Glean:GO 2026 Keynote  🔗 **Source:** [Glean:GO 2026 Keynote](https://www.youtube.com/watch?v=vGmmt7p2Hyk) > [!note]+ Document Notes > --- > tags: [enterprise-ai, ai-agents, model-routing, token-economics, ai-governance, enterprise-search] > author: Glean > source: Glean (YouTube) > reference: https://www.youtube.com/watch?v=vGmmt7p2Hyk > date created: 2026-09-21 09:00 > --- > > [!info] Generated analysis — not the author's words. Sections below are machine-written > > from the captured source. Only text inside `> [!quote]` blocks is verbatim from the author. > > Do not quote or attribute any other line to the named author. > # 🏷️ Title: Glean:GO 2026 Keynote > ## 🧭 Table of contents > - [📝 Summary](#-summary) > - [📖 Main content](#-main-content) > - [🗒️ Notes](#️-notes) > - [📚 References](#-references) > - [🗣️ Paraphrase](#️-paraphrase) > - [📰 Newsletter](#-newsletter) > - [✅ Action items](#-action-items) > - [🧪 Hypotheses](#-hypotheses) > - [📊 Methodology](#-methodology) > - [❓ Questions](#-questions) > - [🔮 Future research](#-future-research) > - [💡 Implications](#-implications) > - [🧠 Learning styles assessment](#-learning-styles-assessment) > - [📊 Accessibility assessment](#-accessibility-assessment) > - [🌍 DEIB assessment](#-deib-assessment) > - [✅ Logical validity](#-logical-validity) > - [🛡️ DARVO analysis](#️-darvo-analysis) > - [🎯 Conclusion](#-conclusion) > - [🔄 Refactor appendix](#-refactor-appendix--gleango-2026-keynote) > ### 🔗 Source links > - [Glean:GO 2026 Keynote (YouTube, 1:48:58)](https://www.youtube.com/watch?v=vGmmt7p2Hyk) > - Transcript captured locally: `/Users/courtneyrobertson/.youtube_analyses/transcripts/vGmmt7p2Hyk.txt` (16,470 words) > > [!warning] Transcription reliability > > This is an auto-generated transcript with heavy speech-recognition noise. "Glean" appears as > > "Gleam," "Gleen," "Clean," and "Le." Product and person names are garbled: Glean Tau appears as > > "Clean Tow," "Glean Tao," "Gleantow," and "Glean Towels"; Nemotron as "Neotron"/"Neatron"; Kimi > > as "Kimmy"; DaVita as "Dvita"/"Dvidita." Dollar figures lose their "$0." prefix. Every number and > > name below is recorded as spoken; where the transcript contradicts itself, both readings are shown. > ## 📝 Summary > - **What happened.** Glean ran a 109-minute opening keynote at Glean:GO 2026 in San Francisco (Fort Mason) on 2026-09-21, streamed to what the host put at 5,000+ attendees. The company announced a proactive-AI product line (interactive dashboards, independent agents, proactive task management, team chat), Glean Transform, auto routing under a new "Glean Intelligence" umbrella, agent governance policies, context-aware AI threat detection, the Glean AI gateway, and Glean Tau, a desktop app for code execution and co-working. > - **Why it matters if you use Glean at work.** Three of these change day-to-day operating decisions rather than positioning: auto routing plus AI usage controls change what a Glean seat costs, AI gateway changes whether your memories and skills stay portable across Cursor, Copilot, and ChatGPT, and agent governance policies plus threat detection change who signs off before an agent gets write access. > - **Key stakeholders.** Glean: Arvind Jain (founder and CEO, transcribed "Arvin Jane"), Amar Malitira (COO), head of product (name garbled — introduced as "Amrajan," self-introduced as "John"), Tony (token yield), Eddie (evals), Michael (finance), Aaron Grove (VP solutions engineering), Judith Bramwell (EMEA lead, addressed as "Juliet" at the close), plus demo presenters Selene and Blair. External: Phil Lutka (executive director, core IT, General Motors), Kate Rooney (CNBC), Alexander (OpenAI, enterprise products, formerly leading Codex agent work), Carrie (NVIDIA), Evan Spiegel (Snap, recorded), "VC" (SVP engineering, Snowflake), Madu Nasharim (CIO, DaVita). > - **Risks and tradeoffs.** The headline 81% token-cost saving and 78% preference rate come from Glean's own benchmark against one named competitor, Claude Co-work, with the query set, harness, and evaluation environment disclosed only in a single sentence and not published on stage. Adoption numbers cited by customers are impressive and internally inconsistent in at least two places (DaVita's 17,000 vs 70,000 users; Snowflake's "16x swing" on a 15/60-to-60/15 NPS shift). Nothing announced carried an explicit GA, beta, or preview label. > - **Bottom line.** Treat the product list as a real roadmap signal and the numbers as vendor-supplied until you rerun them on your own corpus. The two items most worth a procurement conversation this quarter are auto routing with department-level spend profiles and AI gateway memory portability, because both have contract and lock-in consequences beyond the feature itself. > ## 📖 Main content > ### The argument the keynote makes > The keynote runs a three-problem structure that Arvind Jain sets up at roughly 13 minutes and every subsequent segment answers in turn. The three problems, in his framing: enterprises can't get the right context to their agents, so quality is poor and business value doesn't land; AI costs more than most enterprises can afford; and AI has created new problems of its own, specifically a proliferation of AI-written documents, skills, and agents that are hard to manage and secure. > Each block of the keynote maps to one problem. Context is answered by the context engine (retrieval plus enterprise graph) and by proactive AI. Cost is answered by model choice, auto routing, and AI usage controls, bundled as Glean Intelligence. Sprawl and security are answered by Glean Protect additions, agent governance policies, context-aware AI threat detection, and the AI gateway. > ### Context: retrieval plus an enterprise graph, positioned against MCP > Glean's definition of context is the collective human intelligence and experience of an enterprise, packed so agents can operate at the level of human experts. The stated data problem is that structured data and data lakes capture only 5 to 10% of enterprise knowledge and that playbooks and SOPs cover only a slice more; the rest lives in people's heads and surfaces in emails, messages, meeting notes, and exception approvals. > The mechanism: retrieval surfaces current high-quality information across systems; the enterprise graph maps people, customers, teams, projects, and their relationships. Glean says it builds these ontologies largely automatically and exposes them for you to visualize, work with, and change. > The competitive framing here is the most load-bearing and least supported claim in the keynote. Jain positions the graph against "the industry's typical approach of delivering shallow contexts via MCP" and then states Glean is the only context and intelligence company that truly understands your enterprise. No comparison, test, or definition of "shallow" accompanies that claim. Note the tension with a later segment: the AI gateway is itself described as a control plane for LLMs *and MCPs*, and Glean reports 78% of its customers reaching Glean through APIs and MCP. > ### Proactive AI: five things, one bet > The head of product frames human attention, not AI capability, as the bottleneck, then announces proactive AI in five parts: > 1. **Proactive task management** — surfaces what needs attention, prioritizes it, and prepares the next step. Demoed as Glean having already drafted a customer slide deck, carrying the engineering root cause into it. > 2. **Interactive dashboards** — called "our newest artifact." A dynamic, shareable view combining CRM structured data, warehouse usage telemetry, and unstructured customer data. Demoed on a renewal-health dashboard where a fictional account, Frasier Automotive, shows red because an API failure hit during performance reviews. > 3. **Independent agents** — long-running agents with their own distinct identity, memory, and context, able to act proactively and coordinate across teams, with governance so they "don't go off the rails." Demoed as an on-call agent in Slack identifying an escalation, analyzing the issue, reading code, opening a pull request, and tagging engineers. > 4. **Memory and skills** that learn and self-heal continuously. > 5. **Team chat** — a shared space where multiple teammates and Glean work in one thread. This is the segment's actual thesis: AI moving from single-player to multiplayer. > ### Glean Transform > Announced as a [00:00:00] Heat up here. Heat. Heat. Heat. Heat. Heat. Heat. Heat. Heat. >> [music] >> Heat. Heat. >> [music] >> Heat. Heat. All right, we are getting ready to get started with clean go 202026. Please come grab a seat in the keynote area. Thank you. Heat. Heat. [music] [music] Heat. Heat. [music] >> [music] [music] >> back. Hey [music] Heat. Heat. [music] [music] [music] >> [music] >> Heat. Heat. [music] Heat. Heat. [music] Heat. Heat. [music] Heat. Heat. N. [music] >> [music] [music] >> Good morning, San Francisco. My name's Judith Bramwell and I lead our gleam business in air. I'm delighted to be here today all the way from sunny London to kick off our annual Glingo [00:06:28] conference. Thank you so much for coming all the way here. And to all those joining us around the world on the live stream, welcome to Glingo. It's actually also my wedding anniversary today. >> [applause] >> So, what better way than to celebrate it with you? Given this special day, I was wondering, is there a way I can make a link between a wedding and AI? So, here goes. Like AI, all good wedding speeches start with good context. Who you are, how you met, what's important to you. Like AI, the cost of a wedding will probably be more than you originally planned. Maybe the wedding was expensive, but what about the ROI? Surely, it's got to [00:07:17] be worth it in the long run. Unlike AI, you typically only get one model choice at your wedding. Well, we certainly have a great congregation here today. I'm feeling a lot of love and a lot of energy in this room. It's going to be an amazing two days. When we came together last year, enterprise AI was already moving quickly. 12 months on quickly feels like a massive understatement. Frontier and open models now coexist with frontier systems pushing the boundaries of capability while open alternatives broaden access, customization, and innovation. At the same time, the price performance curve is resetting as the cost of [00:08:04] increasingly powerful AI continues to fall. The number of ways employees can access and use AI has multiplied. Agents are moving into day-to-day work and experiences from what AI can do inside every company have changed enormously. But most important progress is happening inside your organizations. What if every employee could tap into the full context of their company right when they need it? That's what we see our customers doing each and every day. Putting that context at every employees fingertips across every function. In the sales function, before a seller walks into a meeting to discuss a renewal with a customer, they know the [00:08:54] promises made, the issues still open, and what could put that deal at risk. For developer teams, a developer opens a pull request. Glean shows the systems it touches, a similar outage from last year, and who needs to review it before the code ships. If you look at support, when they receive a major escalation, instead of searching multiple systems, the team sees the account history, the product issue, fixes already tried, and the engineer most likely to solve it. And for new employees, when they join, Lean instantly connects them to the people, tools, best practices, policies, and password they need to get up to [00:09:36] speed. So work no longer moves through your organization as a series of handoffs, creating bottlenecks, losing context, and waiting for someone to connect the dots. Instead, knowledge moves ahead of the work. Every person across every function has the context of the whole company behind them so they can take the next best action to move the business forward. And that progress is already represented in this room. Over the past year, you've pushed the boundaries of AI and put it to work against the real business problems. You have connected knowledge across your enterprise systems. You've turned years of institutional knowledge into usable [00:10:25] intelligence. You've built agents for real workflows. And you have worked through the hard questions around security, governance, and adoption. You've not just simply embraced AI, you are