Member Info for Whatif18

MBG US Capital Market Event & AMG World Premiere, Los Angeles, 19th May. We have seen this week how their adww is all over the place.

Taking this as the key phrase - "the program marks a significant expansion of magna’s dms/oms across global vehicle platform"

VW Group made 9m Cars in 2024. Stellantis made 5.5m.

Renault, BMW and Mercedes are all around 2.4m

Looking for why, late this afternoon (19Th May)

Stellantis have an Investor Day on May 21, 2026,

What looks like two Very small single LEDs above the central screen. Perhaps each side of a camera

21:43 to go

https://www.youtube.com/watch?v=HttzghRJ4Ls&t=765s

I doubt this is SEE

And 1 SEK = 7.95p

Rav 4 Pod behind steering wheel. Does not appear very often.

42:25

https://www.youtube.com/watch?v=7nvboYcebpY

Another question to AI

How strong is the link between toyota and Renesas

The link between Toyota and Renesas is extremely strong and deeply strategic, functioning as a critical partnership for advanced automotive technology rather than just a simple buyer-supplier relationship. As of 2026, Renesas is a core semiconductor provider for Toyota’s next-generation vehicles. [1, 2]

Key Aspects of the Relationship:

Next-Gen ADAS & AI (2026): Renesas’s R-Car V4H system-on-chip (SoC) has been selected for the ADAS (Advanced Driver Assistance System) control unit in the 2026 Toyota RAV4, as well as AI-powered features.

Deep Integration via Denso: The collaboration often flows through Denso (a key Toyota affiliate), which uses Renesas’s technology for Electronic Control Units (ECUs) in Toyota’s autonomous-driving and high-end multimedia systems.

Strategic Investment: Following a 2018 investment where Denso purchased a 4.5% stake in Renesas, the two companies work closely on developing technologies for "connected cars" and autonomous driving.

Long-Term "Rescue" Bond: The partnership has historical depth; in 2012, Toyota and its affiliates (including Denso) invested in Renesas to help stabilize the chipmaker during a financial crisis, cementing a partnership focused on ensuring supply.

Mission-Critical Supply: Renesas is a vital, high-volume supplier of microcontrollers and system-on-chips. Disruptions at Renesas plants (such as the 2021 fire) directly cause production stoppages at Toyota, highlighting their deep dependence. [1, 2, 3, 4, 5, 6, 7]

In summary, Toyota relies on Renesas for its automotive "brains," while Renesas relies on Toyota for high-volume adoption of its advanced R-Car and MCU products, making them interconnected allies in the software-defined vehicle era. [1, 2]

I asked AI how does Renesas and seeing machines would involve Ceva. This was the reply. Very interesting.

The integration of Renesas, Seeing Machines, and Ceva forms a classic automotive silicon "sandwich"—where Renesas serves as the physical silicon platform, Ceva provides the internal hardware accelerator IP, and Seeing Machines supplies the specialized software layer running on top.The three companies come together through a layered chip architecture and a shared customer ecosystem:1. The Silicon Architecture (The Core Link)Renesas does not design every component of its R-Car System-on-Chip (SoC) from scratch; instead, they license highly specialized sub-processors from IP companies like Ceva.The Hardware Acceleration: Renesas integrates Ceva's NeuPro-S or SensPro AI Digital Signal Processors (DSPs) directly onto the silicon floorplan of the R-Car V4H platform. This hardware is explicitly designed to handle intensive computer vision workloads using minimal power.The Software Execution: When Seeing Machines deploys its driver-monitoring software onto a Renesas R-Car chip, its eye-tracking algorithms and deep learning networks are processed and accelerated by the Ceva AI hardware embedded inside that chip.2. Tier-1 & OEM System IntegrationFor a car manufacturer like Toyota, these three distinct layers are presented as a singular, unified solution inside an Advanced Driver Assistance Systems (ADAS) or domain control unit.Ceva licenses its AI architecture to Renesas.Renesas manufactures the finished R-Car V4H chip.Seeing Machines optimizes its code to run flawlessly on Renesas' hardware.Tier-1 Suppliers (like iMotion or Denso) buy the Renesas hardware, flash the Seeing Machines software onto it, and deliver the final complete "smart camera" module to the automaker.3. Shared Industry PipelinesBecause the automotive industry requires strict safety compliance (like ISO 26262 / ASIL-B standards), all three companies design their respective parts to complement each other:Ceva's DSPs are certified for functional automotive safety.Renesas wraps that IP into a safety-certified computing platform.Seeing Machines utilizes that verified safety-critical processing layer to comply with strict global safety regulations (like Europe's General Safety Regulation for driver drowsiness detection).

