Nvidia Sets Open vs. Closed AI Debate for Disrupt 2026
Nvidia leaders Nader Khalil and Sydney Sykes will unpack open, proprietary and hybrid AI strategies at TechCrunch Disrupt 2026 in San Francisco this October.
Summary
Nvidia executives Nader Khalil and Sydney Sykes will lead “The Open vs. Closed AI Debate Is Just Getting Started” on the Builders Stage at TechCrunch Disrupt 2026, running October 13 to 15 in San Francisco for more than 10,000 tech leaders. Passes are discounted by up to $200 until September 25 at 11:59 p.m. PT, while exhibit bookings close September 18.
Khalil, Nvidia’s Director of Developer Tech overseeing open source and local AI, co-founded Brev.dev, which Nvidia acquired in July 2024. Brev.dev helps developers deploy across public cloud, private cloud and on-premises infrastructure without relying on one compute provider. Sykes, Nvidia’s Global Head of VC Partnerships, will add the investor perspective on scalability and defensibility.
Nvidia said in July that 145 papers accepted at ICML 2026 cited its Nemotron open models and datasets, while other Nvidia open model families supported robotics, autonomous vehicles and biomedical research. Nemotron 3 Super, launched in March, is an open 120 billion parameter model for agentic workloads, and companies are combining it with proprietary systems. CEO Jensen Huang argued at GTC earlier in 2026 that AI’s future includes both approaches. Founders must balance cost, margins, speed, control, infrastructure, differentiation and provider dependence; investors must locate durable value, while business leaders face procurement, security, data control and switching decisions.
Positives
- 145 papers accepted at ICML 2026 cited Nvidia’s Nemotron open models and datasets.
- Nemotron 3 Super provides an open 120 billion parameter model designed for agentic workloads.
- Brev.dev supports deployment across public cloud, private cloud and on-premises infrastructure without dependence on one compute source.
- Companies are combining open and proprietary models, giving startups flexibility to match models with workloads.
Risks & concerns
- Competitors using the same proprietary API must find differentiation in data, workflows, distribution, customer relationships or specialized technology.
- Open models transfer deployment, optimization and infrastructure responsibilities to the companies adopting them.
- AI economics vary by workload and scale, making cost advantages difficult to generalize.
- Tight dependence on one model can constrain startups when capabilities and economics change.
- A poor model strategy can weaken margins, fundraising, product roadmaps, security and future provider flexibility.