helping define what it becomes. That is why Gleingo is not only about what we have built, it is a celebration of what we're building with it. Throughout today, you'll hear directly from some of the companies leading that change. From our customers at General Motors, Dvita, Ericson, and eBay to our partners at Open AI, Nvidia, Snowflake, Cisco, Deote, and Data Bricks. They'll share what they're building, the results they're seeing, and what they've [00:11:12] learned along the way. Over the next 90 minutes, we are going to show you where this goes next. You will see how AI is moving from simply answering questions to anticipating what needs attention and helping move work forward. How companies can get more value from every model and every token and how an open secure platform gives organizations the freedom to keep using the best technology as it evolves. So to begin, please join me in welcoming Glean's founder and CEO, Arvin Jane. [music] [music] Hello everyone. Thank you so much all the 5,000 plus people who are joining us today for Glean Go from this room as well as online. And a special thank you [00:12:16] to all of our customers who will be sharing over the next two days how they're using Glean to drive business value within their own organizations. General Motors, they're using Glean to help engineers design better cars faster. McCarthy is transforming construction management so teams can arrive at the job site ready with project plans and safety procedures you know on hand. David is rethinking patient support helping healthcare employees surface trusted information that improve patient outcomes. What you what you could do with AI a year ago doesn't resemble at all things that we can do today. But there's still a lot of work to do. [00:13:06] Uh, every enterprise that we talk to has three things on top of their mind today. They're struggling to bring the right context to their agents, which results in poor quality and makes it hard to realize business value. And of course, AI costs are running very, very high, absolutely more than what most enterprises can afford today. And while AI is solving some problems for us today, it's also introducing some new ones. Proliferation of AI written documents, also known as slop, skills, and agents, which are all getting very, very hard to manage and secure. We'll spend the next two days talking about how Glean will help you tackle [00:13:53] these problems and make AI work for your enterprise. Let's talk about context first. Since bringing enterprise knowhow to agents isn't easy. Today, employees spend a lot of time coaching AI agents and are feeling overburdened with that. This is bots setitting. We are spending too much time working for AI as opposed to AI working for us. This is not how it's supposed to be. And bot sitting is just one negative consequence of shallow context. There are many more. It's also why token costs keep climbing. It's also why every new AI app that you bring in your enterprise feels like one more disconnected front door. To understand shallow versus deep [00:14:46] context, we wanted to hear we wanted you to hear from experts in the field. So we asked them what context means to them. So let's take a look at that. >> Context is what transforms intelligence into relevance. >> The most critical part of getting the right and precise output from the model. >> The understanding the tribal knowledge of all of the information across all of your systems that fundamentally makes up your business. anything fed in alongside your prompts to help shape the response you [music] receive. Context is what gives information meaning. many different things to each one of them. But for us at Gleam, we recognize [00:15:37] context as the collective human intelligence and the experience of your enterprise packed in a form that AI agents can use to operate at the level of your human experts. As a pioneer in enterprise AI when we started in 2019, Gle has been thinking about the need for rich context in AI years before the industry caught on to this concept. We built the context foundation that now feeds enterprise AI across millions of users. Why is context so hard to build and where is the data to build it from? Structured data and data links capture only 5 to 10% of all enterprise knowledge and documents like playbooks, SOPs go only as far. [00:16:31] Much of the real knowhow of how work happens lives in people's heads. But fortunately it does surface in emails, messages, meeting notes, exception approvals. Um there are there you know all of that data is available somewhere. People are expressing it every day as they do their regular work. The challenge is in capturing those moments of human judgment and turning it into something useful. So, Glean's context engine, it combines advanced retrieval with a powerful enterprise graph. Retrieval surfaces the most current highquality information across your systems. And the enterprise gra graph maps the people, customers, [00:17:18] teams, projects, and relationships between them that essentially define how work actually gets done in your enterprise. Glean builds these ontologies largely automatically, but we do give you the ability to visualize it, to work with it, and also to change it when you need to change it. That deep understanding of your business, how it operates is what Glean sets itself apart from the industry's typical approach of delivering shallow contexts via MCP. Glean is the only context and intelligence company that truly understands your enterprise. I would like to talk bit about other things that powerful context can enable [00:18:03] AI to do. It can make AI proactive as it understands you, your role, your work, things that you need to get today and clean can do work for you before you ask for it. It also allows AI to become a true teammate as opposed to an agent that uh does a very specific task when you're commanded to do that. Glean's longunning independent agents use rich context to learn how to become as good as an escalation engineer or a customer support agent or a technical account manager and deliver business value at a completely different level. Think of agents now, these long-running independent agents as um as real human beings, you know, who are operating at [00:18:52] the level of, you know, that expertise and that experience that everybody in the company has learned over the years. So, let's switch topics now. Now, I want to talk a little bit about AI sprawl that we have all been experiencing. Most businesses I speak with have ended up with multiple AI front doors and agent platforms. that are all connected partially and inconsistently with your company's data. There's also a proliferation of content skills and agents as AI has made it so easy to produce them. Glean's AI gateway makes your AI tools think alike and work consistently. And how do you bring the right content skills and sub aents to your agents? [00:19:42] That's where our core search expertise come into play. Glean identifies the highest quality content, skills, memories based on human and successful agent usage and solves that problem. Search is more important in AI than it has ever been. This year, Glean also evolved into becoming an enterprise AI platform. In fact, 78% of our customers now access glean through APIs and MCP. That means we are making cloud code, Microsoft copilot, chat GPD, all of those all of these tools better with Glean AI front doors. Even our competitors, they're all getting better because of Glean. While we're talking about the Glean's AI [00:20:35] gateway, I want to highlight highlight two more capabilities that it has. First is our memory platform. Glean learns about your employees, how they work with AI, their working preferences by looking at their activity across all of your enterprise systems. It also learns how agents compound their intelligence over time with that human feedback. This memory is is a key asset and it is available to you as a platform that you can leverage in all of your AI tools and agents even as your AI tooling actually changes. This is AI sovereignty. You being in control of your own AI. While we're talking about that, Glean also brings all closed and open- source [00:21:27] models via this AI gateway so that you can choose the models that meet your data retention and other sovereignty requirements. Well, there's one more important issue that we need to address today, probably the biggest one, which is token costs. Solving for soaring costs start first with model choice, of course. Our goal is to bring the best of all AI innovation happening in the industry to our customers but at the cost that they can afford. To do this, Glean aggregates intelligence across all open and closed source models. Most systems still point the most expensive models at the simplest of the tasks. That's not really a limitation of [00:22:17] AI. It's an architecture choice. Glean doesn't want you to be spending tokens where you don't need to spend them. But choosing the right model for a given task is not an easy problem. It requires you to fundamentally understand how models perform across the huge range of enterprise work. Glean has built a sophisticated evaluation and training framework on real world enterprise data uh with all the customers that we've been working with which has helped us build a really effective model routing technology which we use you know every day in our products today. We released a benchmark today showing that Glean saves 81% on token costs [00:23:08] compared to cloud co-work with green response preferred 78% of the time. Context produces higher quality results and it delivers token efficiency. That's because quality context stops AI from having to reconstruct it at runtime, which burns tokens and hands you back answers you can't use. Now to a very important topic. 2026 is the year of open source in AI. And I'm not I'm not just talking about models which as we discussed are allowing us to optimize AI costs. They're really getting as good as close close models. I'm also talking about harnesses. Open- source harnesses often beat closed ones in benchmarks. And they're allowing us to create [00:24:07] amazing AI products very quickly. And one of these products is Clean Tow, a new desktop app that does everything from codegen to general co-working. It's named after one of its creators. We took an open-source uh coding harness, combined it with the Glean's context platform and AI gateway and built this incredible product which now majority of our company internally is already using as a primary code gen and co-working product. Alan is going to talk about clean more later. To conclude, clean is an enterprise AI platform that truly understands how work gets done. Now, let me invite Amrajan, our head of product, to show you the [00:24:57] products that we're building on top of this context foundation. Thank you. [music] Welcome everyone. I'm John and I'm thrilled to be here with you all today to cover what we have been building at clean. Let's get started, shall we? AI has made our lives easier. It has made such that we all can accomplish a lot more work from my lance. As a product builder, we are able to ship more features faster. But our human attention hasn't really changed. We all have a million tasks on our plate, a rapidly growing number of AI features, AI tools available to us. And yet, in our moment of need, when we actually need AI to help, we are left [00:26:00] not knowing what feature to turn to and what job it solves. So if human attention is now the bottleneck, then we have to meet the user the moment they need it and help them get the most value out of AI. This is AI that works unprompted. Today I'm excited to show you what the future of work looks like with AI that is proactive. We are announcing proactive AI that first prioritizes and executes work with task management, email triage, meeting coach. Second, informs business decisions with interactive dashboards. Third, coordinates and executes work before you even ask with independent agents. Fourth, improves over time with memory [00:26:57] and skills that learn and selfheal continuously. And we are moving AI from single player to multiplayer. >> Lovely. [applause] Thank you. The team chat product a shared space for teams to build and collaborate with AI. Let me invite Selene so that we can show you some of what we have been building. In this demo, we will show how a fictional HR SAS company might use