We've been quietly shipping all summer long. Here is what we've been up to: We’ve shipped a lot at Crisp this summer. ☀️ But if we had to summarize the direction in one sentence: AI customer support is becoming more powerful — so the tools around it need to become more transparent, controllable and easier to operate. A few highlights from our Summer 2026 Product Update: 🔍 See exactly what your AI is doing With @hugoaiagent Observability, teams can now inspect the full timeline of an AI-powered conversation: which models were used, which tools were called, how the AI handled each step, how long it took and where things went wrong. AI shouldn’t be a black box, especially when it’s talking to your customers. 🧠 More choice in the models powering Hugo We’ve expanded the AI models available inside Crisp and improved our own models to deliver more reliable answers and fewer hallucinations. 👥 A better way to organize support teams Operator Teams can now be created once and reused across mentions, routing and Sub-Inboxes, with memberships staying synchronized automatically. 🎯 Separate instructions for Hugo Copilot Your customer-facing AI Agent and the Copilot helping your support team don’t necessarily need to behave the same way. You can now configure dedicated instructions specifically for Copilot. 📊 Better visibility into human support activity New messaging metrics make it easier to understand message volume, workload and activity by operator and message type. And beyond those larger releases, we’ve also been working on the fundamentals: → A refreshed Crisp interface → A completely redesigned Dark Mode → Stronger bot and spam protection → Better multi-domain security → More control over email tracking and privacy → Redesigned Sub-Inboxes → Faster workflows and analytics → A ~30% smaller initial Chat Widget bundle → Improvements across routing, translations, integrations, SDKs and mobile apps Individually, some of these changes are small. Together, they’re part of the same goal: Make AI-powered customer support easier to understand, safer to operate and genuinely pleasant for support teams to use every day. Full Summer 2026 Product Update 👇 crisp.chat/en/blog/summer-20…

SECURITY ADVISORY — TanStack npm packages A supply-chain compromise affecting 42 @tanstack/* packages (84 versions total) was published to npm earlier today at approximately 19:20 and 19:26 UTC. Two malicious versions per package. Status: ACTIVE — packages are deprecated, npm security engaged, publish path being shut down. Severity: HIGH — payload exfiltrates AWS, GCP, Kubernetes, and Vault credentials, GitHub tokens, .npmrc contents, and SSH keys. If you installed any @tanstack/* package between 19:20 and 19:30 UTC today, treat the host as potentially compromised: • Rotate cloud, GitHub, and SSH credentials immediately • Audit cloud audit logs for the last several hours • Pin to a prior known-good version and reinstall from a clean lockfile Detection — the malicious manifest contains: "optionalDependencies": { "@tanstack/setup": "github:tanstack/router#79ac49ee..." } Any version with this entry is compromised. The payload is delivered via a git-resolved optionalDependency whose prepare script runs router_init.js (~2.3 MB, smuggled into each tarball at the package root). Unpublish is blocked by npm policy for most affected packages due to existing third-party dependents. All 84 versions are being deprecated with a SECURITY warning, and npm security has been engaged to pull tarballs at the registry level. Full technical breakdown, complete package and version list, and rolling status updates: github.com/TanStack/router/i… Credit to the security researcher for responsible disclosure.

🚨 MIT proved you can delete 90% of a neural network without losing accuracy. Researchers found that inside every massive model, there is a "winning ticket”, a tiny subnetwork that does all the heavy lifting. They proved if you find it and reset it to its original state, it performs exactly like the giant version. But there was a catch that killed adoption instantly.. you had to train the massive model first to find the ticket. nobody wanted to train twice just to deploy once. it was a cool academic flex, but useless for production. The original 2018 paper was mind-blowing: But today, after 8 years… We finally have the silicon-level breakthrough we were waiting for: structured sparsity. Modern GPUs (NVIDIA Ampere+) don’t just “simulate” pruning anymore. They have native support for block sparsity (2:4 patterns) built directly into the hardware. It’s not theoretical, it’s silicon-level acceleration. The math is terrifyingly good: a 90% sparse network = 50% less memory bandwidth + 2× compute throughput. Real speed.. zero accuracy loss. Three things just made this production-ready in 2026: - pruning-aware training (you train sparse from day one) - native support in pytorch 2.0 and the apple neural engine - the realization that ai models are 90% redundant by design Evolution over-parameterizes everything. We’re finally learning how to prune. The era of bloated, inefficient models is officially over. The tooling finally caught up to the theory, and the winners are going to be the ones who stop paying for 90% of weights they don’t even need. The future of AI is smaller, faster, and smarter.

After more than 6 months of non-stop work on a secret Crisp product, I'm incredible proud to announce that it's now out! It's called Hugo. ⬇️ 👉Check it out at hugo.ai Hugo helps you automate a large chunk of your Crisp customer support with a top notch AI Agent, so that you can focus on providing human customer support on what matters: high value customers, gathering feedback, handing bug reports, etc. Hugo handles all the rest for you while you sleep, or work on your product. You can already try & train Hugo from a single dashboard from your Crisp app, feeding it with your own website, CSV documents, Q&A and your Crisp KB. If you already run a Crisp chatbox on your website, enabling Hugo only takes 2 minutes, and it's included for free in your Crisp paid subscription! This is great pair work with @baptistejamin and the rest of the Crisp team.