The TechCrunch Disrupt 2026 AI Stage is back, and this year’s lineup is one of the strongest yet. If you’ve spent any time this year trying to keep up with how fast artificial intelligence is reshaping the startup world, you already know the ground hasn’t stopped moving — new models ship monthly, entire job categories appear seemingly overnight, and the security assumptions enterprise software has run on for two decades are being tested by systems that can now act on their own.
That’s the backdrop for this year’s AI Stage, which this October brings together speakers from Anthropic and OpenAI — arguably the two most closely watched companies in the industry right now — alongside leaders from Databricks, AWS, Okta, Glean, Monte Carlo, and several founders currently building at the frontier of applied AI.
Below, we’ll walk through everything confirmed so far for the event: the sessions, the speakers, the themes tying it all together, and why this year’s edition feels like a meaningfully different conversation than the one the industry was having a year ago.
When and Where Is TechCrunch Disrupt 2026?
TechCrunch Disrupt 2026 runs October 13–15 at Moscone Center in San Francisco. The AI Stage, presented by Google for Startups, is one of several dedicated tracks running alongside the main stage, Startup Battlefield, and the exhibition floor — but it’s consistently been one of the most heavily attended parts of the event in recent years, and this year’s speaker list suggests that trend isn’t slowing down.
Disrupt as a whole typically draws more than 10,000 founders, investors, operators, and engineers over its three days, and the AI Stage has become something of a barometer for where the industry’s attention is actually pointed, as opposed to where the headlines are pointed. If you want the full event schedule and ticket options, you can check the official TechCrunch Disrupt registration page directly.
Why This Year’s AI Stage Feels Different
For the past few years, AI Stage programming has tracked pretty closely with the industry’s mood at the time — a lot of excitement about what generative models could technically do, a lot of speculation about what might be possible next. This year’s TechCrunch Disrupt 2026 AI Stage lineup reads differently. The sessions are organized around three questions founders are wrestling with in real time, not hypothetically:
How do you price an AI product when the underlying model layer is becoming a commodity? As foundation models from multiple providers converge on similar capabilities, the old logic of “charge more because the AI is impressive” stops working. Founders now have to figure out what they’re actually charging for — workflow integration, outcomes, domain expertise, trust — because the model itself is no longer the differentiator it once was.
What does real security look like when AI systems are making autonomous decisions inside enterprise infrastructure? This is a genuinely new problem. Traditional application security was built around the assumption that a human initiates each action and that permissions can be scoped around predictable, rule-based behavior. Agentic AI breaks that assumption entirely.
What happens to go-to-market strategy when the traditional playbook doesn’t apply anymore? Sales and marketing organizations built around human-scale processes are running into AI-native competitors that operate at a completely different velocity. New roles are emerging to fill that gap, and some of them didn’t exist as job titles two years ago.
These aren’t abstract concerns for panel discussion — they’re the exact problems founders in the Disrupt audience are trying to solve for their own companies right now.
AI Stage Sessions: A Closer Look
Inside Enterprise Claude Deployments, with Anthropic
One of the most anticipated sessions on this year’s AI Stage comes from Cat de Jong, Anthropic’s Head of Applied AI. Her role puts her in a somewhat unusual position within the AI industry: rather than working on model development or research, she spends her time directly embedded with the enterprises actually putting Claude to work inside their most critical workflows.
That vantage point matters, because most of what the public hears about enterprise AI comes either from vendor marketing (which tends to emphasize success stories) or from analyst reports (which tend to lag reality by months). De Jong’s session promises something closer to ground truth — the patterns that show up again and again once you’ve watched dozens of large organizations attempt to operationalize AI at scale.
Expect the conversation to dig into which deployments succeed almost immediately and what they have in common, where organizations get stuck, and what separates companies that move from pilot to production within a few months from the ones still running “pilot” programs eighteen months later with nothing to show for it.
The Rise of AI-Native Go-to-Market, with OpenAI
The second headline session comes from Tara Seshan, OpenAI’s Head of Productivity, and it tackles a phenomenon that’s been quietly reshaping go-to-market teams across the industry: the emergence of the “GTM engineer” as a distinct professional discipline.
Two years ago, this wasn’t a job title anyone used. Today, it’s one of the fastest-growing roles in tech, and a meaningful number of the people doing this work aren’t employees of any single company — they’re independent practitioners who’ve built businesses generating seven figures in revenue by helping companies rebuild their sales and marketing motion around AI-native tooling.
Seshan’s session traces how this shift happened: AI didn’t just make each step of the traditional GTM funnel faster, it collapsed several of them into a single automated workflow, which meant the skill set required to build and manage that workflow started looking less like traditional sales operations and more like software engineering. Learn more about OpenAI’s enterprise work ahead of the session.
Rebuilding Enterprise Security for the AI Era
Security gets its own dedicated block this year, approached from two different angles across two separate sessions.
The first features Arsalan Tavakoli, co-founder and SVP of Field Engineering at Databricks, under the framing “The Enterprise Isn’t Broken. Your Assumptions About It Are.” The premise: AI is now making autonomous decisions inside some of the most sensitive systems enterprises run, at a speed traditional security review processes were never designed to accommodate.
The second is a broader panel featuring Chet Kapoor (AWS), Katie Moussouris (Luta Security), and Wendy Nather (1Password), titled “Securing the AI Enterprise.” This session takes the infrastructure-level view — what it actually takes to build AI systems that large, risk-averse organizations are willing to run at scale.
