Member Info for Whatif18

page 3

magna's shift to nvidia has a complex impact on qualcomm. while it highlights a direct loss in the "tier 1 integration" space, qualcomm remains a dominant force by winning in the "digital ****pit" and "mass-market" segments. the impact can be broken down into three key areas:

direct competition in high-compute adas magna's use of nvidia drive agx thor (delivering 2,000 tops) directly competes with the qualcomm snapdragon ride flex platform. • the loss: magna was originally set to acquire veoneer, which would have made it a major partner for qualcomm's arriver software. however, qualcomm outbid magna to buy veoneer itself, effectively turning magna into a competitor in the adas software stack market.

performance gap: nvidia's thor is currently positioned for level 4 autonomy, whereas qualcomm's snapdragon ride is primarily dominating the level 2/2+ mass market due to better power efficiency and lower costs.

market segmentation

automakers often split their business between these two giants: • nvidia (performance): used by brands like mercedes-benz and lucid for high-end autonomous driving where raw ai power is the priority.

qualcomm (connectivity & ****pit): qualcomm leads in the digital ****pit space (infotainment and 5g connectivity). for example, volvo uses nvidia for its adas "brain" but relies on qualcomm for its in-car screens and audio.

collaboration with seeing machines

interestingly, both nvidia and qualcomm are on the same team when it comes to the driver monitoring (dms) inside the mirrors. • seeing machines has formed strategic partnerships with both companies.

whether a car uses a magna/nvidia setup or a qualcomm snapdragon digital chassis, the actual eye-tracking software often remains the same, meaning qualcomm still benefits from the trend even if nvidia wins the main adas chip slot.

in summary: qualcomm has lost the chance to have magna as its primary "integration arm" for adas, but it is maintaining its lead in the software-defined vehicle (sdv)market through its broader snapdragon elite platform.

Page 2

Key Benefits of this Combined Stack

Compliance: This integrated solution allows automakers to meet strict Euro NCAP and global regulatory standards for driver distraction and impairment.

Space Efficiency: By moving the DMS camera into the mirror and the processing to the central NVIDIA chip, it eliminates the need for bulky standalone camera modules on the steering column.

Advanced AI: NVIDIA Thor's architecture is optimized for the Transformer models that next-gen Seeing Machines algorithms use to track eye-gaze and cognitive state more accurately.

This shift is a significant blow to Mobileye’s market dominance in the ADAS space. While Mobileye remains a powerhouse, Magna’s move toward NVIDIA highlights a critical trend: the industry's departure from "closed" vision chips to "open" high-compute platforms.

Loss of Tier 1 Integration Volume

For years, Magna was one of Mobileye’s largest Tier 1 partners, building high-volume "front-camera" modules based on the EyeQ series. As Magna pivots its development focus toward NVIDIA, it reduces the future "pipeline" of vehicles that will come factory-equipped with Mobileye hardware via Magna.2.

The Battle for "Software-Defined Vehicles" (SDV)

The Mobileye Model (The "Black Box"): Traditionally, Mobileye EyeQ chips came as a bundled package of hardware and proprietary software. This "closed" ecosystem is becoming less attractive to major OEMs (like Mercedes and GM) who want to write their own driving algorithms.

The NVIDIA/Magna Model (The "Supercomputer"): NVIDIA Thor provides the raw horsepower, and Magna provides the integration. This allows car brands to own their "software stack," which is the primary reason brands are defecting from Mobileye's end-to-end solutions.

Financial Pressure

Mobileye has recently faced revenue headwinds and lowered guidance as major customers (like Volkswagen and Ford) re-evaluate their long-term autonomous strategies. Magna’s strategic partnership with NVIDIA accelerates the "commoditisation" of the standard ADAS features Mobileye once monopolised.

Where Mobileye is Fighting Back

Mobileye is not standing still; they are moving away from simple chips to advanced platforms like SuperVision and Chauffeur. However, by partnering with NVIDIA, Magna is positioning itself to capture the Level 3 and Level 4 autonomous market, which requires the massive 2,000 TOPS of compute that Mobileye’s current EyeQ Ultra is still working to match in sheer scale.