Glean to surface renewal risks, manage customer escalations, and bring teams, AI, and agents together to resolve critical business issues. Interactive dashboards are our newest artifact. provides a dynamic sharable view combining business data and insights [00:27:57] letting users drill into what matters. Let's begin in Glean. I'm going to open up an interactive dashboard. This was in this one is focused on renewal health providing a live view into all of our renewals. Before Glean I would be limited to only seeing CRM data. Now we combine structured data from our CRM, usage telemetry from our warehouse, combine it with unstructured customer data, all in one view. What we see here is that Frasier Automotives renewal is red, putting revenue at risk. Glean proactively identified this risk and shows that they have been impacted by an API failure right during the performance reviews. [00:28:44] Now, we need to make sure our teams are aware and act quickly to save this renewal. Team chat is that cool new feature that lets teammates and [snorts] Glean proactively work together. So, collaboration builds on shared context. Instead of copying it between tools in a few clicks, I add the AE I add the CSM into a shared thread. I take in the AE to understand what happened. The AE asks Glean to find out more and asks for help to draft a communication plan. Pay attention here. This might appear simple but represents a huge unlock for your business. To date, AI has been predominantly single player. One person going back and forth with AI alone. [00:29:37] Team chat is what multiplayer AI looks like. teammates working together with AI to drive better customer outcomes. The CSM asks Gleen to update the next steps in our interactive dashboard to keep everyone aligned on what's happening without any manual work. So we saw how Gleen is providing a living view, living understanding into which renewals are at risk and we have a strategy to save this renewal. Now let's see how our engineering team is collaborating with an independent agent to fix the API failure directly in Slack. Independent agents are longunning agents with their own distinct identity, memory and context. [00:30:31] They can be trusted to act proactively and they can coordinate work across teams. But they also have governance in place so that they don't go off the rails. Let's see this in action. In Slackle, Glean's independent on call agent is working to resolve that very same critical API failure we looked at alongside the engineering team. The agent is proactively identifying the escalation, analyzing the issue, looking through the code, opening a pull request, and taking the right engineers to fix the issue collaboratively. Glean superior enterprise context is what makes this possible. This agent learns how to proactively fix API issues [00:31:21] because of its memory. So, it can use that deep knowledge to resolve future issues even faster. This is AI not doing everything but making sure everything gets done. Now that we have a fix from engineering, let's see how we can now communicate this back to our customer. Proactive task management. It uses enterprise context to surface what needs your attention. It prioritizes it and prepares the next step before you even ask. Let's go back to Glean assistant. It looks like Gle has prepared something for me here. Once I click on this task, Gleen has already built the first draft of the slide deck for our upcoming [00:32:12] customer meeting. It is even bringing that context from engineering about the root cause and the resolution of the API issue. the resolution we just shipped. Again, this feels really easy because it is. This is how context unlocks better productivity for the enterprise, faster, higher quality execution and definitely not bots sitting. What Lee and I just woke you through is how Gleen delivers proactive AI interactive dashboards that help leaders get a full view of the business and coordinate work across all levels of the enterprise. Independent agents coordinate work across teams to resolve issues faster. Proactive task management gets work done [00:33:01] on your behalf and team chat shows what is possible with multiplayer AI. These new capabilities are unlocked all with Glean's powerful enterprise context. Now I want to pivot a bit and show you how that very same context can identify which processes to transform. Our forward deployed engineers work closely with our customers on their AI transformation journeys and strategies and our customers tell us the same thing. Enterprises need a reliable real time view of how work actually gets done in their environment. Transformation starts with that visibility knowing what actually needs transforming. And to date that visibility has been rather [00:33:55] fragmented and still leaving immense opportunities completely untouched by AI. With the enterprise graph, Gleen already has that visibility and we now can identify which workflows are the strongest candidates for that transformation. Today we are announcing glean transform. This is a living representation of how work gets done in a company. Let's take a deeper look. We'll show how we bring AI transformation to our teams at Gle. Here, Gleen has already transformed 78 workflows and highlighted 272 opportunities across all departments. But let's actually dig into the sales department. In green transform you can see there are 38 opportunities ripe for [00:34:52] transformation just within our sales team. One click glean highlights how our sales team gets work done along with a prioritized list of transformation opportunities. Let's dig into one of them our discovery call and use case mapping process. Here we can see specific recommendations for each step of the workflow mined from the enterprise graph. In the account research step, you can see that our sales development reps are constantly context switching between different tools like Gmail, Google Drive and it is slowing down progress. Glean transform understands this work at a granular level. So we can see how teams [00:35:42] gather account information and the shared context and why this is worth automating. Transform now has built a blueprint for a call research agent. In one click I can see that agent that we have built. Now this agent includes all of the correct instructions, tools, knowledge, other agent config details so that we can set up that automation with one click. And because green transform understands the way that work really got work really gets done in my environment. I have confidence that the agent will fit into my process and will make a real impact. But it is one thing to feel like an agent is having an impact. It is another [00:36:32] thing to prove you have made a difference for your business. If we go back to the main transform dashboard now, you all can see how Glean transform measures AI impact across our engineering workflows. Our customer escalation workflow has become 46% faster. For example, one click deeper and you can now see how that happened, why that exactly happened. With the new diagnostic agent that we deployed, you can see that we are resolving escalations faster, assigning tickets more accurately, saving 40 minutes per ticket. So, transform enables us to see all of that impact in one place. Transformation is not a one-time initiative. It's a constant evolution [00:37:22] opening up new possibilities to reimagine how work gets done in your environment in your enterprise. Now let me turn it over to Ammer and Philip to talk about how GM is transforming their business with Gle. Thank you. [music] Hello everyone. I'm Amar Malitira. I'm the chief operating officer at Glean and uh today with me is Phil Lutka, executive director for core IT at General Motors. Um thank you Phil and thanks for being here and u you know GM is driving an incredible transformation with AI. So let's start with a with a big picture strategic question here. Okay. So what strategic problem was General Motors trying to solve for when [00:38:17] you decided to invest in Glean? And uh why did Glean become a a platform of choice for that challenge? >> It's a good question to start with. Uh first I'm just so thankful to be here and just in this incredible energy. But for GM the the problem we were really trying to solve was an access issue. Uh we needed to have a platform that could help us find all of this vast knowledge that exists within our enterprise corpus. GM's a big company, thousands of employees, tens of thousands of employees, tons of beus, just so much knowledge, 100 years of history and finding the relevant important information. It's hard and Glean has sort of bulldozed [00:38:57] that challenge to the ground and made it very easy to access. The second piece was we needed a product that we could grow with and learn from and one that we felt we would get a good partnership from and I think we've seen that for the last eight months now with Glean is we're building together, we're learning together and we're delivering just incredible value to the business. So those are the core areas that were really part of that initial decision. Uh and and I think the last sort of key piece was like we wanted a great user experience like this needed to feel clean, modern, easy. Uh and and that certainly is what we found [00:39:32] with clean. >> That is fantastic. So talking about user experience, you know, one of the things I noticed was um the incredible pace at which you guys drove adoption, right? So they went from zero to 60,000 users. >> Yeah. >> In just 90 days. >> Um >> and u and you know, so how did [applause] That was just amazing, right? Um there's a lot of lessons to be learned on how you guys drove it. So how did GM really plan for validate and roll it out so that every single employee could you know use it as opposed to what we generally see in most of the companies is you know the technical teams the early adopters using it. So [00:40:12] >> give us some thoughts on it. How did you do it? >> Um I mean first I'm just like really proud of this accomplishment. This is of all of the IT projects that I've done in my career, Glean has been the highlight, the most like impactful, iconic, and like successful, which is a pretty Thank you. Yeah, it's a big deal. We um it took us a lot of work. Like this was not like a we'll just turn it on and make it easy for everyone. Obviously, we started with a product that just was easy to use, but we came at it with our customer and our end user in mind. So we really were curious. We we wanted to understand the problems they were trying to solve. [00:40:50] What's the friction that exists in everyone's daily life? How do we, you know, help people find the person that knows how our, you know, battery chemistry works or a 10-speed transmission works? How do we help them find a relevant like current document and all of those sort of scenarios? And we also wanted to bring it to folks in the sense of we had a lot of employees, thousands that had really not used AI before. And we had another huge population that had and was ready to get cooking, building agents and all that fun stuff. And so we we really met people where they were in that journey and just spent a tremendous amount of [00:41:27] energy and calories on enablement training, joining all people meetings, staff meetings and just again like leading with curiosity like what problems can we help you solve? How can we help you get comfortable through that? On top of just the reality that Glean is super easy to plum, we had it integrated into all of our systems. We made it so that like that first experience on day one through day 90 was great. And I'll say too, you know, when we talk about 60,000 users adopted, like we're pretty serious about that. We we sort of look at the typical metrics, but 5D30 is where we anchor. And like I looked this morning, we're still at over [00:42:05] 80% 5D30 for our whole organization. So it's it's real adoption like sustained continued use throughout the day, the month, and the week. Yeah, >> that's fantastic. So you know one of the questions every single enterprises around the world they're grappling with is how do you measure the impact of AI >> and so how are you guys doing it at uh at General Motors and more importantly um if you could give us one glean use case that had sort of material impact on your business. There are there are a lot of use cases so it's it'll be tough to pick but we do the baseline measurements