Open or closed AI? Nvidia’s Nader Khalil and Sydney Sykes take on one of the decisions shaping next-gen startups at TechCrunch Disrupt 2026

Nvidia executives Nader Khalil and Sydney Sykes will lead a session titled “The Open vs. Closed AI Debate Is Just Getting Started” on the Builders Stage at TechCrunch Disrupt 2026 in San Francisco, October 13-15. The discussion will examine trade-offs between building on open models, proprietary frontier models, or hybrid approaches and how those choices affect startups' costs, control, and competitive positioning.

By AI Newsroom· Reviewed by Pranav, Founder & Editor-in-ChiefPublished 29 minutes agoUpdated 29 minutes ago0 views

Open or closed AI? Nvidia’s Nader Khalil and Sydney Sykes take on one of the decisions shaping next-gen startups at TechCrunch Disrupt 2026

Why It Matters

Founders, investors and technical leaders are actively choosing between open and proprietary AI approaches — a decision that can materially influence infrastructure, margins, differentiation and fundraising. The session brings both developer/infrastructure and venture perspectives to help attendees weigh practical business trade-offs rather than abstract philosophy.

Key Facts

Event: TechCrunch Disrupt 2026, Builders Stage
Dates: October 13-15, 2026
Speakers: Nader Khalil (Nvidia Director of Developer Tech) and Sydney Sykes (Nvidia Global Head of VC Partnerships)
Nader Khalil background: Co-founded Brev.dev, an AI infrastructure company acquired by Nvidia in July 2024
Nvidia open models cited: 145 ICML 2026 papers cited Nvidia's Nemotron open models and datasets (reported by Nvidia)

TechCrunch Disrupt 2026 will feature a session focused on a pressing, practical choice for AI startups: whether to build on open models, license proprietary frontier models, fine-tune privately, run models locally, or pursue hybrid strategies. Nvidia’s Nader Khalil and Sydney Sykes will lead the Builders Stage discussion, framing the issue as a business decision with implications for cost, infrastructure, control, speed to market and competitive differentiation.

Khalil will approach the topic from a builder and infrastructure angle; he previously co-founded Brev.dev, which Nvidia bought in July 2024, and now leads open source and local AI efforts as Director of Developer Tech. Sykes will contribute a venture perspective as Nvidia’s Global Head of VC Partnerships, highlighting how these technical choices affect investability and scalability. Together they aim to unpack how different model strategies map to product roadmaps, fundraising narratives and operational requirements.

The rise of open models has narrowed capability gaps: Nvidia said 145 ICML 2026 papers cited its Nemotron open models and datasets, and the company has invested in open families such as Nemotron 3 Super, a 120-billion-parameter model released in March designed for agentic workloads. At the same time, proprietary frontier labs continue to push model performance, creating a landscape where hybrid approaches are already common and where the right technical decision depends on workload, economics and how quickly capabilities evolve.

Panel topics will include trade-offs around cost versus control, where competitive advantage resides if multiple parties can access similar models, and the operational burdens of deploying and maintaining open models. The session is pitched as practical guidance rather than a philosophical debate, intended for founders, investors, line-of-business leaders, developers and students who must align technical decisions with business objectives.

TechCrunch is inviting attendees to join the discussion at Disrupt in San Francisco from October 13-15, 2026; the event materials note early pricing savings available through September 25 at 11:59 p.m. PT. Khalil and Sykes’s session aims to help participants better understand when an open, closed or combined approach makes commercial sense and what trade-offs each path entails.