Meta Horizons World probably puts up similar numbers if you sum up the hardware/software tech stack to get this number.
The new game is finding a single sentence with the most instances of "safe" or "safety". My current high score is 4..
The real technical challenge is rappresented by edge cases: a software that is excellent 99.9% of the time can still be unacceptable if the remaining 0.1% contains rare but catastrophic scenarios. And that's why we still don't see many self-driving vehicles on the roads today.
However, NVIDIA has a credible shot because it controls much of the loop - hardware, training infrastructure and simulation environment. If it works they will impose a huge vendor lock-in, difficult to replicate for other competitors.
Here's Archive.org's copy of the page from 2025 in September, their earliest copy:
https://web.archive.org/web/20250920031549/https://www.nvidi...
"Safety transistors safety assessed" exists in this version too
> 18,600+ Engineering years invested in vehicle safety to date
What does this even mean?
> 7,000,000 Lines of safety-assessed code
Are we seriously using LoC as a measure of productivity again?
Not to mention the em-dashes
If it means what I think it means, you take every engineer working on it (and maybe the years of research involved) and add it all up. Say you have a room with about 10 engineers with 10 years of experience per developer, you can claim there's 100 years of developer experience between all of them (maybe the overlaps not unique enough and its more like 30 to 50 years? but in this case I think they're rounding up, and I assume it means thousands of engineers involved in the project) that's how I took it.
My first interview in tech I was asked what the heck I was even doing with the D programming language, followed by the remark that in the next room (where all the devs were) there was at least 100 years of experience between everybody there, and not a single one knew what D was, my manager clearly did, which cracked me up.
Yes, sadly. Because its how everyone justifies LLMs. "Look at how much code it writes!" is the only measure they can come up with to sell its usefulness, completely forgetting that it'll be more useful if we started talking about how much code they remove.
It means over 18,600 engineering hours have been spent working on vehicle safety. This is a pretty common metric.
I don't even know what that was supposed to mean. Hopefully all the safety transistors in the safety graphics card of my safety-PC were safety-assessed, too /s
Hot take here, but personally I feel they should safety assess the danger transistors, reducing the need for so many safety transistors.
It's also hard to blame Nvidia for the pivot, from where I'm standing. Their proprietary middleware like PhysX, DLSS and RTX has been memed to death by PC gamers, while high-margin edge and datacenter customers are chomping at the bit for CUDA compute. Nvidia's raster stack is more-or-less complete, the things that PC gamers are asking from them are not realistic or fairly priced at this moment in time.
That was my point. It's not even a pivot! They're still making consumer cards! They've even product-differentiated enough that the consumer cards are still on the shelves at close-to-MSRP, despite world-historic demand for adjacent parts of the lineup.
Being _mad_ at Nvidia in this setting is weirdly possessive - a business that was 90% gaming is now 10x larger and 9% gaming[1]. You haven't lost ground!
[1]: numbers made up but you get the point
Ensure full AV stack safety with NVIDIA Halos OS, the unified safety foundation for physical AI production.
Overview
NVIDIA Halos is a full-stack, comprehensive safety system that unifies safety elements across vehicle architecture, AI models, chips, software, tools, and services to ensure the safe development and deployment of autonomous vehicles (AVs) from cloud to car.
The system covers the full development lifecycle with design-time, deployment-time, and validation-time guardrails that collectively build safety and explainability into AI-based AV stacks. These guardrails are implemented using three powerful computers—NVIDIA DGX™ for model training, NVIDIA Omniverse™ and Cosmos™ for simulation, and NVIDIA DRIVE AGX™ for deployment. At the heart of the vehicle, NVIDIA Halos OS provides the unified software foundation necessary to bridge these AI capabilities with production-ready safety.
NVIDIA Halos complements existing industry-standard safety practices, while introducing unique elements for autonomous vehicles. This ensures regulatory compliance and advances safe and reliable AV stacks, together with NVIDIA’s Halos AI Systems Inspection Lab.
Halos is also extending its comprehensive safety framework beyond AVs to robotics, further enhancing the reliability and safety of intelligent systems.