obviously adoption we talked about we [00:42:40] look at sort of the impact that it's having ease of use is it finding the right relevant information sort of all the great tools that glean provides us we we leverage and then on top of that we really focus on impact like and not just like did we save a bunch of time but like did this really enable the next great idea a faster decision or just a a better outcome altogether for the business. Nquille in room 11 at 11 o'clock. You should join. It's going to be a great session. Is going to talk about our uh DR agent which is the design release engineer agent that we built in partnership with Glean's uh for deployed [00:43:16] engineering team. Uh, and it effectively takes this like massive amount of like issue, warranty, like telemetry, like engineering, mathad, all of this data across just a ton of disparate systems, and it brings it into one clean user experience, and it helps like our 6,000ish uh, design release engineers bring all of that context together, solve problems quicker, get to customer outcomes faster, and really just deliver good value. uh and like improve their overall sort of enjoyment with the work they're doing. They're spending less time swiveing trying to find information and actually like shipping good quality [00:43:56] outcomes. >> Fantastic. Thank you very much for that. And looking ahead uh given what you guys have already done so far, >> how is GM um thinking about continuing to innovate and stay ahead of the curve? >> Gosh, I mean everything that what we've been seeing here is is a big part of that. I think transformation is the big thing. Like we don't want to just bolt AI on to processes that maybe aren't great or maybe we don't need anymore. And so I think that's really the mindset we're driving across all of GM is like look at this problem that you have or this way we work and assume it can be completely different. And so so really [00:44:31] that transformation is a big part of it. Obviously like continuing to focus on the outcomes. You know Arvin talked about token use. we want to do it efficiently, but we also want to make it like easy to experiment and safe. Um, so those are really the the key areas that we're looking at and just really learning together with you. And I think again like that's the partnership that we appreciate so much and why we select vendors the way we do is we want to be able to build together and deliver those cool experiences. >> You know, um, Phil, we are so grateful for your partnership. I think we are going to clean is going to be in you [00:45:08] know with you in your transformation journey. We're going to learn a ton. We learned a lot so far from your transformation. So thank you for being here with us and for sharing your perspective with the audience here. >> Great. >> Thanks folks. [applause] >> Appreciate it. >> Well done. Okay. [music] All right. Good morning. Hi, everyone. Uh, this will be a fun panel. We have about 12 minutes, so we're going to go do some speed questions. U, I'm Kate Rooney. I'm a tech correspondent at CNBC covering AI, which has been not busy at all. It's been a very chill couple months. Nothing going on. Um, but thank you so much for having us, Arvin. And I [00:45:58] want these guys to introduce themselves. I was just saying Alexander long resume at OpenAI running Codeex. Um, but let's go around the horn. Everybody probably knows Arvin, but Alexander, tell us a little bit about your role at OpenAI, what your day job looks like. >> Hey folks, great to be here. Thanks for having me. Um, for the past year or so, I was leading a lot of our agent work with Codex and now I lead enterprise products for OpenAI. >> Terrific. um a tiny startup you may have heard of open AAI just um Arvin I think everybody knows Arvin you know tell but thank you for having us we'll kick off with you um one of the big topics we [00:46:35] want to talk about is agents we were talking a little bit about this backstage and this this move from reactive to proactive and I think a lot of people myself included want to live in a world where AI is a little more proactive and can do a little bit more for It seems like there's a gap still. I wonder what you hear from your partners, from customers, and what needs to change in order to bridge the gap, make AI a little bit more proactive and a little bit more useful, I think, for the average company, the average executive, and just the average Joe like me who wants it to do more. >> Yeah. I mean this is this is absolutely [00:47:14] like you know we believed in I think AI can make true impact when we don't have to go and direct it all the time and give it tasks that can do for us because what we've seen what we've noticed you know even at the the most advanced technology companies in the world there are people are barely using AI to its full potential um because they just simply don't know what AI is capable of doing for them so we always thought about this like you know why why is it like that why can't AI actually come to you um and figure out like you know who you are um and what you need to get done and actually just start to sort of [00:47:52] propose and actually do those things for you that sort of will completely change the dynamics of adoption and you will see a lot more like you know AI sort of helping everybody and and and the adoption sort of becoming more uniform inside inside your enterprise but that's so that's the trick though like you know what's missing is to make AI proactive Ive think about personal co-working if AI knew everything about me as an individual what's my what my role is what do I do in the company what are my OKRs you know for the next 3 months what are the weekly tasks I need to get done and it also kind of knows like how I [00:48:29] like to you know do that work how I how I write how I read with that full context it can then actually come and figure out that these are the priorities for me these are the things I need to get done >> and and actually you know propose and get those things done for me. So that's that's what it needs. So it needs you know that deep context, deep understanding of your people and your business >> to to actually become proactive. >> Alexander, I wonder what it looks like at your day job. You have conversations, I'm assuming, with Fortune 500 companies. OpenAI works with some of the biggest companies in the world. what are [00:49:04] they saying about adoption about AI agents and this conversation about maybe making them a little bit more proactive? >> Yeah, so I actually kind of want to start with the risky move which is pulling some people here. >> Ah okay. >> Um but you know I totally agree with what you were saying about how AI is just it requires so much thoughtfulness. You know I like to joke that it's sort of a little bit bottlenecked by our own creativity and even typing speed. So raise your hand if you think here you are running enough loops in parallel in the background. Okay. Raise your hand if you feel slightly bad about that and you're like, [00:49:37] "Oh, I should be running more loops." >> Yeah. Okay. Okay. Thank you for reacting team. >> Um yeah. So like we need to get if with with the right context agents can be much more powerful. And then the thing I would add to that is the form factor is really meaningful, >> right? Like right now we've built a lot of tooling and tooling is great. We can achieve mastery over tooling and we can make it awesome for ourselves. But that puts a lot of onus on us. And so a lot of what I hear from enterprise leaders when I'm talking to them about AI is they want to know, hey, how can I reduce the amount of change management required [00:50:09] for AI to be helpful for me? How can I make it so that I don't have to train every single person in my company how to prompt the right way, how to manually inject the right context, and how to even know that AI could help for a task. So we think a lot about okay, how do you get the right context by default? We think about what is the right form factor. You know, you use a coding agent differently than you use a product that's shaped like automations, which you use even more differently than you talk to a teammate, right? Where it's like really natural. If we were colleagues, I could delegate to you or you might proactively suggest an idea to [00:50:37] me. >> So, this is kind of the journey we're trying to take people along. Starting from, okay, we all know how to use, you know, chatbt, we know how to use codeex, we know how to use like glean just like search manually. How do we then build products such that people at a company can sort of like by default get value from deeper AI workflows? >> One of the things you mentioned Arvin was sort of the context. I think about security is such a big issue and that tradeoff between at least as an average consumer. Yes, you might want to give AI more information so that it can make better decisions. But you're also worried about personal security, cyber [00:51:11] security for companies. How do you think about that entire topic and keeping companies data safe? >> Yeah. Well, so first I I think security has so many different um areas, you know, where we need to actually do a lot of work in. When you think about agents, agents can run either on behalf of me um or they can have their own identity. If they're running on behalf of me, these agents should not be given access to enterprise data and information that I did not have rights to see or use or for the same on the same token they should not have the ability to make any changes into the system that I could not make. Mhm. [00:51:54] >> So that's that's the basic you know that's that's the first thing that you have to think about uh as you build your agent infrastructure is making sure that you're properly restricting your agents to things that they should be able to do. Um but also when an agent is running on my behalf, I actually have a lot of powers like as an employee I can go and change you know uh things in many different systems but an agent is probably has been built with a specific um with a specific sort of set of tasks that supposed to do. So you to sort of further go and restrict these agents like you have to think about like you [00:52:29] know you know the the the range and the scope of task that it's supposed to do and and and accordingly restrict you know their permissions to those kind of actions. So these are these are some of the basic things that you have to actually get you know do to make sure that agents suddenly don't start to actually create problems inside your enterprise. But then the the new thing you know which I'm actually you know we're really excited about is this concept of independent agents. So these are agents that are not running on my behalf. You know they are treat them more like employees or colleagues of yours. Um say you know for example take [00:53:04] your customer support team maybe you have 10 people in the team today and now this agent is the 11th member of that team. So when it is like that when the agents have their own identity you have to again um think about like you know what is the right permissions model the security model for these agents what information can they see what they can't and what's the again the guardrails in which they should operate so that you know as a team member they know they're nice to people they're actually adding positive value and and and and not actually escalating and you know fighting with you know with the humans inside. So these are these are like you [00:53:41] know interesting you know topics that now we have to actually solve for as agents become more and more independent. >> It seems like a lot needs to happen behind the scenes to to take out some of the friction if that's you need to have certain permissions. I would think that's probably one of the challenges too. I wonder how you see it at OpenAI. Um and also if you want to respond to any of Arvin's commentary on it. >> No, I agree with that. Um I think we're about to see the rise of maybe two big shifts are going to happen, right? independent agents and agents execution shifting to the cloud off our devices which which will be way more secure and [00:54:13] a lot of our enterprise customers are very excited about. I do have to pause for a beat though and say if we're talking about security and cyber a lot of our customer conversations today start with cyber one of the things they're