Agent Security, Unfiltered
Closely related is a conversation with Ric Smith, President of Product & Technology at Okta, focused specifically on agent security. The framing is blunt: agentic AI is powerful, but it was never actually built to be secure, and enterprises adopting it are now having to rebuild fundamental cybersecurity concepts from the ground up.
Smith’s session is expected to dig into why application-level permission models — the standard approach to access control for years — are fundamentally mismatched to how AI agents actually operate.
The Video Intelligence Race: Real-Time, Reasoning, and What Comes Next
Not every session on the AI Stage is about enterprise infrastructure. One of the more forward-looking conversations brings together Dean Leitersdorf, co-founder and CEO of Decart, and Amit Jain, co-founder and CEO of Luma AI, two founders building at the edge of visual AI.
The framing captures something that’s been happening quietly over the past year: AI-generated video has moved past being a novelty into something with real technical depth — systems that don’t just generate plausible-looking footage, but model how objects and environments actually behave and interact.
Is SaaS as We Know It Over?
Perhaps the most existential question on this year’s agenda gets its own dedicated panel: “Rewriting SaaS: Why AI Breaks the Old Business Model.” The panel brings together Arvind Jain (Glean), Barr Moses (Monte Carlo), Cathy Gao (Sapphire Ventures), and Aaron Jacobson (NEA) — a mix of operators actually running SaaS companies through this transition and investors evaluating how to underwrite the next generation of software businesses.
The core tension: the traditional SaaS model — charge a per-seat subscription fee for access to software — assumed value scaled with the number of humans using it. AI breaks that assumption in two directions at once: agents can now do work that used to require a human seat, and as foundation models commoditize, the software layer itself becomes harder to differentiate.
The GTM Engineer: How AI Created Tech’s Next Big Job Category
Rounding out the go-to-market thread is a session with Kareem Amin, co-founder and CEO of Clay, digging further into the GTM engineering phenomenon. Amin’s session promises a founder’s-eye view of how AI-native go-to-market actually gets built in practice — the specific tooling and workflow decisions separating teams that have made this transition from teams still figuring it out.
What Ties These Sessions Together
Step back from the individual sessions and a clear thread emerges. This year’s TechCrunch Disrupt 2026 AI Stage isn’t organized around “AI is amazing, look what it can do” — the conversation the industry has largely already had. It’s organized around the much harder second-order questions that come after the initial excitement wears off: How do you make money from this sustainably? How do you keep it secure once it’s making decisions without a human in the loop? And how do you restructure sales, security, and engineering teams so they’re built for how AI actually works, rather than retrofitted from how things used to work?
That’s a notably more mature set of questions than the industry was asking even a year or two ago — and it reflects where a lot of founders and enterprise buyers actually are right now: past the initial pilot phase, and into the much harder work of figuring out what sustainable, secure, AI-native businesses actually look like. For more background on how enterprises are approaching this shift, see our related guide to enterprise AI adoption and our breakdown of agentic AI security.
Should You Attend the TechCrunch Disrupt 2026 AI Stage?
If you’re building anything in or around AI — a product built on foundation models, a security layer for agentic systems, or a new category of go-to-market tooling — the AI Stage is likely to be one of the more useful rooms of the year. A lot of the most valuable conversations at events like this happen in the hallways around the formal sessions, and with speakers from Anthropic, OpenAI, Databricks, AWS, and Okta all in the same building over three days, there’s a good chance the conversation you have in line for coffee is as useful as the one on stage.
Beyond the AI Stage specifically, attendees get access to the full Disrupt program: the main stage lineup, Startup Battlefield, the exhibition floor, and networking with the broader pool of founders, investors, and operators expected to attend. For related coverage, check out our guide to everything founders should know about Disrupt.
Practical Details
- Dates: October 13–15, 2026
- Location: Moscone Center, San Francisco
- Track: AI Stage, presented by Google for Startups
- Confirmed speakers: Cat de Jong (Anthropic), Tara Seshan (OpenAI), Arsalan Tavakoli (Databricks), Ric Smith (Okta), Dean Leitersdorf (Decart), Amit Jain (Luma AI), Arvind Jain (Glean), Barr Moses (Monte Carlo), Cathy Gao (Sapphire Ventures), Aaron Jacobson (NEA), Kareem Amin (Clay), Chet Kapoor (AWS), Katie Moussouris (Luta Security), Wendy Nather (1Password)
Ticket pricing for Disrupt typically increases as the event approaches, so if attending is on your radar, it’s worth locking in a pass sooner rather than later via the official ticket page. More sessions and speakers are expected to be announced ahead of the event.
Frequently Asked Questions About the TechCrunch Disrupt 2026 AI Stage
When is TechCrunch Disrupt 2026? October 13–15, 2026, at Moscone Center in San Francisco.
Who is speaking on the TechCrunch Disrupt 2026 AI Stage? Confirmed speakers include representatives from Anthropic, OpenAI, Databricks, AWS, Okta, Glean, Monte Carlo, Decart, Luma AI, Clay, Sapphire Ventures, NEA, Luta Security, and 1Password, with more announcements expected.
What topics does the AI Stage cover this year? AI pricing and business models, enterprise and agent security, the rise of AI-native go-to-market roles, and the future of visual/video AI.
Who presents the AI Stage at TechCrunch Disrupt 2026? The AI Stage is presented by Google for Startups.
How many people typically attend TechCrunch Disrupt? More than 10,000 founders, investors, and tech leaders typically attend across the full three-day event.
This article is based on publicly available programming details for TechCrunch Disrupt 2026. Session details, speaker lists, and additional programming may be updated by TechCrunch ahead of the event.