NVIDIA Halos OS delivers a certified OS, standardized interfaces, AI guardrails, and pre-deployment validation for L4 robotaxi deployments at scale.
NVIDIA unifies vehicle architecture to AI models; chips, software, and tools to services for safely developing AVs from cloud to car.
Highlights
NVIDIA Halos is the result of continuous investment in autonomous vehicle safety—from research to engineering to active engagement with international safety standards—validated by independent third-party assessments.
Engineering years invested in vehicle safety to date
Safety transistors safety assessed
Lines of safety-assessed code
Daily end-to-end integration tests for validation
Platform safety monitors
Hours of safety test data
Patents filed
Research papers published on AV safety
Certificates and assessment reports issued
Technology
As AV companies transition to AI-based, end-to-end architectures, NVIDIA Halos provides the critical safety foundation to ensure system-level reliability and iterative improvement for automated driving systems. This includes integration of third party-assessed hardware, software, and processes with a diverse algorithmic architecture and validation pipelines.
Design-time safety guardrails for built-in hardware/software safety and trustworthy development processes.
Validation-time guardrails for data generation, simulation, evaluation, and lifelong safety assurances.
Deployment-time guardrails for run-time monitoring and real-time introspection.
Benefits
NVIDIA Halos helps to ensure AI-driven AV systems are safe and secure. Partners can tap into NVIDIA’s investments in AI safety to accelerate development and enhance AV reliability. NVIDIA Halos is also open to developers, enabling adoption or customization of safety elements to drive the shared mission to create safe and reliable autonomous vehicle technology.
Design-time, deployment-time, and validation-time guardrails collectively build safety and explainability into several layers of technologies spanning platform safety, AI algorithmic safety, and ecosystem safety.
At the top of the NVIDIA Halos elements sits the NVIDIA Halos AI Systems Inspection Lab, which allows customers and ecosystem partners to verify the safe integration of their products with NVIDIA Halos elements. The lab is the first worldwide program to be accredited by ANAB for AI functional safety.
Use Cases
NVIDIA Halos integrates foundational models and a diverse algorithmic stack, combining classical and AI-based, end-to-end models to drive system-level safety in the shift toward AI-driven AV architectures.
The robust foundation for autonomous driving systems includes:
NVIDIA Halos OS is a unified, production-ready software foundation for AI-driven vehicles, built on three in-vehicle software layers and a cloud-side infrastructure layer:
“Joining NVIDIA's Halos AI Systems Inspection Lab marks our commitment to advancing driving safety. By combining Bosch's comprehensive in-house ADAS sensor expertise with NVIDIA's AI validation framework, we're setting new standards for safe and reliable ADAS solutions.”
— Dennis Raabe, Senior Vice President ADAS Components, Bosch
“NVIDIA’s latest evaluation with ANAB verifies the demonstration of competence and compliance with internationally recognized standards, helping ensure that developers of autonomous machines—from automotive to robotics—can meet the highest benchmarks for functional safety.”
— R. Douglas Leonard Jr., Executive Director, ANAB
“Aligned with our shared mission to enhance safety, efficiency and productivity, we congratulate NVIDIA on the launch of its Halos AI Systems Inspection Lab. The Lab's reports will provide valuable insights to support our certification efforts.”
— Thomas Steffens, Head of Certification Body Functional Safety and Cybersecurity, TUV Rheinland
“We are pleased to hear of NVIDIA's commitment to advancing autonomous vehicle safety and welcome NVIDIA's Halos Inspection Lab efforts for structured and thorough development of AI for safety-relevant applications.”
— Dominik Strixner, Global Lead Functional Safety Auto and Mobility, TÜV Rheinland (automotive)
“UL Solutions is a global leader in applied safety science. We are pleased to announce our intent to collaborate with the NVIDIA Halos AI Systems Inspection Lab to harmonize testing activities and reports for companies who are pursuing safety certification with us.”