most interesting in is just like how to prepare for the new sort of landscape cyber landscape. Um and we obviously have great cyber models but coming back to to running agents in enterprise we kind of see there's a few layers and tackling them sort of incrementally seems to be one of the best ways to go about it. So you know one of the layers is um just control like how do you decide what agents can [00:54:44] do what they cannot do and interestingly enough this is like somewhat easy by default because you know over time you want to connect the agents to all your context so they can actually be incredibly useful but when you start they're not connected to anything right >> um and so we've actually seen that with a lot of our customers just starting with like the most important pieces of context or tools to connect them to and incrementally building your way out might not be exactly what your CEO or the average employee wants But your security team will be very happy if you add one and then you add the other one. You can actually move really fast if you [00:55:12] do it incrementally. So okay, so you have to like lock down like what you know what configuration, what permissions the agent have. The next side of it is actually observability, right? You can be much more confident in giving your agents access to things if you have really good understanding of like what they're doing. Um and any any manager or IT admin can go in and say like, okay, here are the different actions this agent took. I like that. That's great. Let me actually optimize that or let me lock this thing down so it can't happen in the future. So basically pairing these two ideas um or maybe three. One is the control of what [00:55:43] the agent could do. The second is the observability of that. And then lastly starting to shift away from agents always impersonating us as people and having more agents that have a specific role with exactly the right permission set for that workflow and that's an independent agent. >> So is that like an agent has an avatar and is its own employee? What does that actually look like? >> This is a fun one. I'd be curious what you think. We talk a lot about like embody not embody. I think the answer is both, right? Like some things you just want an automation. Like every time someone writes code, we want a code review to happen, you know, that doesn't [00:56:13] need to be an embodied thing. >> How do you think about it, Arvin, too? >> Well, I think with independent agents, I like the concept of them sort of like having a human-like identity, like, you know, an avatar and and and a name. uh because because I think the the when you do those things, you know, of course they're gimmicky in some way, but but when you do it that way, that's when people will start to truly treat them like a a real teammate of theirs and it'll unlock different types of collaboration with agents. >> I wonder, we're running out of time, but big picture implications for companies or investors in the audience, what are [00:56:48] the economic implications? is what are the implications for productivity, profitability for these companies if there is actually a world where companies can harness this well and sort of close that gap that we talked about in the beginning. Arvin, maybe you want to take this and Alex. >> Well, I mean like AI of course promises um significant productivity and fundamentally um in the long run we're going to see work is going to change. you know, we're not going to be doing most of the things that we do today anymore. Like, you know, that we're going to have these amazing agents and assistants and co-workers that actually help us, you [00:57:22] know, do all of this work. So, the economic impact is like it's very hard to actually even, you know, measure like, you know, it's so so huge, so big. We just have to make sure that we you know are getting there in a way that we actually can like today like unfortunately what's happening is AI is often becoming more expensive than the business value that's that is delivering in short in the short order like you know so that's that's the challenge that we have to solve long terms there's no confusion this technology is going to be great it's going to create a lot of economic value uh we we we need to make sure companies don't run out of money [00:57:57] before that happens >> we didn't even talk about token maxing I can't believe it didn't come up. Um, >> value maxing, not token. >> Well, last word, Alexander. Uh, how do you think about this and how should companies and some of Glean's partners, investors maybe even who might be in the audience think about the implications and how to sort of prepare their businesses. >> I think individuals and teams are able now to be vastly more productive and that is only going to increase. But there is now going to be and and by the way, this is not a new thing. I think since since the history of our species, we've created tools and tools always [00:58:30] mean that someone who is good at using tools is vastly more productive than someone who is not. So we've seen this problem before, but we are now going to enter a situation where the most productive individuals could be running unicorn companies and folks who are less productive are well where they are today, right? And so I think the most important implication and takeaway is just to have really high agency. Make sure you're hiring really high agency people and think about how to structure your team to unlock the creativity of those high agency people and then scale that to the rest of the company. It's >> fascinating. We're past the shot clock. [00:59:01] This flew by. Uh but thank you all for listening and tuning in. Norvin, thanks for having us. Alexander, thanks for being here. I'm going to welcome Tony up next. Uh let's hear it for Tony. Thank you. Thanks, guys. Good morning. I'll start with something structural about Glean. Our business model gets better when your tokens do more work, not when you burn more of them. A lot of vendors are built the other way around. Arjan showed you a few minutes ago what happens when proactive takes on more of the work. But that only pencils out if you're spending AI on tasks that actually need it, matched to the right level of [00:59:53] intelligence. Tokens are money. With unlimited tokens, we could all do millions of things, but everything has a cost. So our goal is simple. Drive down the cost of AI doing work and then dramatically many more things become worthwhile doing with AI. token yield and what I'll talk about today comes down to three levers. Model choice, auto routing, and AI usage controls. Now, I'm going to invite one of the smartest engineers I know, Eddie, to the stage to talk to me about our evals and our latest benchmark results. Eddie, >> hey, Tony. So, yeah, let's start with the first one, model choice. Um, you really can't optimize anything [01:00:45] with token spend without without having a real model choice. That's why Glean gives you over 40 open and Frontier models and why we work so hard to make sure literally every other week as they drop uh that we're getting that day zero support. >> Yeah, definitely. Uh, speaking of models, uh, you know, this is the year that we think open models really caught up. uh mostly because we stopped measuring them on maximum intelligence and instead measuring them on cost performance and on that curve open models have closed the gap. >> Yeah, you said curve. Uh I want to take a look at this. This is called the Pareto frontier. Um not not potato [01:01:21] frontier. I thought that uh the first time I heard it. Pareto frontier. So what is this? This uh this this is a model that shows us price performance. So a a a model is Pareto optimal um if nothing else beats it on performance at the same cost or matches its performance for less. So we see cheaper less capable models in the lower left quadrant and we see the more expensive more capable models in the right quadrant. And when we talk about the frontier that's that curve that traces the best option at every cost point in between them. >> Yeah. So this chart you guys are looking at we built on enterprise data capturing [01:02:01] the variety of tasks and different works across departments that are in the enterprise today. So it's important to note you know these are our results on our benchmark. So they involve you know our query set composition. They do convey these models across the frontier but they're conditioned on the choice of harness as well as the evals in the environment. So if you guys look closely in this setting GPT 5.6 6 Luna does dominate a big part of the frontier given its incredible bang for buck. Um in the middle we see an appearance from open models like Kimmy K3 GLM52 and actually as of this morning you know we haven't [01:02:38] recreated this plot yet but that ox alpha model from the weekend is known to be GLM53 flash. Now I have a strong suspicion it's going to be at the frontier here as well. Uh and then on the top right there we have Opus 5 giving us the strongest intelligence for those who are willing to pay for it. And if you did go back about a month, the graph would look pretty different, right? So Luna would be at a different cost point. So it might not be shaped exactly like this because OpenAI hadn't done that awesome big price cut. Uh Kimmy K3 didn't exist yet. So it hadn't shipped and it wouldn't be So those two wouldn't be these sort of intelligent [01:03:14] affordable options. >> Yeah, that's really the thing. I mean, we just made this slide, right? And it's already it's already obsolete. So it the frontier doesn't hold still. And if you're locked into one provider, if you're looking at one ecosystem, you're not taking advantage of every cost savings or every intelligence opportunity that comes up simply cannot bet on one provider here. >> Yeah. So in our data, the same same pattern can kind of play out task by task. So over here we have information seeking tasks and you can see for those um a lot of them get served by pretty inexpensive models. Uh and you can see [01:03:49] that in the the sort of slope of the chart being pretty flat. Uh in contrast for coding uh you see the opposite that the slope is a little bit steeper and you can see that performance jump as you move up to those frontier open models as well as opus 5. Um and this is sort of in contrast to that lower left left right. So this kind of analysis it makes the case for auto routing uh not just by you know tracing that overall paro curve but also task by task. >> Yeah. And we we don't just like run all these eval benchmarks so we can put a pretty slide up on screen. These eval are what we use to guide our auto routing decisions, which brings us to [01:04:27] that second big lever that I talked about. So today we're announcing auto routing in Glean. And it played a key role in Glean saving 81% on token costs versus claude co-work a benchmark that we also released this morning. [applause] Thank you. Yeah. Uh let me put that 81% in perspective here. So that means we're talking about 58 per task with Glean versus $2.98 with Cloud Co-Work on a similar task. Yeah. So Glean's advantage comes from routing across those, you know, 40 plus available models, picking the right model at the right reasoning level for the given task. So for example uh opus 5 we know we found excels at visual tastes [01:05:20] and visual artifact creation. So for those kinds of tasks we route to it there. Um but it's not just choosing the model right it's also deciding how hard the model should think. So for simpler tasks we might use 56 Luna but on low reasoning rather than high reasoning. Um and to make that all happen at runtime we've postrained our own specialist model Glean Waldo. um that's post- trained on strong open- source base models such as the Neatron family uh and that model's capable of routing to the right uh reasoning level on demand. Yeah. And you know as an engineer as a product builder the part I really love is we didn't we didn't win that [01:05:58] benchmark on on just this this uh choice the price per token alone. We actually use 70% fewer tokens. Can you tell us why? >> Yeah, totally. All these things kind of compound for token efficiency, right? So it comes back to our roots in search. Um with search we have a very purposeful choice to premputee a central index build this enterprise graph that has all these signals and that way the model can much