I have asked AI several questions about Magna , Nvidia / Qualcomm. I know it could be wrong but it seems very possible

Part 1

Magna has significantly shifted its primary chip partnerships for next-generation ADAS and mirror-integrated systems. While they have a long-standing history with Mobileye, they have recently formed a major strategic partnership with NVIDIA to power their most advanced upcoming platforms.

Key changes in Magna's chip and technology suppliers include:

Move to NVIDIA: For its next-generation (Gen 2) vehicle intelligence and ADAS applications, Magna is integrating the NVIDIA DRIVE AGX Thor system-on-a-chip. This shift supports "software-defined vehicles" and high-level autonomous driving (L2+ to L4).

Continuation of Mobileye: Magna has historically used Mobileye EyeQ chips (notably the EyeQ5) for its "one-box" front-camera systems. While NVIDIA is the focus for future high-compute platforms, Mobileye remains a established partner for many existing vehicle models.

Driver Monitoring (DMS) Partnerships: For its mirror-integrated DMS (which recently scaled production with a major German OEM), Magna collaborates with Seeing Machines for the underlying vision technology.

Strategic Silicon Carbide (SiC) Partnership: For power electronics related to electrification, Magna signed a long-term agreement with onsemi.

Seeing Machines remains a core technology partner for Magna's mirror-integrated monitoring systems, even as Magna transitions its underlying compute platform to NVIDIA DRIVE AGX Thor.

While NVIDIA provides the high-performance "brain" for the vehicle, Seeing Machines provides the specialized vision algorithms and optical path expertise specifically for driver and occupant monitoring.

How the Partnership Continues

• Exclusive Collaboration: Magna and Seeing Machines have an exclusive agreement to co-market driver and occupant monitoring systems (DMS/OMS) specifically integrated into interior rear-view mirrors.

Centralized Integration: The NVIDIA DRIVE Thor platform is designed to centralize multiple functions—including ADAS, infotainment, and driver monitoring—onto a single chip. Magna will integrate Seeing Machines' FOVIO algorithms into this centralized environment to handle cabin safety.• Mass Production Scaling:

Magna confirmed in late 2025 that its DMS-integrated mirrors, powered by Seeing Machines technology, are being launched with multiple global automakers, with volumes expected to reach several million units annually.

System Responsibilities: In this setup, Magna acts as the Tier 1 integrator, coordinating the NVIDIA hardware, the Seeing Machines software, and the physical mirror assembly.

Volvo, Polestar, and Rivian together are about 1/10th Volkswagen and about 3% of the known SEE customers. Like everyone else I would like to know what the other OEMs are going to do.

Ai seems to think its 32million no billions.

On January 29, 2026, Mitsubishi Electric announced a significant strategic deal to invest ¥5 billion (approximately $32 million) in Akari Inc., a construction-tech and robotics startup.

Key Details of the Deal

The Investment: Mitsubishi Electric is backing Akari to accelerate the development of AI-driven factory robots.

Core Objective: The partnership focuses on teaching robots to recalibrate their movements in real-time, allowing them to master multiple, complex tasks rather than being limited to single, repetitive actions.

Strategic Goal: This deal is a major step toward full factory automation, a critical priority for Japanese manufacturers facing severe labor shortages due to a shrinking population.

Technology Synergy: Mitsubishi plans to integrate Akari's AI and sensing know-how into its existing industrial electronics and factory automation systems to create more adaptable "situational" robotics.

My theory - it could be the new Nvidia thor chip running SEE software.

I thought the one single LED, although a Purple/ pink colour was odd. Nvidia did say they had a new Thor chip at the end of this month which was very fast. So I asked AI if a fast chip could reduce the number of LEDs. I know it could be wrong but with CB not wanting to say who else do you ask! Also worth a thought- Collins had a link to the new Thor chip as well.

AI response

While a faster chip like the NVIDIA DRIVE Thor does not inherently change the physics of how light works, its advanced processing power can indirectly lead to a more efficient Driver Monitoring System (DMS) that uses fewer or smaller hardware components.

Centralisation and Integration

The primary advantage of the DRIVE Thor chip is its ability to centralise multiple functions—including automated driving, infotainment, and driver/occupant monitoring—onto a single system-on-a-chip (SoC).

