Story
Anthropic
Gradient LabsClaude Powers Regulatory-Compliant AI Customer Support Automation for Financial Services
Gradient Labs, an AI customer support platform for financial services, resolved the limitations of basic automation by leveraging Claude to achieve 80-90% resolution rates while maintaining regulatory compliance.
Value results
| Category | Value result |
|---|---|
| Revenue | Outcome-based pricing model enabling profitable scaling of complex query resolution |
| Experience | 98% customer satisfaction outcomes with AI-powered support automation |
| Risk and compliance | Regulatory compliance built into platform architecture for financial services requirements |
| Capability | 80-90% resolution rates for customer support automation without human intervention |
| Capability | Automation of approximately 80% of tasks within customer support and operations functions |
Gradient Labs identified a critical gap in financial services customer support automation. While industry competitors reported 40-50% automation rates, they were only automating simple, time-consuming queries with minimal operational impact. Founded by Dimitri Masin, Neal Lathia, and Danai Antoniou—former leaders at fintech company Monzo—Gradient Labs built an AI-powered platform specifically designed for regulated financial institutions. The platform automates complex, multi-step customer queries while embedding regulatory guardrails such as vulnerable customer detection and unauthorized financial advice prevention directly into its core architecture.
Gradient Labs selected Claude as the foundation for its platform after evaluating multiple AI models. Claude powers critical functions including intent classification, standard operating procedure (SOP) execution, response generation, and knowledge graph integration. According to Masin, Claude's "ability to generate more natural, human-like, and authentic responses" combined with its superior reasoning capabilities made it the optimal choice for understanding customer queries within context and determining appropriate next steps in complex financial workflows. The platform leverages Claude to follow multi-step regulatory processes with precision while maintaining conversational quality expected in premium financial services.
The results demonstrate substantial operational value for Gradient Labs' financial institution clients. The platform achieves 80-90% resolution rates for end-to-end query handling without human intervention—translating to meaningful time savings compared to competitors focusing solely on automation percentages. Customer satisfaction reaches 98%, while the platform automates approximately 80% of tasks within customer support and operations functions. Gradient Labs' outcome-based pricing model aligns incentives with client success, with tiered pricing for increasingly complex resolutions. Looking forward, the company is extending Claude's capabilities to advanced use cases including fraud detection and money laundering investigations, enabling financial institutions to transform how they deliver intelligent, compliant customer support.
