Intercom

We’ve just published new research that explores how end users feel about AI Agents and how much they trust them. While sentiment is rising and people trust AI more than ever before, their trust is still conditional. But trust can be built. In the 2026 AI Sentiment Report, we show you how to do this. Read the key findings from our research in this edition of the Blueprint newsletter. P.S. We’re very excited to share that today Fin officially joins Salesforce. Find out more in the newsletter.

Today Fin officially joins Salesforce. It’s a huge moment for us. Fin was the very first AI Agent for CX and started an entirely new category. Today Fin is the highest performing Customer Agent on the market, running on our custom AI models trained specifically for CX, and resolving over 2 million customer issues every week. Thank you to our customers, teammates, partners, and supporters who’ve been a huge part of our journey so far. Now with Salesforce, Fin will make perfect customer experiences possible for millions more people. You can read the full press release linked in the comments. 

AI Agents can resolve the majority of queries end to end, but sometimes you'll still want to have a human involved, like for validating high-stakes decisions or handling tightly regulated questions. Deciding who does what, when handoffs and escalations should happen, and why is an important part of CX design. In the latest edition of the Blueprint newsletter, we share advice from our VP of Support at Fin, Declan I., on how to do it.

Your customer support experience is a product, so the team behind it should operate like a product team. When people interact with your business, they don’t experience sales, product, and support as separate things – they experience your brand as a whole. Great support experiences can do as much for loyalty and expansion as your product itself. Now that AI Agents handle the majority of customer-facing queries, your support team focuses on building and optimizing the customer experience with the same rigor and accountability product teams apply to your core product. In this edition of The Blueprint Newsletter, Ruth O'Brien, Senior Director of AI Support at Fin, unpacks how to run support like a product organization so you can define, build, and own your customer experience just like a product org owns a product.

Alannah C., AI Support Engineer at Gamma, has been instrumental in transforming their support to over 100M users. Using Fin and now Operator, Alannah designs complex workflows, tests them, spots gaps in performance, and improves the knowledge Fin draws on—all in plain language. Hear how she and Gamma continue to reinvent their support operations, at this year’s Pioneer. Register to attend via the link in our comments.

When an AI Agent gets held up at security review, it's rarely because of the technology. It's usually because your CISO is missing the information they need. They don't know how you plan to use the Agent, what data it touches, or what actions it can take. In this edition of The Blueprint Newsletter, Fin’s CISO, Thibault Candebat covers everything you need to prepare for your security team to give an AI Agent a confident “yes”.

AI can happen to your people, or for them. Which one comes down to how you lead them through the change. In this edition of the Blueprint Newsletter, we share five ways to bring your people with you as you roll out AI, from being honest about how the work is changing to showing them where their careers can go next.

CSAT was never built for a world where AI handles most of your conversations. It captures less than 10% of them, and the answers you do get skew to the extremes. As AI takes on more, that blind spot only grows. This week, William Lane wrote about how we're closing that gap at Fin by scoring every conversation instead of sampling a handful. In this edition of the Blueprint Newsletter, we break down how we do this, how to set targets from your own data instead of carrying over old CSAT thresholds, and how full visibility turns recurring problems into things you can actually fix. If you're making CX decisions on a fraction of the picture, this one's worth a read.

When your product team is shipping fast, your AI Agent needs to keep up. We've shipped 191 product updates at Fin in the first half of this year alone. Our New Product Introduction Manager Sean Reid wrote about how we make sure our Agent is ready for every single one. He breaks down the process: getting support into launch planning early, the three-step checklist we run every time (prepare → review → test), and why a single owner keeps it all from falling through the cracks. If your product team is shipping faster than you can update your knowledge base, this might help.