Reduced Complexity: By consolidating these features, manufacturers can reduce the overall number of distributed electronic control units (ECUs), cables, and hardware components.

System Synergy: A more powerful chip allows for more sophisticated AI models (like Transformer engines) that can extract more detailed information from lower-quality or less-illuminated camera feeds.

Potential for Fewer LEDs

A faster, more intelligent chip like DRIVE Thor could potentially allow a DMS to function with fewer IR LEDs for several reasons:

Enhanced Image Processing: With 2,000 teraflops of performance, Thor can run complex "Reasoning Vision Language Action" models that are better at interpreting images in low-light conditions. This may reduce the need for high-intensity infrared illumination (fewer LEDs).

Dynamic Exposure Control: The chip's high-speed sensor processing (via 4x 25 GbE networking and dedicated accelerators) enables real-time adjustments to camera exposure and LED pulse timing, potentially making the system more efficient.

Better Data Analysis: Higher processing speeds mean the system can perform neural rendering and advanced noise reduction, allowing it to "see" the driver's eyes and head position more clearly even if the physical lighting is minimal.

Standard DMS Hardware Requirements

Regardless of the chip's speed, a DMS still requires specialized IR-LED drivers to ensure consistent brightness and reliability.

Infrared Necessity: IR LEDs (typically 850nm or 940nm) are necessary for DMS to work in complete darkness without distracting the driver with visible light.

Hardware Limitation: Even the fastest chip cannot replace the need for at least one or two IR sources to create the necessary contrast for the camera to track facial features at night.

These articles explore how NVIDIA's DRIVE Thor chip can reduce the number of LEDs in driver monitoring systems by improving image processing and integration.

tried an ai search - afari, stradvision, geeley, seeing machines. time will tell.

in january 2026, afari, stradvision, geely, and seeing machines have converged as key players in a unified ecosystem for ai-native autonomous driving, specifically through the global launch of the g-asd (geely afari smart driving) platform at ces 2026.

core partnership: g-asd and stradvision

g-asd platform: developed by afari (geely’s intelligent driving subsidiary), this system represents geely’s shift to an "end-to-end" ai architecture. it replaces traditional code with multimodal foundation models to coordinate driving, ****pit, and safety domains.

stradvision perception: stradvision is a primary technical partner, providing its svnet frontvision software for the g-asd platform. this software enables the high-performance object detection and scene understanding required for the 16 geely, zeekr, and lynk & co models already equipped with the system.

integrated safety: stradvision and seeing machines

joint collaboration: at ces 2026, stradvision and seeing machines announced their first public collaboration to demonstrate the fusion of exterior perception with interior driver monitoring.

holistic safety: the partnership integrates stradvision's external road sensing with seeing machines' driver monitoring systems (dms) to create a more responsive adas ecosystem. this allows the vehicle to react based on both external threats and the driver's real-time cognitive state (e.g., fatigue or distraction).

geely’s strategic role

the integrator: geely acts as the overarching oem that integrates these technologies into its full-domain ai 2.0 architecture.

commercial scale: geely has already deployed these integrated solutions in over 300,000 vehicles, providing the massive real-world data set needed to refine the ai models used by afari and its partners.

Horzabora - thanks very much . Futurelogic may come back with more information on whether non embedded software makes a certified chip. Best not ask AI

I know AI could be wrong. I did post the Zack report with the information and the source. Have you found any other evidence either way ? Where would you look to find out what software is preloaded onto a Qualcomm chip ? Always willing to learn.

I asked whether it contains Smart Eye and it said no !!

AI says Qualcomm’s Snapdragon Ride Flex SoC contains SEE software.

Qualcomm Incorporated ((QCOM Quick QuoteQCOM - Free Report) ) and Hyundai Mobis have signed a comprehensive agreement to co-develop next-generation solutions for Software-Defined Vehicles (SDVs) and Advanced Driver Assistance Systems (ADAS). The deal will allow Qualcomm to develop an integrated automotive solution designed to meet the needs of emerging markets and expand its reach in the global automotive supply chain.

