Open AI or closed AI? Learn what to build at Disrupt 2026

For AI startups, model selection is no longer necessarily a one-time architectural decision.

The open model has been improved. Frontier API continues to evolve. Models can be customized for specific workloads, and some companies are building products that use multiple models rather than focusing on one model. This gives founders more ways to build and makes more decisions about where to spend money, what to own, and how much flexibility to maintain as technology changes.

TechCrunch Disrupt 2026 will feature four conversations that approach these decisions from different points in the AI ​​stack, from multi-model applications and customized models to the infrastructure and the chips underneath.

Secure your Disrupt pass to hear how our AI leaders are navigating these choices. Now you can save up to $100 on a pass and get a second pass for 50% off. Or, if you’ve been affected by a layoff, get a $75 Expo+ Pass for a limited time.

When one AI model is not enough

The choice between open and proprietary models assumes that a company must first choose one or the other. Increasingly, AI products can invoke different models for different jobs.

The session “The Real Tokenmaxxing: How the Best AI Companies Navigate a Multi-Model World” will include Mo Jomaa, Partner at CapitalG. Vipul Ved Prakash, co-founder and CEO of Together AI. Zuzanna Stamirowska, CEO and co-founder of Pathway, took to the Builders Stage.

They investigate why companies use multiple models. How do you balance cost, performance, and flexibility? And when open models can outperform proprietary alternatives.

For founders, that flexibility can impact more than model performance. This can impact operating costs, product decisions, and how quickly companies can take advantage of better models as they emerge.

Get your ticket to Disrupt to hear how companies decide which models are right for which jobs. Buy now and save up to $100 and get your second pass at 50% off.

How much of your AI stack should you own?

The choice of model also raises the larger question of what the company should own.

Manos Koukouumidis, CEO and co-founder of Oumi, will be on the Real World AI stage with the theme, “Which AI Should You Really Deploy: Rent, Customize, or Build?” He considers whether startups should build their own models, when customization can be a competitive advantage, and choosing between open and proprietary AI.

Koukoumidis will help attendees evaluate frontier APIs, customized open weights, and full ownership of AI through audience voting, startup scenarios, and practical frameworks. Participants will gain three decision principles they can bring to their next architectural conversation.

The choices founders make can shape both the product and the business behind it. Building more AI stacks allows for greater control and differentiation, but it can also require more time, people, and resources than relying on existing models.

Add the Oumi session to your Disrupt agenda as a practical framework for determining how much of an AI stack your company really needs to own. Get your pass now and have a front row seat to this discussion.

Where is open and proprietary AI suitable?

For some startups, the decision will still come down to trade-offs between open and proprietary models.

Nader Khalil, Director of Developer Tech at Nvidia, and Sydney Sykes, Global Head of VC Partnerships at NVIDIA, will be on the Builder Stage for Building AI Startups Worth Betting on. They explore what founders are choosing today, the tradeoffs between frontier APIs and open-weight models, and how those choices can impact product strategy and long-term differentiation.

These trade-offs have business implications. Your choice of model will determine everything from cost and infrastructure requirements to how much control you have over your product. It can also determine where and how easily startups can create differentiation that competitors cannot imitate.

Secure your Disrupt pass to hear how builder and venture perspectives shape model decisions. Save up to $100 on a pass and get 50% off your second pass.

When architecture reaches hardware

Model architecture is only part of the equation. AI performance ultimately depends on its underlying hardware, and advances in AI are beginning to change the way that hardware is designed.

Anna Goldie, founder and CEO of Ricursive Intelligence, and Azalia Mirhoseini, founder and CTO, take that idea to the disruption stage in When AI Starts Designing Its Own Hardware. They explore how AI is optimizing chips and hardware, why model architectures and hardware are becoming more tightly coupled, and what an increasingly open AI ecosystem means for its underlying infrastructure.

For founders, the relationship between AI and hardware can impact how quickly new capabilities reach the market. Faster chip development could also change the infrastructure available to startups building the next generation of AI products.

Get your ticket to Disrupt to hear what happens when AI starts helping design the hardware that powers it.

Learn how to keep your AI options open at Disrupt 2026

These conversations are part of more than 200 sessions across six industry stages, roundtables, and breakout sessions at Disrupt, taking place October 13-15 at Moscone West in San Francisco. More than 10,000 founders, investors, executives, and technology leaders are expected to attend, with more than 250 speakers and more than 300 startups exhibiting.

Additionally, matchmaking, dealmaking, and ad hoc networking will give participants the opportunity to connect with founders, investors, potential partners, and other builders who are facing many of the same technology and business decisions.

Startups may be using Frontier APIs today and customizing open models tomorrow, or eventually moving workloads between multiple models. Staying flexible can be just as important as choosing the right model now.

Secure your pass to TechCrunch Disrupt 2026 and hear how AI leaders are approaching the choices that determine what and how they build. Now you can save up to $100 and get 50% off your second pass. Have you been fired? Get an Expo+ Pass for just $75.