— Alex Dadakis, EVP, Chief Business Ops and Innovation, UL Solutions
“We are committed to fostering digital trust and delivering rigorous AI assurance at scale. By recognizing the inspection reports of the NVIDIA Halos AI Systems Inspection Lab, we’re supporting the industry’s move toward more transparent, reliable, and secure AI — while enabling developers to bring innovative systems to market faster and more safely.”
— Vincent Sabot, CEO, CertX
Algorithmic AI safety spans:
Building a safer AV ecosystem includes:
Certification
Independent third-party safety and cybersecurity assessments of NVIDIA Halos elements demonstrate NVIDIA’s ongoing commitment to AV safety.
ANAB accredited the NVIDIA Halos AI Systems Inspection Lab as an ISO/IEC 17020 Inspection Body. NVIDIA is the first company accredited by ANAB for an inspection plan that combines cybersecurity, AI, and functional safety.
TÜV SÜD certified the core NVIDIA hardware and software process to Automotive Safety Integrity Level (ASIL) D. Under the ISO 26262 standard, NVIDIA DriveOS 6.0 is certified ASIL D conformant and Thor-X SoC is assessed as ASIL D conformant. NVIDIA also received ISO/SAE 21434 Cybersecurity Process certification for its automotive system-on-a-chip, platform, and software engineering processes.
TÜV Rheinland performed an independent United Nations Economic Commission for Europe safety assessment of NVIDIA DRIVE AV related to safety requirements for complex electronic systems.
Research
Our research and development have published 330+ research papers on autonomous vehicle safety.
Provides tools and guidelines to build a credible safety case from L2 ADAS to L4 robotaxis, drawing on 330+ research papers and 1,000+ patents within NVIDIA Halos.
Comprehensive evaluations with open-loop metrics, closed-loop simulation, and real-world vehicle tests demonstrate that Alpamayo 1 is state-of-the-art in multiple aspects (including reasoning, trajectory generation, alignment, safety, latency, and more).
Collecting and annotating real-world data for safety-critical physical AI systems, such as Autonomous Vehicles (AVs), is time-consuming and costly. To address this challenge, we introduce the Cosmos-Drive-Dreams, a synthetic data generation (SDG) pipeline for generating challenging scenarios to facilitate downstream tasks such as perception and driving policy training.
Safety remains a fundamental challenge in autonomous driving, with a key step being the development of a safety evaluator that can reliably identify unsafe (i.e., collision-prone) scenarios.
HydraSafe is a framework that addresses the challenge of ensuring autonomous vehicle safety in hazardous scenarios by improving data availability and planner robustness.
The advent of end-to-end autonomy stacks—often lacking interpretable intermediate modules—has placed an increased burden on ensuring that the final output, i.e., the motion plan, is safe in order to validate the safety of the entire stack.
For a list of additional AV Research papers, click here.
Designed to handle real-world scenarios, Alpamayo accelerates safe Level 4 AV development by enabling reasoning-based autonomy through a family of open VLA models, physical AI datasets, and simulation frameworks.
Partners
Leading robotaxi companies, OEMs, industry safety pioneers, mapping and simulation companies, and software and sensor providers worldwide are using the system to deliver autonomous vehicle safety at all levels of automation.
Resources
How do you scale autonomous vehicles to large fleets while meeting the world’s strictest safety standards? This livestream presented by NVIDIA safety experts shows how the DriveOS stack and Hyperion reference architecture solve the safety-scalability problem.
In this livestream, we showcase how the newly launched NVIDIA Alpamayo autonomous vehicles open dataset creates new opportunities to advance AV safety.
Learn about NVIDIA Halos AI Systems Inspection Lab—the first ANSI National Accreditation Board (ANAB)-accredited lab dedicated to physical AI systems.
In this livestream, we present a general, theoretically grounded framework for AV safety validation whereby real-world tests are paired with simulated tests on corresponding reconstructed scenarios.
Learn how cutting-edge AI, rigorous validation frameworks, and global standards are shaping autonomous vehicle safety.
Sign up for the latest news and updates from NVIDIA.
NVIDIA solutions give you the performance and scalability to design, visualize, develop, and simulate the future of driving.