Qualcomm’s Snapdragon Ride Flex SoC is a powerful automotive chip designed to support ADAS and autonomous driving, offering high-performance AI processing, real-time sensor fusion, and advanced safety features. It enables vehicles to handle complex driving and parking tasks efficiently while reducing power consumption and system complexity. Riding on this chip, Qualcomm and Hyundai Mobis will jointly develop advanced driving and parking solutions that support multiple vehicle functions on a single chip, helping automakers reduce costs, improve efficiency, and simplify system design.

The two companies further plan to combine their software and chip technologies in the near future to build a scalable, high-performance SDV architecture that is more efficient, stable, and upgradeable over the air.

https://www.zacks.com/stock/news/2815129/can-qualcomms-snapdragon-ride-flex-for-hyundai-work-wonders?cid=CS-NEWSNOW-HL-analyst_blog|quick_take-2815129

Finally, Qualcomm's MOU with Hyundai covers software-defined vehicles (SDVs) and autonomous driving systems. The agreement is also focused on Qualcomm's Snapdragon Ride platform, which will work together with Hyundai's Mobis smart mobility platform.

https://wccftech.com/qualcomm-expands-automotive-ai-portfolio-through-deals-with-hyundai-others/

The 6k are Aftermarket fittings. Come July will we not have 6k Aftermarket and a slice of about 25,000 large trucks per month when they are built? I think it should be more like 6k + 15k in the summer at least.

Not sure of the implications.

https://www.webwire.com/ViewPressRel.asp?aId=348972

It would help insurance companies assess driver attention, seat belt compliance. Attention at time of crash.

If it was built into the top of your TV it could assess response to adverts

Isb1974 - I think the Aviation contract may be waiting for Nvidia IGX Thor to be available later this month then Collins can announce their product . It could be soon in my opinion.

I doubt SEE will be taken over for the profit it makes. It will be taken over for the control of the technology or the potential for other things. It will depend on how many tech companies are interested and will pay a premium to take that control.

From Nvidia

https://nvidianews.nvidia.com/news/nvidia-releases-new-physical-ai-models-as-global-partners-unveil-next-generation-robots

NVIDIA IGX Thor, which will be available later this month, extends robotics to the industrial edge, offering high-performance AI computing with enterprise software support and functional safety. Archer is using IGX Thor to bring AI to aviation, advancing critical capabilities in aircraft safety, airspace integration and autonomy-ready systems.

Article also includes:

In addition, Caterpillar is expanding its collaboration with NVIDIA to bring advanced AI and autonomy to equipment and job sites in construction and mining. Caterpillar CEO Joe Creed will share details alongside NVIDIA Vice President of Robotics and Edge AI Deepu Talla during a CES keynote on Wednesday, Jan. 7.

I asked AI whether Collins and Seeing machines was waiting for Nvidia IGX Thor and got this reply. I know its AI and it could be wrong.

As of January 2026, Collins Aerospace and Seeing Machines are indeed utilizing NVIDIA’s newest computing platforms, including IGX Thor, to power their advanced aviation safety solutions.

Key details regarding their status and relationship with the Thor platform include:

Platform Availability: NVIDIA announced on January 6, 2026, that IGX Thor will be generally available later this month.

Integration in Aviation: Collins Aerospace and Seeing Machines are co-developing a Fatigue Risk Management System for the commercial aviation market. Seeing Machines’ eye-tracking and cognitive state monitoring technology is being integrated into Collins’ avionics suite.

Aviation Partners: While Collins is a primary partner for Seeing Machines in aviation, NVIDIA confirmed that Archer Aviation is already using IGX Thor to advance aircraft safety and autonomy-ready systems.

Future Mobility and Scale: Seeing Machines has recently established a Future Mobility Group (January 2026) to support the transition of automated vehicle programs into commercial deployment, which aligns with the production-ready capabilities of the Thor family.

Market Timeline: Production for the related NVIDIA DRIVE Thor platform (targeting Level 4 automotive applications) is slated for 2025–2026.

In summary, the companies are not just "waiting"; they are active development partners transitioning these high-performance AI platforms into production-ready aviation and industrial systems as the hardware becomes available this month.