. You only see what survives.
0
Pipeline Stages
0
Verification Agents
0
Production Modules
0
Benchmark Cases
How it works
Nine stages between your question and the answer.
Every answer passes through all nine before you see it.
Classify
Determines query type, complexity, risk level, and which FCA sourcebooks apply before anything else runs.
Retrieve
Expands your question into multiple phrasings. Searches vectors and the knowledge graph. Follows cross-references, exemptions, and amendment chains.
Generate
Produces the initial answer using retrieved evidence. The next six stages check whether that answer is actually correct.
Verify
Ten agents check the answer independently with different evidence subsets. Sentence-by-sentence. If a cited section does not exist, the score gets hard-capped.
Analyse
Aggregates agent verdicts with weighted voting. Tier 1 agents carry 1.5x weight. Detects disagreement patterns and flags low-consensus claims.
Cite
Links every claim to its source. Confirms each cited FCA section exists, is current, and matches the rule type the answer implies.
Calibrate
Adjusts raw confidence so the percentages mean something. If the score says 80%, the system should be right about 80% of the time. Calibration corrects for overconfidence.
Explain
Generates counterfactual explanations. Shows what would need to change to flip the verdict. Makes the reasoning auditable.
Score
Produces the final trust score from five weighted components: verification, dimensional analysis, citations, calibration, and retrieval quality.
Query
Scroll to begin...
Scroll to see verification
Modules
Fourteen modules.
Each module has a production backend, API endpoints, and test coverage.
Compliance Q&A
Ask anything about FCA rules in plain English. The answer goes through nine stages before you see it. Every citation is checked against the actual handbook.
Decision Vault
Encrypted decision records for Senior Managers & Certification Regime (SM&CR) compliance. SHA-256 signed approvals, escalation history, five-year retention. Generates certificates auditors can verify.
Regulatory Genome
Maps every obligation that applies to your firm type across all FCA sourcebooks. Graph traversal across SYSC, COBS, SUP, and nine other sourcebooks, with enforcement data and risk scoring per obligation.
Regulatory Forecast
Reads consultation papers, Dear CEO letters, and policy statements. Scores each one against your firm type and tells you which upcoming changes will affect you.
Compliance Gateway
Checks AI-generated content before it reaches clients. Multi-stage analysis, data boundary checks, PII scanning. Sits between your AI systems and your output.
Policy Generator
Drafts internal policies mapped to specific FCA provisions. Every statement links to the regulation it satisfies. Auditors can trace each policy back to its source.
Persona Auditor
Maps SM&CR roles to AI agent behaviour. Detects when an agent starts drifting from its risk profile. Flags gaps before the regulator asks about them.
Pulse Engine
Monitors FCA feeds around the clock. Parses enforcement notices, extracts fines and breach details, clusters signals by topic. Summary in your inbox by morning.
Regulatory Arbitrage
Compares rules across jurisdictions. Where do UK and EU overlap? Where do they conflict? Runs parallel queries across regulatory frameworks and scores the friction.
Enforcement Intelligence
Analyses the graph of FCA enforcement actions. Finds repeat offenders, co-violation patterns, and how fines are trending. Tells you which rules the FCA is actually enforcing right now.
Inclusion Scanner
Consumer Duty vulnerability assessment. Checks against twelve FCA vulnerable customer categories for access barriers, comprehension issues, and pricing problems.
Evidence Bundle
Generates a ZIP file ready for FCA inspection. Cover sheets, verification certificates with per-agent breakdowns, audit logs, decision records, and your firm compliance profile.
Code Scanner
Analyses source code against FCA compliance rules. Flags data handling, consent flows, and record-keeping gaps that standard code review misses.
Sandbox Simulator
Tests fintech products against sandbox acceptance criteria across multiple jurisdictions. Run the scenario before it is real.
Verification agents
Ten agents. None of them trust each other.
They run in parallel. The important ones carry more weight. If the two most critical agents fail, the score gets capped at 40% no matter what the others say.
Citation Verifier
Tier 1 · 1.5xChecks every cited rule character by character against the FCA Handbook. If you cite SYSC 10.1.8R, it confirms that exact section exists and says what you claim it says.
Temporal Validity
Tier 1 · 1.5xKnows which rules are current, which have been superseded, and which are pending. Catches answers that cite revoked provisions as if they still apply.
Logical Consistency
Tier 2 · 1.2xFinds contradictions. If the answer says "firms must" but cites Guidance rather than a Rule, that is a logical error. Guidance is not binding.
Cross-Jurisdiction
Tier 2 · 1.2xCatches when UK-specific advice accidentally references EU law that no longer applies post-Brexit. Validates the regulatory framework matches the question.
