The Dodge Ram is a large pickup truck used for tasks like carrying cargo and towing trailers. It also has a built-in screen and electronics for things like navigation and phone connections. If the podcast is talking about memory or connectivity, it’s likely referring to the truck’s infotainment system needing enough resources to run those features.
ADAS means “driver-assistance tech.” It’s the stuff in modern cars that helps you drive—like keeping you in the lane or braking automatically if there’s danger. Because it has to think quickly, it needs enough computer power and memory.
Infotainment is the car’s entertainment and information system—like the touchscreen, navigation, and music apps. “Advanced” infotainment usually means it’s more powerful and does more things, so it needs more computer memory.
Decontenting means taking features out of a product to save money. If chips are expensive or hard to get, a carmaker might remove some tech so the car doesn’t need as much of those parts.
Tier one suppliers are big companies that make major parts that go straight into cars. If a key chip gets more expensive, these suppliers often feel the cost pressure first because they’re the ones building the tech modules.
OEMs are the car companies that build the vehicles. The question is whether the automaker has to pay the extra chip costs, or whether the companies that supply parts pay first. Sometimes the OEM can negotiate or reimburse suppliers.
LIVE
This podcast is brought to you by Reynolds and Reynolds.
There are three waves of AI transforming our industry.
Explore what they are, where AI technology is headed,
and what these emerging innovations mean
for automotive retailers at rayray.com
slash AI dash waves for more info.
Hey, it's executive producer Jake Nier here in Detroit.
Welcome to this bonus episode of Daily Drive.
Earlier this week on Wednesday's show,
we heard from Kushal Fernandez,
a partner at Kearney, about a new report
on the DRAM memory chip shortage
and what it means for the auto industry.
Prices for DRAM chips have surged dramatically
in recent months as AI infrastructure companies
outbid virtually every other sector for supply.
Automakers are reporting hundreds of millions of dollars
in losses, and the question is,
how long it lasts and who weathers it best?
We only had time for part of that conversation on Wednesday,
today we're gonna hear the rest of it,
including why the auto industry actually sits
in a more favorable position than you might expect,
why EV makers have a structural advantage
over traditional automakers when it comes to securing supply
and what feature decontenting could mean
for the software defined vehicle push.
Here's the rest of that conversation.
You outline how different sectors are positioned
to compete with the AI industry for memory,
and some industries appear to be really in a tough spot
because they can't afford to outbid
AI infrastructure buyers on price,
but the auto industry is kind of at this middle category,
which actually surprised me to tell you the truth
because it appears to me that the auto industry
is squeezed from so many different directions,
but maybe it has the capital expenditure potential
to invest in memory contracts for a long time to come,
or can you just explain a little bit about why
the auto industry is sort of in that middle ground position?
Yeah, I think we look at it on two axes.
One is the ability to absorb price increases,
which either means you have the margins to absorb it,
or the ability to pass on the increase cost
to your customers.
And the second axis is really your ability to adapt,
which either means securing supply,
which in this situation is not,
you cannot outspend the AI players,
or your ability to streamline your designs
to quote unquote be a customer of choice
to the memory players,
that IE have less complexity or more modern components,
align better with supplier roadmaps and so on.
Now, the best way to talk about this
is to illustrate the spectrum.
If you look at a budget smartphone
that sells hundreds of millions of units every year,
largely to the developing countries,
they sell at $150 to $200,
in many cases even less than $100.
Now, all of these require a minimum of one gigabyte
to two gigabytes of RAM to operate
for you to be able to connect to the internet,
use WhatsApp, things like that.
That one gigabyte to two gigabytes of RAM
is adding $30 to $50 of cost.
So on a $100 to $200 device,
there are no margins that can absorb that,
and there is no ability to pass that on
to the end customer,
because that's a 20% price increase
that customers in those markets are not gonna accept.
So that side of the market is gonna face the brunt of it.
Now, if you go to automotive,
there's RAM is primarily a consumed
in higher quantities in more advanced cars,
infotainment systems, ADAS, these are the primary users.
So those tend to be more premium.
And in general, obviously the more high end,
the car, the more memory content in there.