According to AI, (could be wrong), the RS5 is using Smart eye and the Q3 is using Seeing machines. It said the Q3 was using the same platform as the Golf.

In the Audi RS5 at 9:50. Pod behind steering wheel. Two LEDs to lefthand side. One single LED to RHS

If you are quick you can pause at 9:50, restart and pause quickly again you can catch both patterns

https://www.youtube.com/watch?v=d0HYXGHQgGY

But in the Audi Q3 its in the mirror at 11:50 ( its faint and you need to concentrate)

https://www.youtube.com/watch?v=sW1N2QfkisY

So, looks like the Audi is dual sourced with the RS5 having 3 LEDs

I have tried pausing the videos . With the Mercedes you seem to get either the two left LEDs or the two right. But on an Audi I got two LEDs on the left and one on the right. I think it was a Q9. Bit more digging required. Perhaps it depends on the camera used to film the video ! or is it the SEE Patent !

Mercedes S-Class through wheel on dash

LEDs at 19:02

https://www.youtube.com/watch?v=T8qXTGjMEHQ

AI seems to think these are Seeing Machines but I am not sure. Any inputs.

Mercedes S-Class through wheel on dash

LEDs at 19:02

https://www.youtube.com/watch?v=T8qXTGjMEHQ

Mercedes c class LEDs edge of dash in centre of car. 16:48

https://www.youtube.com/watch?v=eFxdCbTM1lk

Looks like it

8:42 onwards in the mirror

https://www.youtube.com/watch?v=qQusLTq6tcc

I then asked

How much is the market worth and what is it worth to seeing machines ?

The total global robotics market is projected to be worth US$53.64 billion, with the highly lucrative AI-driven robotics sector tracking toward US$38.0 billion. If this corporate trio targets specialized industrial automation, the global industrial robot market represents an independent US$25.66 billion opportunity. [1, 2, 3]

Within this alliance, revenue distribution is highly unequal because hardware manufacturers absorb the majority of top-line sales, while software vendors operate on ultra-high margin percentages.

What Seeing Machines Would Receive

Seeing Machines functions as an intellectual property (IP) and software-licensing entity. Their standard commercial blueprint—heavily proven by their existing automotive contracts—reveals exactly how they would extract value from a joint Mitsubishi and Qualcomm product: [1]

Software Licensing / Per-Unit Royalties: Seeing Machines does not bill for the physical robot. Instead, they charge a fixed royalty fee for every individual machine produced that activates their software layer. In their automotive business, this exact model yields a 100% gross margin on royalty inflows, meaning every pound or dollar received drops directly into cash flow and corporate earnings.

Upfront Non-Recurring Engineering (NRE) Fees: Before mass production begins, Mitsubishi would pay Seeing Machines upfront NRE fees. These cash payments cover the engineering hours required to customize the eye- and attention-tracking algorithms for Qualcomm's chips and Mitsubishi's custom physical camera placements.

Annualised Recurring Revenue (ARR): If the robotic product utilizes continuous safety tracking, fleet tracking, or cloud analytics (similar to Seeing Machines' existing Guardian fleet ecosystem), they would receive ongoing monthly or annual software-as-a-service (SaaS) subscription fees per active robot.[1, 2, 3, 4, 5]

The Financial Split: Who Gets What?

If a specialized co-bot or industrial vehicle built by this trio sells commercially for $50,000, the structural revenue split roughly mirrors this distribution:

[ $50,000 Total Robot Sale Price ]

├── Mitsubishi (Hardware & Integration) ──► ~$43,000 (86% - Covers steel, motors, assembly, sales)

├── Qualcomm (SoC Board & 5G Modem) ──► ~$5,000 (10% - Silicon hardware hardware layer)

└── Seeing Machines (Human AI License) ──► ~$2,000 (4% - 100% pure-profit software royalty)

Mitsubishi captures the lion's share (~80-85%) of gross revenue to offset the heavy capital expenditures of manufacturing, metal casting, physical component sourcing, and international product distribution.