Completeness
Tier 2 · 1.2xUses the Regulatory Genome to find obligations the answer missed. A correct but incomplete answer is still a problem in compliance.
Adversarial
Tier 2 · 1.2xTries to break the answer. Runs six attack types: edge cases, outdated rules, bad assumptions, missing information, misinterpretation risks, and scope problems.
Regulatory Intent
Tier 3 · 1.0xChecks if the answer aligns with why the rule exists, not just what it says. A technically correct answer that misses the point still gets flagged.
Plain Language
Tier 4 · 0.6xMakes sure the answer is understandable to someone who is not a lawyer. Flags jargon and ambiguous phrasing.
Precedent
Tier 4 · 0.6xChecks the answer against how the FCA has actually enforced the rules. What the handbook says and what the FCA does about it are sometimes different things.
Audit Trail
Tier 4 · 0.6xRecords every step. Who checked what, when, what they found. The verification process itself needs to be auditable.
Transparency
We show you how it works. And where it falls short.
How we verify
- The answer gets broken into individual sentences. Each sentence is classified and verified against source evidence independently.
- Each agent gets a different subset of evidence on purpose. The adversarial agent deliberately gets the weakest evidence. This prevents groupthink.
- Deterministic checks confirm cited sections exist, are current, and have the right rule type. If the section does not exist, the score gets capped regardless of what agents say.
- Agents challenge each other through four rounds of structured debate. They can concede, rebut, or maintain their position. Fully deterministic and reproducible.
What we cannot do
- AI makes mistakes. Our verification catches more than a single model, but it will not catch everything. The trust score tells you how much the agents agreed. It does not mean the answer is definitely right.
- This is not legal advice. Always talk to a qualified advisor before making compliance decisions. Your team makes the final call.
- FINGAURD is not regulated by the FCA. We build technology tools, not regulated financial services.
- Confidence calibration improves with more data. Early scores may be over- or under-confident until we have enough labelled queries to calibrate properly.
ICO Registered
UK data protection compliant
Encrypted
TLS 1.2+ in transit, AES-256 at rest
SCCs in Place
International transfers covered
No Model Training
API data not used for training
Pricing
Pick a plan. Cancel anytime.
All prices exclude VAT at 20%. Billed monthly. Every plan includes the full 9-stage verification pipeline and 10 agents. Higher tiers unlock more modules, MCP server, and scheduled compliance monitoring.
Early Adopter
- 100 queries/mo
- 5 core modules
- 10 verification agents
- 9-stage pipeline
- Email support
Starter
- 500 queries/mo
- 10 modules
- 10 agents + debate verification
- Decision Vault audit trails
- Priority support
Professional
- 5,000 queries/mo
- All 14 modules
- Full verification + hallucination detection
- Evidence Bundle generation
- Regulatory Genome mapping
- Dedicated CSM
Enterprise
- Unlimited queries
- All 14 modules
- MCP server (10 compliance tools)
- Scheduled compliance monitoring
- UK-hosted option
- SSO, API & custom SLA
FAQ
Questions compliance teams ask us.
How is this different from ChatGPT with a compliance prompt?+
ChatGPT gives you one answer from one model with no verification. We run the answer through nine more stages after generation. One agent breaks it into individual sentences and verifies each one. Another confirms every cited FCA section actually exists. A third tries to break the answer on purpose. If cited sections do not exist, the trust score gets capped.
Is this regulated by the FCA?+
No. FINGAURD is a technology tool. We do not provide financial advice or regulated services. Your compliance team makes the decisions. The system includes a guardrail that blocks any answer that crosses into giving legal advice.
What happens when the AI gets it wrong?+
You see it. Each claim is classified as supported, unsupported, contradicted, unverifiable, or fabricated. If more than 30% of claims fail verification, the answer gets blocked entirely. The explainability engine shows what would need to change to flip the verdict.
Can we use this for audit evidence?+
Yes. The Decision Vault stores encrypted, signed decision records with five-year retention. The Evidence Bundle generates a ZIP file with cover sheets, per-agent verification certificates, audit logs, and your compliance profile. Built for SM&CR record-keeping.
Where is our data stored?+
Queries go through OpenAI API, which does not train on API data per their current policy. Frontend on Vercel, backend on Railway. Vector embeddings in Pinecone with per-tenant isolation. Encrypted in transit and at rest. Enterprise plans offer dedicated UK-hosted infrastructure.
What is the MCP server?+
Model Context Protocol. It exposes ten compliance tools that any AI agent can call directly: query regulations, verify answers, check firm status on the FCA Register, get enforcement history. Claude, GPT, LangChain, CrewAI, and AutoGen can all connect. Available on Enterprise plans.
Contact
Want to see it?
Try the platform directly, or request a 20-minute walkthrough.
Or request a demo below and we'll reach out.