more efficiently find that relevant context at runtime instead of sort of building it up incrementally wasting a lot of loops collecting it. Right? So cloud co-work and other harnesses other setups might take a different approach [01:06:34] when they rely on federated MCPs. um those services will end up searching live across a lot of sources and they all those different sources have different search qualities. They're not reconciled. All that data then needs to be dduped, filtered, joined all while your tokens are burning, right? So that's a pretty token intensive operation. Uh and it shows up in this token difference. >> Yeah. And that's that's the cost side, right? But uh token yield is about pairing costs and quality that balance. Um, cheap, cheap and wrong would be worthless, right? So, on quality, Glean was actually preferred 78% of the time [01:07:12] over Claude co-work. Thanks for that. [laughter] And the cool thing is that preference showed up showed up across every individual department as well too. >> Yeah. So, what these results come back to is the question of usability, right? How useful is this output in my actual work? Can I trust it? Is it correct? Is it complete? And a big reason that glean was preferred comes back to our search foundations. So it allows us to pair highquality context with the right model for the task. Not always the most expensive model. And that gives us significant gains in both quality and cost together. >> Yeah. Um and we actually got we got so [01:07:55] excited about auto routing here that we made a quick video about it. So let's roll that. Heat. Heat. [music] >> [music] [music] [music] [music] >> And I said three things. So, you know, I can't leave this stage yet. Good, good, good, good applause for the little video. We got carried away, but I can't leave the stage yet without talking about AI usage controls. We're of course building the caps that enterprise need, but caps alone aren't a strategy. Picture this. Your best engineer hits their limit on the 20th of the month. What are you going to do? Send them home for the rest of the month or just raise it? The goal here is to [01:09:22] have the visibility and the levers that let you coach this usage instead of just capping it. So auto routing is spend aware. It knows whether you're on track for the month or burning fast enough that costs need raining in. I think of it like my phone battery. When the juice gets low, it goes into low power mode. Similarly for models, we move to more costefficient options instead of just cutting users off. Admins can set these profiles too on a per department basis um between costs efficient balance or frontier level intelligence. And not everything just comes through the system. Some of this you have to coach through process as well too. So [01:10:07] we're shipping uh self-s serve analytics over your AI usage data. Then you can know why is a user over budget. Uh you can ask questions like who needs a nudge off of Opus 5? uh whose agents are stuck running that somebody forgot about and they're just continuing in the background. So caps here are really the backs stop that help you avoid getting a surprise bill. Um but in order to really coach that we needed to add spend forecasting. So this lets you see where usage is trending and budget with confidence. model choice, auto routing, and AI usage controls. These all fall under the new glean intelligence umbrella. Delivering [01:10:53] the right intelligence for the right task, for the right token yield for your business. Now, let's hear Michael's take on our token costs. So, we've learned a lot about tokens in the last year. My name is Michael and I lead finance at for the last 3 years, we've been telling everyone the same thing. Use AI, use a lot of it, token max, build with it, put it into your workflow, and guess what? It [music] worked. And they actually did use AI. In fact, they used a lot of it. And token costs went up and the CFO took notice. And then we [music] entered the token whiplash. Now, we're all asking, how do we control the costs without killing the [01:11:34] behavior we work so hard to create? [music] And the answer is in model choice, different models for different tasks. Because right now, [music] too many employees are using Frontier models to do non-frontier work. That means tasks that are simple are getting handed to a really expensive Frontier model. Using a Frontier model for a non-fronttered [music] task is like asking your CTO every time you need to reset a password. >> Why is everyone asking me to reset their password? >> And also speaking for CFOs, they aren't the villain here. And this isn't less [music] use of AI. It's just more responsible use of AI and smarter AI [01:12:07] economics. Tokens are just dollars in another currency. Giving every employee unconstrained tokens is like giving every employee a credit card with no spin limit. That would be insane. Limits exist for a reason. They build responsible AI use. And they help enterprises stay in control. Give people access to the right model for the right task. Frontier models when they need it, small models when they don't. Because if AI is going to scale across the enterprise, every token has to earn its keep. I love that video. >> Now, please keep the applause going and help me welcome uh bring out in the stage here Carrie from Nvidia who's [01:12:48] going to talk to us more about open models and the AI ecosystem. >> Hey there, >> Harry. Thank you for joining us. I'm looking forward to your your insights here. Yeah. >> So, yeah, it' be it'd be a real it would be an understatement to say that uh there's a lot of discussion on open models on an open ecosystem. Um tell me about why this is so important to the broader ecosystem. >> Well, hi everyone. Um I would say that open models help the entire ecosystem. We need open models now more than ever. um for reasons from uh you know researchers to startups to enterprises. Open models allow them to uh be able to evaluate, understand what's inside, [01:13:38] customize, guard and inspect. And so researchers, you know, have an ability to uh learn and innovate. And um startups have this foundation that they can build on top of. And we saw a video about token efficiency. And then enterprises, you know, really are coming around to the idea that they need to own their AI instead of rent it out. And so being able to take their IP and customize models and then run it even more costefficient efficiently. >> Yeah, those I I love those points. Uh it makes total sense. Um at Gleam, we're definitely big believers in open models as well, too. This is why you know our Waldo like Eddie was saying is trained [01:14:20] on Neotron 3 Nano and why of course we offer Ultra as well too in the product. Um and I let me see so like I know you're continuing to do a lot of work put a lot of things into open source. Um can you tell us any more about open source and NVIDIA and what you're doing there? >> Well what what one thing that most people don't know is that NVIDIA is one of the largest if not the largest contributor to open source AI out there today. Um, and so that's something that's really exciting. We put out not just the open weights of the models. We put out all the data sets that we train those models with. We put out the, uh, [01:14:56] open source of the frameworks, the exact frameworks and recipes that we use to train those models so that everyone can pick them up and either recreate or create their own data blends. And so, uh, so that's something I'm really excited about. And, and so now that we've done that, we've seen such great engagement from from the community. And so builders from sovereign AI uh you know we have neighbor in in Korea, Sarbomb in India uh you know we have Sufi in Germany and then so that's sovereign AI but then also in enterprises and startups so we have code rabbit uh who is doing code reviews uh we have um perplexity doing AI search [01:15:32] and then of course you mentioned Glean Waldo who is uh using Nvidia's Neotron models. We use it personally at NVIDIA. I use Glean every day with our Neatron Ultra model which allows us to have lower latency and lower uh you know tokens too. >> I get so excited uh hearing about uh people using things that that that we build and that we work on here. Maybe you could share a little bit more about how you use Glean at NVIDIA. >> Yeah, so uh Glean is in this the back end of our IT systems. It connects to you know I think it's like 20 plus enterprise tools. Um uh you know we have 90% of Invidians where I think we're [01:16:08] something like 42,000 employees. like 90% of Nvidians use Glean every single day for this IT search. It was like one of the highest ranked of an internal IT survey. Yeah. And then we also [laughter] uh and what's more exciting too is that we allow our employees to create their own agents. Um and so we have over 400 agents uh running in Slack based off of glean and grounded in in factual information from these systems. Yeah. >> Yeah. I mean you and your team at NVIDIA you have you have the best engineers the best agent builders in the world. Uh tell me like what did you find valuable about building agents with glean? [01:16:45] >> Yeah well I think that like rag is only one part of the challenge when you're building agents. I mean agents are the entire system. I mean their models their harnesses their factual grounded information. Uh but then when you put this all together you still need uh I mentioned connectors but you still need permissions governance security and compliance. And uh you know our our engineers we need to focus on our core competency and so uh so we want we need trusted tools in the back end to do that and so Glee provides that that enterprise ready platform for us. >> Yeah. Uh maybe maybe last question a wild card like so favorite favorite [01:17:19] query or favorite agent uh with Glean? >> Uh well we definitely we have this thing called Envy Info and Envy Bugs and so so I'd say those compete for my top two in the back end. And so what I love is uh I first of all, if you know me, my favorite thing is I love bugs because I say it's love letters to my products. And so when I'm able to to go in and say, you know, what is the bug that's happening on on this product? What's going wrong and why? So that's that's probably my favorite. Yeah. >> Yeah. Oh, nice. Okay. Yeah, that that that resonates uh you know, everything that you're saying, we're seeing that a lot across our our user base as well, [01:17:52] too. Uh Carrie, thank you so much for your partnership on this on on on this journey. I look at like every one of our customers as partners and this is this is amazing working with you and I appreciate you sharing your insights. >> Thank you for having me. Cong congratulations. [applause] >> Hi everyone. I'm Evan Spiegel and thanks so much for having me at Glean Go. I wanted to share a quick message about how we're thinking about AI at Snap and why Glean has become such an important partner for us on this journey and helped transform the way that we operate. I've absolutely loved meeting Arvin and learning from the Glean team. [01:18:28] First of all, across Snap, Glean is helping our team move faster and accomplish more. We've spun up an internal program with folks from across the company who share new Glean agents every month. That's helped the best ideas spread really quickly because you can see what someone in another part of Snap built and just start using it or adopting it for your own team. I want to walk through a few examples of how we're using Glean. Today, one of the most popular is a quarterly playback agent that helps with performance reviews. It pulls together relevant context about someone's work over a period of time and formats it so managers can better [01:18:59] understand what their team has been doing and give more thoughtful feedback. Glean also powers our tech lounge agent which handles a lot of common IT questions and self-s served troubleshooting. When things are more complex or sensitive, it routes them to the right person and that saves a ton of time on basic issues while actually improving response times on the hardest problems. We're also using Glean to change how we support our annual R&D tax credit work. Before, people were spending a lot of time hunting for documents and asking engineers to recreate their projects from memory. Now, we have an agent that finds and [01:19:30] organizes the