Best of LinkedIn: Next-Gen Vehicle Intelligence CW 09/ 10

00:00:00: Brought to you by Thomas Allgeier and Frennis, this Adrien highlights key LinkedIn posts on next-gen vehicle intelligence in weeks nine and ten.

00:00:09: Frenis supports automotive enterprises and consultancies with market and competitive intelligence decoding disruptive technologies customer needs regulatory change And competitive moves.

00:00:20: so product teams and strategy leaders don't just react but shape the future of mobility.

00:00:25: Welcome to the deep dive, everyone.

00:00:27: We are really excited to have you with us today.

00:00:29: Yeah!

00:00:30: we've got an amazing lineup.

00:00:31: Today were exploring the top next-gen vehicle intelligence trends that was seen across LinkedIn during calendar weeks nine and ten.

00:00:39: It is a packed agenda the reality of vehicle software platforms, the geopolitical race happening in mobility right now and that big pivot in autonomous driving.

00:00:49: Right!

00:00:49: And we'll also get into ecosystem collaboration...and of course the invisible threats in automotive cybersecurity…it's a lot to cover

00:00:56: It really is.

00:00:57: So let's just dive right into it.

00:00:59: Let's start with the foundational shift that is basically defining the whole industry now, The transition from hardware-first engineering to software defined vehicles or SDVs.

00:01:09: Yeah...the big buzzword!

00:01:11: Exactly.

00:01:12: and I want kick things off a post for Michael Beller.

00:01:16: He had this really optimistic take about vehicles becoming these updatable intelligent systems.

00:01:21: Which was the dream right?

00:01:22: Right.

00:01:23: But then Sabine McGough came in with a massive reality check.

00:01:28: She compared this whole shift to SDVs, to astronomy's great filter.

00:01:32: Oh wow that is an aggressive analogy.

00:01:34: It

00:01:35: is but it makes so much sense.

00:01:36: like the Great Filter Is This Structural Barrier where OEMs essentially have to get much slower and endure all these rising costs before they can actually get faster.

00:01:46: And reap the benefits of a platform?

00:01:48: Yeah, I mean it's a structural desert.

00:01:49: They have to cross in Vinicius to do design really built on this with an economic perspective.

00:01:54: What did he say?

00:01:55: well He argued that SDV fragmentation isn't just a technical issue.

00:01:59: You know It fundamentally changes The cost curve.

00:02:01: so if you Have fragmented life cycles across All your different car models Your costs don't Just grow linearly.

00:02:08: they Grow super Linearly

00:02:10: Wow.

00:02:10: Okay, let's unpack this for a second.

00:02:12: is building an SDV platform essentially like trying to replace the foundation of a house while The family still living inside it?

00:02:19: that Is exactly what?

00:02:20: It's like you have To keep the legacy systems running to pay the bills While ripping out the floorboards.

00:02:26: and That brings me to Marcus Rett Stetz provocative thought on This.

00:02:29: he said that if in OEMs still needs Like A spreadsheet to figure out what breaks when a requirement changes, then it's not a software-defined vehicle.

00:02:39: Oh I saw that!

00:02:39: Yeah he has the PowerPoint defined vehicle.

00:02:42: The truth has to live in git Not in management dashboards

00:02:45: Powerpoint Defined Vehicle.

00:02:47: That is brilliant and so true For you listening whether your an engineer or strategy leader... ...the takeaway here pretty stark Which

00:02:53: Is

00:02:54: If your organization isn't willing to endure really painful structural transition of shifting power away from local vehicle programs, the technical architecture isn't going to save you.

00:03:06: It's culture over code basically?

00:03:08: Exactly and if that internal struggle wasn't hard enough we have to look at how different regions are handling this pressure right.

00:03:14: moving from the internal OEM struggles to the global battlefield.

00:03:18: yeah and the velocity of innovation here is just wild.

00:03:22: It really

00:03:23: is.

00:03:23: H-Ming Wino shared this fascinating data point, did you see the numbers on fleet ages?

00:03:28: I did but break it down for everyone.

00:03:30: So China is iterating its SDVs on a fleet with an average age of just five point five years.

00:03:36: Meanwhile The US and Europe are trying to debug things On legacy fleets that averaged twelve plus years old.

00:03:42: That is massive gap!

00:03:44: It's

00:03:44: huge!

00:03:45: Mangrino pointed out that China competes on velocity while the West still competing on life cycles.

00:03:50: Yeah Syed Hasip Hasan layered into this perfectly.

00:03:53: He brought up Chancellor Friedrich Merz's recent comments on Germany's productivity gap,

00:03:58: right?

00:03:58: And just to jump in here for you listening we want to be super clear.

00:04:02: We are not endorsing any political viewpoints regarding Trump or tariffs or anything like that.

00:04:08: We're just impartially reporting on the strategic market dynamics from these sources

00:04:12: Exactly.

