SSAD2026 workshop debates open-loop evaluation flaws and open-source wave for autonomous driving

At the #ssad2026 workshop, Valeo AI researcher Kashyap Chitta gave a systematic overview of how end-to-end autonomous driving evaluation frameworks have evolved, and called on the automotive industry to replicate the open-source wave seen in the LLM space. Evaluation and open source were framed as the two key topics for accelerating progress in the field, making them well worth watching.

Confirmed

• The researcher traced the evolution of end-to-end autonomous driving evaluation: nuPlan's open-loop evaluation has known issues (see "Is Ego Status All You Need for Open-Loop Autonomous Driving"), which paved the way for NAVSIM's rise.
• In his talk, Kashyap Chitta used software industry analogies (such as the Linux vs. Windows NT rivalry) to argue that open source is the right path for AI development, and recommended Bill Gurley's related articles.
• Bill Gurley published a new piece, "From Open Source Software to Ope...", discussing open-source strategy and the question of whether the automotive sector could see an open-source wave similar to the LLM space.
• A researcher posted asking whether the automotive industry could see an open-source wave like the LLM field and replicate its rapid progress, listing current players and noting that Valeo.ai is among them, while arguing the field still has room for more participants; follow-up replies also shared concerns about the current state of autonomous driving deployment.
• Kashyap Chitta reflected on the changing culture of machine learning evaluation: in the 1990s almost no papers included benchmarks, whereas in the 2020s a paper without a benchmark can hardly be published; building shared test frameworks was no easy process, and the progress made over the years is impressive.

Why it matters

• The shift from open-loop evaluation to new benchmarks like NAVSIM directly shapes the direction and credibility of end-to-end autonomous driving research.
• If the automotive industry replicates the LLM open-source wave, it could lower barriers to entry and speed up technological iteration, though the current deployment landscape suggests a gap remains between the ideal and reality.

2026-09-16 ~ 2026-09-16 · 5 related posts

• [source] Researcher calls for an open-source wave in autonomous driving, citing current deployment reality check — abursuc · 2026-09-16
• Bill Gurley on Open Source Strategy: can automotive AI replicate the LLM open-source wave — abursuc · 2026-09-16
• [source] Kashyap Chitta argues open source is the way, using the Linux vs Windows NT analogy — abursuc · 2026-09-16
• From no benchmarks in the 1990s to no papers without them: a short history of ML evaluation — abursuc · 2026-09-16
• [source] E2E driving evaluation shifts: nuPlan open-loop flaws exposed, NAVSIM benchmarks drove two years of progress — abursuc · 2026-09-16