Financial-services agent platform Gradient Labs runs disputes and insurance claims to FCA, EU AI Act standards, discloses no insurance terms of its own
The company says its agents resolve chargebacks and file insurance claims under SOC 2 Type 2 certification with human sign-off before escalated decisions; its own site lists no insurance, indemnification or liability-cap terms.
Gradient Labs, which builds AI agents for banks, lenders and insurers, says its platform runs collections, dispute resolution, onboarding, KYB checks, insurance-claims intake and general customer service "end to end," escalating to a person only when a decision needs sign-off. On disputes, the company's site says its agent "gathers and verifies the evidence, decides the outcome, escalates to a human for sign-off, and files the chargeback with the card scheme." On insurance claims specifically, it says the agent "captures the details on the call, asks for any additional information needed for intake, fills out the record, and files the claim."
The company markets directly to regulated financial institutions, telling prospective customers its agents are built for "Global regulatory coverage" spanning "FCA Consumer Duty, CONC, Reg E and Reg Z, PSD2, GDPR, and the EU AI Act." It says "20+ guardrails on every turn" run "to keep interactions safe and compliant, every time," and that the platform carries "full SOC 2 type 2 certification" for what it calls "banking-grade data handling security."
On governance, Gradient Labs says it offers "Enterprise ready SSO, comprehensive audit logs, and role-based permissions that scale flexibly," and that its agents are "trained on your SOPs and best human agents," learning "from real conversations, not just your help centre."
Under AI Dispatch's five-dimension trust-gap rubric, Gradient Labs' published material supports real findings on audit trail, runtime governance and compliance — audit logs, per-turn guardrails and SOC 2 Type 2 certification are all named on the same site. Insurance/indemnity and outcome-settlement terms are absent — fields AI Dispatch's index records as no public information for the company. That combination, real operational-governance disclosure paired with silence on who bears the cost of a wrong decision, recurs across vendors AI Dispatch has indexed to date. It is more pointed here than most: Gradient Labs' own marketing names insurance-claims filing among the regulated workflows its agents complete without a person in the loop until after the decision is made.
Entries in this piece 1
Published index entries backed by the same source documents this piece cites.
Sources 7
“the AI agent gathers and verifies the evidence, decides the outcome, escalates to a human for sign-off, and files the chargeback with the card scheme.”
gradient-labs.ai · checked Aug 7, 2026“the AI agent captures the details on the call, asks for any additional information needed for intake, fills out the record, and files the claim.”
gradient-labs.ai · checked Aug 7, 2026“Global regulatory coverage Built for US, UK, and EU rules: FCA Consumer Duty, CONC, Reg E and Reg Z, PSD2, GDPR, and the EU AI Act.”
gradient-labs.ai · checked Aug 7, 2026“20+ guardrails on every turn Financial-services-specific guardrails run on every turn of conversation to keep interactions safe and compliant, every time.”
gradient-labs.ai · checked Aug 7, 2026“SOC 2 certified Banking-grade data handling security with full SOC 2 type 2 certification.”
gradient-labs.ai · checked Aug 7, 2026“Enterprise ready SSO, comprehensive audit logs, and role-based permissions that scale flexibly.”
gradient-labs.ai · checked Aug 7, 2026“Trained on your SOPs and best human agents Learns from real conversations, not just your help centre, and executes natural-language procedures that mirror your operations.”
gradient-labs.ai · checked Aug 7, 2026