Qualcomm secures a hardware component fee (~10-12%) for supplying the high-end processing chipsets, motherboards, and integrated 5G modules.

Seeing Machines secures a smaller portion (~3-5%) of the total machine cost.

I asked Google in AI mode a slightly different question.

Could seeing machines software be integrated into Qualcomm Dragonwing or Snapdragon robotics processors ?

Yes, Seeing Machines software can absolutely be integrated into Qualcomm Dragonwing or Snapdragon robotics processors. While historically optimized for Qualcomm’s automotive chips, there are no hardware or architectural barriers preventing it from running on their robotics silicon. [1, 2]

In fact, industry crossovers are already occurring, as seen with the Motive AI Dashcam Plus which embeds edge-AI vision directly onto a Qualcomm Dragonwing QCS6490 chip. [1]

The feasibility, implementation, and use cases for this specific integration are detailed below:

⚙️ Technical Feasibility & Porting

Shared Architecture: Qualcomm’s Snapdragon automotive platforms and Dragonwing robotics platforms (such as the RB3 Gen 2, RB5, and IQ9 series) share a foundational architecture. They utilize similar Qualcomm Hexagon Neural Processing Units (NPUs), Kryo CPUs, and Adreno GPUs.

Software Libraries: Seeing Machines designs its Embedded Product Strategy (e-DMS) to be processor-agnostic software libraries. Because they already compile their algorithms to leverage Qualcomm's core acceleration blocks, porting the software stack to a Linux- or Ubuntu-based Dragonwing robotics environment is a straightforward software compilation task.

Sensor Ecosystem Partnerships: To assist with edge-AI integration, companies like Airy3D provide 3D passive depth sensing solutions that are already cross-compatible with both Seeing Machines software and Qualcomm Dragonwing robotics platforms (such as the RB5). [1, 2, 3, 4, 5]

💡 Primary Use Cases in Robotics

Integrating Seeing Machines into a robotics processor shifts the software's intent from "Driver Monitoring" to "Human-Robot Interaction and Safety": [1, 2]

Humanoid Robots (Fluid Collaboration): For advanced humanoids built on platforms like the Dragonwing IQ series, the software can track human eye-gaze and attention. This allows a robot to know if a human coworker is looking at it for instruction, is distracted, or requires assistance.

Industrial Operator Safety: In smart factories or heavy industrial yards, a Dragonwing-powered robotic arm or automated guided vehicle (AGV) can utilize the software to continuously monitor the machine operator's fatigue, cognitive awareness, and line-of-sight before engaging high-risk automated tasks.

Home Care Robotics: For robots looking after elderly or vulnerable individuals, tracking subtle facial movement cues, eye gaze, and drowsiness is crucial for assessing a patient's physical state or cognitive distress. [1, 2, 3, 4]

Worth a read in the absence of other news.

https://www.forbes.com/sites/tiriasresearch/2026/05/14/ai-is-the-key-powering-qualcomms-rise-to-top-of-automotive-stack/

Rewind to the other shares that were sold at 16:08. 5.05p was a buy

If it is Tata Motors they make 35% of trucks in India (400,000 trucks) about the same as Daimler in Europe. Could explain the rise in the SP

Issued today. Could it be Tata Motors ?

https://www.valeo.com/en/valeo-selected-by-a-leading-automaker-to-deploy-advanced-adas-in-india/

Https://www.valeo.com/en/valeo-to-showcase-future-ready-trucking-innovations-at-act-expo-2026/

Looks like something was ready well before May 4th

Discussion is from 34:40 to 35:48

By the end of the year a Tier 1 will install a system and Seeing Machines will get a payment and they may get a monitoring contract.

So what happens in 55 days time ? (They actually only have 38 days working days to sign off the GSR certification, from now. )

All the vans and trucks first registered in Europe, after 7th July , need to be compliant. How can it wait 6-9 months for the new business model for Trucks. Can someone explain what I am missing?

Well worth watching. Much more searching questions.