# Open or closed AI? Nvidia leaders debate at Disrupt

> Two Nvidia executives will tackle the biggest decision facing AI startups: whether to build on proprietary models, open models, or both.
- Title: Open or closed AI? Nvidia leaders debate at Disrupt | Shortify
- Summary: Two Nvidia executives will tackle the biggest decision facing AI startups: whether to build on proprietary models, open models, or both. There's no single…
- Keywords: nvidia, nader khalil, sydney sykes, techcrunch disrupt 2026, open source ai, proprietary models, ai startups, ai infrastructure, brev.dev, nemotron, san francisco, founders
- Source: TechCrunch — https://techcrunch.com/2026/09/18/open-or-closed-ai-nvidias-nader-khalil-and-sydney-sykes-take-on-one-of-the-decisions-shaping-next-gen-startups-at-techcrunch-disrupt-2026
- Author: TechCrunch Events
- Published: 2026-09-18T15:30:00+00:00
- Read time: 2 min
- Topics: nvidia, nader khalil, sydney sykes, techcrunch disrupt 2026, open source ai, proprietary models, ai startups, ai infrastructure, brev.dev, nemotron, san francisco, founders
## The AI model choice that shapes everything
Founders building AI products face a critical decision: use proprietary frontier models for speed, open models for control, or fine-tune your own. This choice affects cost, infrastructure, margins, differentiation, and competitive advantage.
## Nvidia brings two perspectives to the debate
Nader Khalil, Director of Developer Tech, brings infrastructure expertise from co-founding Brev.dev. Sydney Sykes, Global Head of VC Partnerships, represents the venture ecosystem. Together they'll examine the open versus closed decision from builder and investor angles.
## Open models are catching up fast
Nvidia's Nemotron open models were cited in 145 papers accepted at ICML 2026, spanning robotics, autonomous vehicles, and biomedical research. The gap between open and proprietary capabilities is narrowing.
## Proprietary labs keep pushing forward
While open models advance, proprietary frontier labs continue improving their capabilities. The real question is no longer whether open models work, but where each approach makes commercial sense.
## Nvidia CEO rejects the either-or framing
Jensen Huang argued at GTC that the future is proprietary and open, not proprietary versus open. But building a company around that decision is far more complex than the philosophy.

> the future is not proprietary versus open, but proprietary and open
## The real trade-offs founders face
Lower cost versus data control. Owning your stack versus maintaining infrastructure. Models changing every few months versus product lock-in. These are the practical questions that determine success or failure.
## Your model isn't your moat yet
If competitors access the same proprietary API, differentiation comes from data, workflow, distribution, or customer relationships. Open models offer flexibility but require decisions around deployment and optimization.
## Hybrid approaches are already winning
Companies are combining Nvidia's Nemotron 3 Super open model with proprietary models rather than choosing one side. The most interesting reality may be treating open and closed as complementary, not competing.
## This decision matters across your org
Founders shape margins and fundraising. Investors distinguish real defensibility from thin layers on someone else's model. Line-of-business leads control procurement and data security. Developers connect technical choices to business models.
## Join the debate at Disrupt 2026
Khalil and Sykes will unpack these trade-offs on the Builders Stage at TechCrunch Disrupt 2026 in San Francisco, October 13-15. Register now and save up to $200 before September 25.
## Key takeaway

There's no single right answer to choosing between open and closed AI models. The decision affects cost, control, speed, and competitive advantage, and smart founders need to understand the trade-offs for their specific business.