00:04:13: Purely looking at the business side, and Hassan explains that the architectural difference is what matters here.

00:04:19: EU OEMs validate for perfection in a lab.

00:04:22: they want it flawless before it hits the road which

00:04:24: takes forever

00:04:25: right?

00:04:25: but Chinese OEM's they validate with millions of users actually out on the road.

00:04:30: They learn in real time

00:04:31: And The brain consequences Of this race are really starting to show.

00:04:35: Pedro Pacheco noted That Audi Actually had A rely On the chinese oem I I am motors for their EV tech.

00:04:42: I remember reading that

00:04:43: yeah, and because of that Audi had to drop their prices And essentially dilute.

00:04:48: they're premium brand power.

00:04:50: They just couldn't move fast enough internally.

00:04:52: it's a tough spot To be in

00:04:53: and Philip brash highlighted something even wilder.

00:04:56: Ford CEO Jim Farley met with the Trump administration to propose bringing Chinese automakers To produce in the?

00:05:03: u.s via joint ventures,

00:05:04: which is just crazy when you think about the history.

00:05:06: totally it's reversing The exact playbook China used against the West forty years ago.

00:05:11: It

00:05:11: really is a full circle moment and it makes me think of Augustine Friedel concept at the four tribes of automotive.

00:05:17: Oh yeah walk us through that.

00:05:19: so he maps out the industry from specialized hardware makers all the way two ecosystem players.

00:05:26: And his main point is that the most dangerous place to be right now isn't just being a legacy hardware maker.

00:05:32: What's worse than that?

00:05:33: Being a legacy Hardware Maker, that deludes itself into thinking it's a nimble software ecosystem player.

00:05:39: Ouch!

00:05:40: Yeah pretending to be tech company when you're still relying on spreadsheets

00:05:44: Exactly You can't fake anymore.

00:05:46: Well and because this geopolitical and technical race Is too fast for anyone OEM To win alone They are forced to collaborate.

00:05:55: Which brings us to the ecosystem conversation.

00:05:58: Right, Brian Carlson and Matt Damascino posted about The SDverse Marketplace celebrating its second anniversary.

00:06:04: It's basically this hub for new collaboration.

00:06:06: Yeah places share in by software components.

00:06:09: But

00:06:09: here is my pointed question For you if everyone Is sharing open-source Software like who actually takes the blame when the brakes fail?

00:06:16: That is the ultimate Question.

00:06:17: right?

00:06:17: And Naveen C broke This down beautifully By comparing Eclipse S core and Alloy Core.

00:06:22: Okay what's the difference?

00:06:23: Well!

00:06:25: That's Desco R is where the open collaboration happens.

00:06:28: It's fast, it's open source

00:06:30: but maybe not road ready

00:06:32: exactly.

00:06:33: you absolutely need a foundational safety certified layer like alloy core sitting underneath right

00:06:39: to handle The liability in the audits.

00:06:41: yes Open Source is great for innovation But it has to be as Naveen says shipable.

00:06:47: You need release discipline when people's lives are on the line.

00:06:50: That makes total sense.

00:06:51: And this ecosystem expansion isn't just happening in software, it's applying to connectivity too!

00:06:56: David Fidalgo pointed out that connectivity is not an add-on radio component anymore.

00:07:01: No its the core architecture.

00:07:02: Right and Keven Carger highlighted BMW partnering with NTT Docomo.

00:07:06: Oh

00:07:07: the telecom partnership.

00:07:08: Yeah,

00:07:08: telecoms are literally becoming tier one automotive suppliers now.

00:07:12: Massive over-the-air updates and cloud assisted A days.

00:07:15: They require a five G digital backbone.

00:07:17: Yeah, the value team has just permanently expanded.

00:07:20: you can't build a smart car with a bad cell connection

00:07:22: Exactly.

00:07:23: So with all this computing power and connectivity, what does it actually mean for the driver?

00:07:27: Like... What about the promise of self-driving cars?

00:07:29: The autonomy reality check?

00:07:31: Yes!

00:07:32: Let's get into safety in sensors.

00:07:35: I loved this fun fact from Sony Andrews Jobudas.

00:07:38: He mentioned that these tiny steering angle sensors are quietly powering lane departure warnings And they've reduced relevant crashes by eleven percent.

00:07:47: That

00:07:48: huge for such a small component

00:07:50: It is, but despite those sensor advancements there's a massive reality check happening in higher level autonomy.

00:07:57: Partha Goswami and Andrea Leitner both pointed out that BMW and Mercedes are actually abandoning eyes off Level Three Autonomy.

00:08:04: They're

00:08:04: stepping back from it?

00:08:04: Yeah to refocus on Level Two Plus.

00:08:06: Well I can explain why this happened.

