Enterprise security and privacy, for one shared system
Two products. One shared security foundation. Intercom Helpdesk and Fin AI Agent run on the same infrastructure, the same controls, and the same team.
Every certification below maps to a real, independently audited control, so you don't have to take our word for it.
Independent audits confirm your data is encrypted, monitored, and protected by enterprise-grade controls. SOC 2 Type II compliant and ISO 27001 certified.
ISO 42001 certified for AI management system governance across the business, and Fin AI Agent is separately AIUC-1 certified, the industry standard built specifically to test AI agents.
Your customer data stays yours, with GDPR/CCPA compliance and transparent processing practices, independently verified by ISO 27701 certification, the international standard for privacy information management.
Fin AI Agent: safety built into every stage
AI safety isn't a checkbox. Fin's self-serve tools give you full visibility and control over how it behaves, before it ever reaches a customer, and every time after.
Fin trains only on your knowledge and Procedures, never on data you haven't approved, and you can opt out of fine-tuning anytime, with data deleted within 30 days. Before Fin ever talks to a real customer, run realistic Simulations against edge cases, including vulnerable-customer and compliance scenarios.
Procedures let Fin handle multi-step workflows, apply business rules, and trigger approvals, all while staying inside the boundaries you define.
Every conversation is monitored against your criteria. When confidence thresholds aren't met, Fin escalates to a human instead of guessing.
Purpose-built models. Your edge stays yours.
Fin runs on models purpose-built for customer service, not a shared, general-purpose model every company draws from. Fin Apex 1.0 is grounded in your knowledge base and post-trained on Fin's own production data, so the expertise your team builds compounds inside your workspace, not a model your competitors can also query. Third-party AI providers are contractually barred from training on your data, with zero retention once a response is generated.
Explore the Fin CX Model SuiteProven scale and reliability
Intercom Helpdesk and Fin AI Agent run on infrastructure built for continuous uptime, with redundancy across regions.
99.8%
Backed by a formal Service Level Agreement across our services, including AI Agent and Core Platform.
US · EU · AU
Data hosted in the region your residency needs require, with redundant systems and no single point of failure.
30,000+
The whole Intercom platform, Helpdesk and Fin AI Agent together, is trusted by over 30,000 customers with their most important conversations.
Trusted with the conversations that matter most
There’s a lot of transparency baked into how you configure Fin and build the workflows, which gives us control over the end-to-end experience. That was the light bulb moment for us; we were going to be letting this thing loose on our support queue, so we needed to have that level of transparency and control over the experience.”
We saw quick success with Fin, reaching around a 50% resolution rate early on, and scaling that to over 70% over time, all while keeping customer satisfaction high.”
Reaching a 91% answer rate and earning customer trust with Fin’s unrestricted, intelligent AI support, without compromising on the compliance standards a regulated exchange has to meet.”
Explore our full Trust Center
Browse compliance documentation, security policies, and sub-processor details, all self-serve.
Security & Privacy FAQs
Intercom and Fin hold SOC 2 Type II and HIPAA compliance along with ISO 27001, ISO 27018, ISO 27701, and ISO 42001 certification, and Fin AI Agent is also AIUC-1 certified (the industry standard for AI agent security). We also comply with GDPR and CCPA. All documentation is available to self-serve at trust.intercom.com.
The Fin AI Engine applies safety checks at multiple stages before any response reaches a customer: it validates the query, retrieves content grounded in your knowledge base, reranks for accuracy, and performs a final validation pass before responding. When confidence thresholds aren't met, Fin escalates to a human agent rather than guessing.
Fin can use anonymized customer data for model fine-tuning, and you can opt out at any time, with data deleted within 30 days. Third-party AI providers are contractually restricted from using your data for training, with zero data retention once an output is generated.
Intercom's services, including AI Agent and Core Platform, are backed by a 99.8% uptime SLA, with redundant systems across regions and no single point of failure. Fin's multi-model resilience means it automatically switches between AI providers to maintain performance.
Intercom integrates with Okta, Azure AD, and OneLogin for SSO, and supports 2FA, SCIM, and IP restrictions, so only the right people reach your workspace and data.
Data is hosted in the US, EU, or Australia based on your residency needs, with redundant systems architected for continuous uptime across regions.