relevant material, so engineers start from a much better place. The outcome is more consistent, higher quality documentation, and our engineering teams get back hundreds of hours they can spend actually building things instead of doing paperwork. Taken together, Glean has given our team a powerful foundation to build agents that are deeply embedded in our workflows while still feeling simple and intuitive for people who use them every day. Taking a step back, AI as a whole is one of the best things that's ever happened to our business. At Snap, AI is a huge accelerant, helping our teams explore many more ideas, execute more [01:20:04] efficiently, and deliver greater impact for our community, partners, and team members. For our community, our AI powered support agent answers roughly 3.9 million questions from Snapchatters each month and has reduced support ticket volume by about 62%. In advertising, AI increased First Pass image review automation from 40% in Q2 last year to nearly 90% resulting in faster approvals for advertisers, stronger content safety, and lower operating costs. For our engineers, our internal AI code reviewer now reviews 90% of poll requests across Snap and has saved an estimated 30,000 hours of code review time. More broadly, I believe AI [01:20:43] will fundamentally change our relationship with computers, which is why we've been hard at work on specs, a new kind of computer built for this future. We'll spend less time operating computers and more time supervising intelligent systems that understand context and complete work on our behalf so people can stay more present with their friends and family out in the world doing the things they love. As we build towards that vision, I want to thank everyone at Glean for helping us stay focused and move quickly. Thanks again for your time and I hope you have an incredible conference. Good morning everyone. Super happy to be [01:21:34] here. As Arvin mentioned, there has never been a more important time to have an enterprise AI platform that drives productivity, helps control costs, and delivers a secure secure AI foundation for your business. Last year, right here in Fort Mason, just a short walk or an even shorter swim if anyone's up for it, across the way, we announced Glean Protect. And think about what has changed since then. A year ago, AI wasn't accessing your data or writing back to enterprise systems. Now, agents are everywhere. A user can spin one up thousands at a time. At that scale, guarding data and actions isn't enough. We have to watch [01:22:24] for patterns. An agent with valid permissions behaving in ways that puts the organization at risk. That's AI as an insider threat. We've all seen the headlines. Models escaping guard rails, hacking other vendors. AI has access to your data and your systems, and it can use a browser to reach beyond a single system and accomplish tasks on its own. The good news is that as AI adoption has matured, Glean Protect has been in lock step with it. We've added models that detect sensitive content at 95% accuracy. We shipped restricted topic policies and alignment models that verify agent actions against the user's intent at [01:23:18] runtime. And today we are adding agent governance policies, controlling what agents can read, restricting high-risk rights, and setting access controls right at the individual agent level. But permissions and policies can only go so far. Something risky can look safe in isolation. And it's only when you see the full picture you understand what's really going on. That's why today I am super excited to announce contextaware AI threat detection. It spots high-risk agentic behavior across sessions and uses enterprise context to distinguish between legitimate activity and potential attack. Contextaware threat detection is how we fight what is [01:24:14] becoming the largest insider risk facing enterprises AI itself to see this in action. Let's welcome Blair. >> Hey everybody, it's great to be here. Looks like you've got the uh dashboard up. Let's walk through it. >> Yes. So if we take a look at this dashboard, we can see that there are four issues detected around about two days ago. And if we click here, we can take a closer look at these findings. You know, one of these items stands out. We just detected a sensitive data Xfill request. Let's open it and uh follow the activity. Here we see Alex was working on a Zenesk ticket for a customer Atlas Labs. He asked Glean assistant to reconstruct the [01:24:55] issue, look for similar cases, and draft an RCA for an internal team. Now, this initial request makes sense, right? Alex is a support engineer and so looking for previous issues to find a root cause to help the customer solve their problem is a reasonable course of action. >> Right. So Glean assistant starts with Atlas Lab support tickets. Then it finds two similar internal RCAs from other customers. Alex then prompts assistant to respond back to the customer Zenes ticket linking the newly generated RCA. >> I think I see the problem. Did Alex not review the summary in detail? >> Exactly. Alex missed that the RCA [01:25:40] included detailed information from two other customers, Northstar Health and Pioneer Bank. This sensitive data should have never reached Atlas Labs, a customer with no business seeing it. So, let's reconstruct this full chain. Glean assistant accessed sensitive information from multiple customers. It aggregated that information into an RCA containing customer names and incident details. Then it shared it too broadly. [snorts] And the enterprise graph was able to catch this issue, right? Because it understands all of the customer data and relates activity back to it. It's able to see when one customer's sensitive [01:26:22] information crosses into another's boundary. And not only that, but it understands the work that Alex does. And so it knows that he should only be sharing data that's related to Atlas Labs, not any other customer that he doesn't work with. Every step appeared business as usual. The assistant and Alex had valid permissions. There was no malicious intent here. But when we connected these individual actions to the overall context of the situation, we accurately detected a high confidence exfiltration. Thanks for the demo, Blair. [applause] Now, we've talked a lot about how the AI landscape has changed with more models, [01:27:11] more front doors, and employees reaching for their favorite AI apps. That's why we are announcing Glean's AI gateway, a unified control plane for LLMs and MCPs that secures and governs models, context, skills, memories across every front door. With it, your apps also inherit the efficiencies and cost controls that Tony walked through earlier. And today, we are announcing that AI gateway also makes memories and skills portable across apps. Glean is aggregating memory from across all your contexts, chats, agents, other AI apps to make it sharable and usable everywhere you work. Now, that's a big deal because right [01:27:57] now, memory is locking you into a single app. That's memory that you built and it's become just another data silo. Your whole history of working with AI, everything you've done trapped in one app. Now, that doesn't seem right. And with Glean AI gateway, it changes. This is Glean making your other AI tools better. Cursor, Copilot, Chat GBT, most vendors want you inside their walls. We want to empower you and give you choice. Now, as Arvin mentioned earlier, one of the first uses of Glean's AI gateway is Glean Tao. Glean's desktop app built for code execution and co-working. Gleantow democratizes coding access for all knowledge workers [01:28:48] so they can delegate complex long-running tasks across authorized tools. And because Glean Tow is configured with Glean's governed enterprise context right out of the box, employees get all the models, MCP server, skills, memories with just one click. No fussy configuration, letting you get right to work. So to pull all this together, let's get the band together and welcome Blair back out here to show towes [applause] and our field marketing team has just put together an intimate executive at Wayfair Tavern. Now we've got 34 attendees who showed up and I jotted down some notes right after the event. Once I get home, I can go ahead and ask [01:29:43] Glean Tao to do some prospect research for me. Right, it's going to automatically pull the attendee list and the voice memo that I took right from my computer and use Gle's enterprise context to enrich both of those files. And to optimize outreach, I want to know a few things, right? Is this company that I met with an existing customer? What role is the person who I spoke to? And what's top of mind for them right now? All these things are really only possible with great enterprise context. And based on that enrichment, Gleantow is able to create a lead score. It weighs up that company information, the past interactions that the customer has [01:30:22] had with us, and the interest signals that we're able to pull in to create this great little ranked list. And after that, we can use glean to target the highest priority with tailored outreach. And just like that, boom, you've got multiple emails all sent for me. But I can take things even a step further with Glean Tow. I can go ahead and update, set up an automation to actually proactively monitor buyer intent signals and keep track of new contacts that I've met over the last 6 months. So when intent signals change, Glean Tao will actually reach out to me and send me a Slack message. >> Wow, that is amazing. As you know, [01:31:05] prospecting has historically been a very manual process. And this is just an example of what's getting reinvented with Glean Towels, Context, and Cody Harness. Thanks again, Blair. >> Thank you. [applause] What we've shared today, model choice, AI usage control, the protect suite, isn't just built into Glean's product. It's built as a platform. So that means these investments benefit you inside Glean and across every front door in your enterprise. Last year, everyone thought one AI product would win it all. This spring, everyone wanted a different app or a different model. Taming AI sprawl isn't about restricting to just a handful of [01:31:55] apps or betting on a monolith. With Glean, that diversity can thrive, all with a platform that lets you do so responsibly. Now, we've heard you loud and clear. You want to do more with AI and we will get there together. Starting with a strong foundation in context, layering on model choice and security and wrapping it all in AI gateway and open APIs so you can scale AI faster, build on AI in new ways, and safely adopt innovations the moment they drop. Let's wrap with a short video on AI Gateway. Thank you. [applause] [music] >> AI doesn't live in one place anymore. It lives in dozens of front doors and dozens of interfaces. And your data and [01:32:50] [music] models flow right through them with no way to see it, control it, or own it. So what you want is one gateway [music] for LLMs and MCP. Yours to own and govern. With LLMs, [music] you choose which run, frontier or open. You set the budgets by user, by group. [music] Take the context, your data, your memories, your skills, interoperable across every front door. Ask the same question anywhere [music] and you get the same answer because it's the same context everywhere. [music] Glean makes other companies or products better because we believe in AI sovereignty. Enterprises should [music] have control and ownership over models, context, [01:33:40] skills, and memories. Whichever front door you use, front doors can come and go, swapped in and out. What's underneath [music] them matters more than ever. So you standardize on what scales, intelligence, context, and security. Own your AI with [music] Glean's AI gateway. >> [music] [music] >> Hello again. Um this time I'm excited to be joined on the stage uh with VC uh one of the most thoughtful engineering leaders I know uh SVP of engineering at at Snowflake uh talking about AI transformation. So um one of the things you said recently that I really liked was um there's a difference between using AI to save yourself 20 minutes and using it to [01:34:39] accomplish 80% of your work. I think a lot of people in the audience, a lot of enterprises probably identify with that problem. the the ROI gap between what you expect from AI and what you actually get. And I know Snowflake has been