00:08:08: Level three has incredibly conflicting requirements

00:08:11: Like what.

00:08:12: So the system tells driver hey you look away But then it demands that they take control instantly if the system fails.

00:08:19: Which is a terrifying handoff!

00:08:20: Exactly, and it carries immense legal liability for the automaker Plus.

00:08:25: as Roger C Langtod points out companies like GM & Tesla are finding level two plus is just massive subscription revenue gold mine.

00:08:33: Oh because it keeps the driver legally engaged?

00:08:36: Yes

00:08:37: The drivers legally responsible but their comfortably assisted.

00:08:40: It's way less risk for the OEM, but they still get the recurring revenue.

00:08:44: So

00:08:44: wait what does this all mean?

00:08:45: For the automotive engineer?

00:08:46: are they just becoming like glorified IT workers?

00:08:50: not at All.

00:08:50: Chauncey had a brilliant insight on this.

00:08:52: he pointed out that A modern vehicle has hundred million lines of code making thousands Of probabilistic decisions.

00:08:59: a

00:08:59: hundred million.

00:09:00: That's hard to even wrap your head around

00:09:02: right.

00:09:02: and his point is there software won't replace engineers it will expose them.

00:09:07: engineering is shifting from precision coding to judgment.

00:09:10: Judgment?

00:09:11: I like that!

00:09:12: Yeah, what should a four-thousand pound machine do when it doesn't know What To

00:09:15: Do?!

00:09:15: That takes deep engineering judgement not just basic coding.

00:09:18: But When vehicles are relying entirely on AI and software And cloud connectivity...what happens when those systems Are maliciously manipulated?

00:09:28: The dark side of SDVs.

00:09:30: Let's dive into our final theme the invisible threats in automotive cybersecurity.

00:09:36: Bob Carver brought up a really terrifying point about AI adversarial attacks.

00:09:40: Oh, the stop sign example?

00:09:41: Yes!

00:09:42: He pointed out that you don't even need to hack a car's computer to crash it.

00:09:45: You just alter a few invisible pixels on a physical stop sign

00:09:49: And a human wouldn't notice it.

00:09:50: Right

00:09:51: But the cars' AI misclassifies as speed limit signs.

00:09:54: That is genuinely scary.

00:09:56: and broader attack surface is exploding.

00:10:00: Tonskip noted that attacks on automotive organizations doubled in twenty-twenty five.

00:10:04: Doubled?

00:10:05: In one year!

00:10:06: Yes,

00:10:06: and they're moving away from the vehicle itself.

00:10:09: now They're targeting cloud APIs.

00:10:11: Yeah And Dennis Laskoff had a warning about those OBD two dongles.

00:10:15: you know The ones people plug in for insurance discounts.

00:10:17: Oh yeah the companion apps.

00:10:19: right there's actually massive backdoors.

00:10:21: You found that seventy percent of them have security vulnerabilities.

00:10:24: That is wild.

00:10:25: You think you're saving money on insurance and just giving hackers the keys to your car?

00:10:29: Exactly, even embedded storage is a critical vulnerability now as Stefan Schumacher pointed out.

00:10:35: So

00:10:35: how do automakers begin to defend against all of this?

00:10:39: Imran Khan uses this great house framework analogy to synthesize these threats.

00:10:45: He says vehicles don't fail neatly inside individual standards anymore.

00:10:49: What do you mean?

00:10:50: They fail at the messy boundaries between different safety protocols So between functional safety, SODIF which is the safety of the intended functionality AI Safety and cybersecurity.

00:11:01: It's all connected

00:11:02: right?

00:11:03: if someone hacks your AI it causes a SODIFT failure Which leads to a physical safety failure.

00:11:08: you can't just isolate them?

00:11:09: Which is exactly why Vikas Chaudhary made that huge prediction for twenty-twenty eight.

00:11:13: He said cybersecurity won't just be an IT headache anymore.

00:11:16: It'll be a brand differentiator

00:11:18: Exactly, and a strict requirement for market access globally.

00:11:21: If you can't prove your car is secure You literally won't be allowed to sell

00:11:24: it.

00:11:38: Also check out our other editions on electrification and battery technology, future mobility in market evolution and commercial fleet insights.

00:11:45: Thank you so much for joining us on this deep dive.

00:11:47: don't forget to subscribe.

00:11:49: And I want to leave you with one final lingering question to ponder.

00:11:53: Let's

00:11:54: hear it

00:11:54: If software updates allow a vehicle to continuously evolve and improve its capabilities years after It leaves the factory.

00:12:01: are we approaching The end of the traditional model year?

00:12:05: Think about it soon.

00:12:06: Buying a twenty-twenty six car might mean absolutely nothing because the car you drive in twenty thirty will be fundamentally smarter than the day You bought.

00:12:14: that is a wild thought.

00:12:15: to end on thanks for tuning In everyone.