through that transformation. So, could you tell us a little bit about about that? >> Yeah, I would love to. First of all, super excited to be here. Um, I think the first thing we did and I, you know, it sounds kind of dead obvious when when you say it uh is we started treating our engineering teams like customers. So, just like customers, we surveyed them to understand their problems. Uh just like [01:35:15] customers uh we had hypothesis for what we could do that would improve their lives and just like customers we ran experiments to see if uh we could improve their experience with uh snowflake engineering. >> Um as we did those experiments our NPS with our internal engineers has gone from 15 satisfied 60 dissatisfied to 60 satisfied 15 dissatisfied. So 16x swing in 18 months. All our metrics are up and to the right. Um of course developer happiness, uh how much code they're writing, how happy they are with their uh information finding experiences, thanks to you guys at at le um all of those metrics. Um the second thing we [01:35:59] did, and I think this stuff took me a little time to to internalize. Uh I used to have people walk into my office and say, "Hey, I went through my seven stages of acceptance on AI today." Right. Um and we realized that uh you know working with coding agents was as much a technological problem as it was a an organizational and a social problem. Right? Bunch of people have gotten really good spent 20 years getting excellent at at something and now there's a coding agent that kind of does some of the things they do for them. Um, and as we did that, we realized uh change was something we're going to have to process through the organization uh [01:36:40] at different levels. Like we had to and to quote Andy Grove, let chaos reign before we reigned in the chaos. Um, and so uh we started off by just letting chaos rain. And as we let chaos rain, we got a bunch of fearless, what I'd call AI superheroes or AI zars. These were people who were, you know, paving the way. These were people who were discovering the new patterns for how to work with coding agents. Um, and we just observed like we it was very chaotic, but we just observed. We found who these people were, what they were doing. Um, and now we're in stage two, I'd say, where we're reining in the chaos a bit. [01:37:14] Um, and we're focusing on two things. The vast majority of engineers don't actually want to spend their day learning new patterns to work. Uh, and so for those engineers, we're focusing on raising the floor, right? We're focusing on seeing all these patterns that we discovered in that first phase uh and enabling the whole organization to do it. Um almost what are the maturity levels of coding agents in the enterprise. Uh the second thing we're doing is we're focused on raising the bar or the ceiling. We're still looking at these superheroes. We're seeing what they're doing. We're giving them more time to do to go find new patterns and [01:37:49] then we're baking them back into the organization. >> Yeah, definitely love the part about treating treating your developers like customer experience. I think everything in the business should be that way. Everything is a customer experience. Like it just uplevels everything. You've also you've also talked about continual learning and I think a lot of people think that's like models getting smarter over time. But there's also a lot of ways we have to bring the context to models, the memories, the skills, you know, bring bring all of that. Can you talk a little bit more about that type of continual learning for AI? [01:38:18] >> Yeah, it's a great question. When I think of enterprises, I think, you know, just simplifying this, I think of four things that enterprises do, right? Uh you ask what happened in the past, that's analytics. Uh you ask what is going on right now and that's often sitting in Slack and Jira and Teams and Gmail. Uh you ask what will happen in the future. Um and that's sitting in various predictive systems. Um and then you have systems of record like your SAPs, your workdays, your Salesforces, your Jeters again, right? I think the opportunity for all of us in the enterprise is to build coding agents and and you know something you can call an [01:38:59] enterprise brain or an agentic control plane that is a living organism that ties all these things together, right? It is a system that um is continuously subscribing to everything that is going on in the enterprise whether it's being done by humans or coding agents. It's a system that operates at uh every level of memory right from what's happening today to what happened last week to what happened last quarter to what happened last year. Uh it's ideally a system that operates at every level of the organizational hierarchy. It could be like me, it could be the my team, it could be the broader organization, it could be the subunit I'm part of. Um, [01:39:37] and it does it in a governed way. So, it does it with ales, with role-based controls, uh, in a way that is safe and governed and secure and is not leaking information. Um, I think the opportunity ahead of us and I see that very much in the vision of Glean um, um, and we at Snowflake share that vision. um is to build a system that actually is that contin continually learning system that continued living system um that you can call it the agentic operating system for the enterprise right um I think that opportunity is ahead of all of us I run a 3500 person team uh the number of times I forget to do something because [01:40:15] data didn't show up at the right time insights didn't show up at the right time is you know if I had a penny for every time I did that I'd be very rich uh so I think that is the opportunity >> totally yeah and I love it. I I appreciate you trusting Glean with that. It's it's it's so so important. And um really quick, we're we in 30 seconds. So, you also talked about making the ambitious possible with AI. So, give me give the audience here just just one ambitious problem you think more people should be giving to AI that they are not doing today. >> I'll give you three. [laughter] >> One is uh things that are verifiable. So [01:40:50] we are doing things like rewriting our core execution engine our compiler because these are verifiable problems right we have a ground truth we can just like have a an agentic factory keep working on it the second is things that need complex reasoning across lots of systems like think AI s right these are systems that AI does really well and one last one migrations I think they combine both you they are verifiable and they need coordination across lots of systems >> vec thank you so much I wish we had an hour uh I could listen to this all day. Please uh round a round of applause for VC and let me turn it over to uh Madu [01:41:25] from Dvita talking with Aaron about Dvita's uh transformation with clean on the far side. You're at the [music] Hello everyone. I'm Aaron Grove, VP of solutions engineering at Glean and I'm joined today by Madu Nasharim, CIO at Dvidita Madu. Thank you so much for joining me. >> Oh, happy to be here. >> Let's get started with your journey with Glean. Can you share with the audience where did you start and what's top of mind now at a healthcare company? >> Sure. Um, I'll start with a little bit about Dvita. So Dvita we are a kidney care company and we care for people who have kidney disease but frequently they have other conditions called [01:42:23] co-orbidities. So they'll have heart disease or hypertension. So they're high engagement with the health system. So that's who we care for and um we started speaking with Glean probably about last year and in April of this year we deployed Glean throughout our organization. So all 17,000 people from the front lines in the clinics all the way to uh the back offices, everyone has glean and we've had some um great uh responses. We've had um of the 70,000 uh users almost 40,000 engage uh on a fairly frequent basis. So within days we had about 5.4 4 million queries since we started and on average people saved about 19 to 20 minutes [01:43:15] worth of time. So, um it's been it's been a really good uh deployment and we're still working our way through the potential possibilities. So, um yeah, so that's that's where we are and these great partner >> 70,000 and a million. I think that's understated a little bit. So, there's a lot of value there. Yeah. >> Uh which AI use cases are creating the most meaningful business value for Dvita today? >> Yeah. So um we we we are very intentional about using AI especially in healthcare and uh Glean is part of an overall AI ecosystem. So we've brought in um AI capabilities to remove administrative friction and things of [01:44:00] that nature. So there is absolutely uh no confusion about the complete level of um asynchrony if you will between the available supply of healthcare labor and the you know needed the demands. There's a big uh disconnect there. So our goal is really to make sure that we allow our caregivers more time with the patients more time to do the care they they uh need to do. So a couple of use cases where we've found tremendous value is um synthesizing information. So usually when we're talking about technology, we forget there was a time before tech. So there's stuff information about patients on paper, faxes, PDFs and APIs and being [01:44:49] able to synthesize across all of that and provide a summary is critically important. So we found a lot of value in that. And then uh we're also using AI for proactive care with some of our patients. Um so dialysis is not easy. It's not like popping a pill. You have to spend 3 4 hours three times a week getting your blood clean. And so we try to watch for indications that patients are going to fall off the wagon if you will. And we proactively engage with them to keep them on their treatment plan. So both in making sure we have enough time for uh our caregivers to care for our patients but also uh engaging with the patients earlier. [01:45:36] >> Patients are very important in your business, right? Um and I know you have corporate teams, field leaders, clinical environments and the emphasis has always been on solving for better patient care. How did that change the way you thought about rolling out Glean? H I think uh so we have glean as I mentioned from the front line all the way to the back offices and people who use AI especially in healthcare that's a very strong highly discerning audience it has to be in workflow so we've seen that as long as it's part of their workflow they'll adopt it has to work right and be reliable every time That's not optional. And the third is of [01:46:24] course with uh it it has to add value. And you know in healthcare we don't have a return button. You know once care is given it's given. So we have to make sure that um it it it works and that has been super helpful with the context that Glean has. It's been able to bring that uh back very effectively to our um to our uh staff. And the other thing we've done is we have this innovation marketplace where our teammates can share the agents they've built with each other and that uh absolutely helps and you know one thing I'll say is you know people look at numbers a lot of numbers have been floated around so I'll [01:47:05] share a few you know we said 5.4 4 million queries at approximately 20 minute savings each. That's 108 million minutes or 75,000 days. But you know what is truly valuable? If our caregiver has five extra minutes >> for the patient, that is the ROI and it's priceless. >> 100%. Well, Moo, thank you so much for joining us at Glingo. It's fantastic to hear how um a healthc care leader is driving thoughtful AI outcomes and AI transformation. So, thank you for sharing your learnings with us. >> Thank you for your partnership. >> All right. And now I'm going to welcome Juliet back to the stage to close us out. [music] Wow, what a morning. Thank you so much [01:48:02] to everybody who made it happen this morning. We're going to wrap up now. Going to go to the breakout sessions and I would like you back here for 3:45. We've got an amazing closing session. Just before I let you go, we have the agenda here, but also I just wanted to remind you that today is a celebration of what you're building. Thank you so much. Have a great rest of the day and I'll see you at 3:45. >> [music] [music] [music] >> Heat., Glean:GO 2026 Keynote, Glean, 01m32jrx1de21h67srm5vnhxfk.