But even at the high end,
if you look at it,
you're looking at maybe a $200 or $300 increase
on a car that already has a $40,000, $50,000 bill of material.
Not to say that automakers have margin to give away.
And as you said, they rightly so they squeeze
from many directions,
but it is relatively speaking
in terms of what the price impact would be,
not a significant amount to pass on, right?
Again, not to say that they aren't, as you said,
pressure from all around that they need to cooperate,
but relative to other industries,
they're not as bad of a spot
and the ability to either absorb that or pass that on.
I think in certain more entry level cars,
there will be an issue,
but there are ways feature decontenting
is an example of that, right?
I think you really have to think about,
I mean, the features that require memory,
as I said, advanced infotainment, ADAS.
So, if you really need to compete,
then you consider decontenting
or taking out those features
and you would not need the memory.
So is it the automakers who are capable of absorbing
or passing on those costs or are the suppliers?
Because I would think it would be
sort of tier one suppliers who would be dealing
with the brunt of that cost pressure,
but it sounds like you're talking about OEMs.
Yeah, I'm talking about OEMs, yes.
Ultimately, I think it is the tier ones
who would be facing the pressure.
Again, it depends on the OEMs.
Some OEMs control their ADAS and infotainment stacks,
obviously not a majority.
So it would be the tier ones,
but I think there is an understanding obviously
from OEMs and there's a reimbursement
that I have heard anecdotally from some suppliers
that they are being reimbursed for higher memory costs.
So I think it is a well understood dynamic
that prices are going up in the market.
And so, that is a mechanism that is happening
and that's why tier ones I think so far.
It may not be evenly across all tier ones,
but that is happening.
Three waves of AI are transforming our industry
as we know it today.
Here from AJ McGowan,
Vice President of Research and Development
at Reynolds and Reynolds
on all three ways of AI and automotive
where we're heading
and what the latest cutting edge innovations
mean for automotive retailers.
So I want you to picture what it would be like
on that same Tuesday morning that we talked about earlier
with a full agentic infrastructure in place.
Your service manager wouldn't start the day
by reading reports.
Ray hands them three ROs that need attention
and tells them why.
Your sales manager wouldn't spot check calls.
Ray already routed the ones that mattered.
The GM doesn't reconstruct yesterday.
Ray reconstructed it overnight and gave them a summary.
And your controller doesn't have to pull
the deal level analysis every Friday.
Ray runs it on Thursday and flags anything
that's worth a conversation.
Decisions can get made earlier in the day
with better information by fewer people.
And that's what the agentic wave
actually feels like in your store.
It feels and sounds a little bit like science fiction
but at the end of the day it's not.
It's just a faster Tuesday.
If you're interested in exploring how AI
can improve operations at your dealership
visit rayray.com slash ai dash waves.
That's r-e-y r-e-y dot com slash ai dash waves.
You also outlined some differences
between the internal combustion engine vehicles
that are seeking access to memory
and electric vehicles that are seeking access to memory.
Why are EVs and I guess EV manufacturers
and tier one suppliers serving EV manufacturers
better positioned to continue to access memory
than internal combustion engine manufacturers?
I think the EV companies and EV manufacturers
for the reason that they control their stack
more holistically, right?
There's a bit of a correlation there
and therefore they can optimize better
their memory requirements and memory footprint, right?
Like you can share memory across
when you have a more centralized architecture
because I mean, certainly use more in ADAS and infotainment
and those do require separate memory
for reliability and safety purposes
but there are other places in the vehicle
that could also consume memory
and having a centralized architecture
would certainly help with that.
Also controlling the software is important
because there's efficiency in how you use memory
and the more the software is decentralized
across many different systems,
the more there will be code that is running
that is not as efficiently structured.
So that is one reason.
The second reason is EV companies tend to have
stronger relationships with some of these
memory suppliers as well.
Just for historical purposes
and the way they've operated more directly with suppliers.
And in some cases like there are other components as well
that go from some of these major conglomerates, right?
Like Samsung is a good example.
They also sell supplier.
They also supply displays, ADAS components and so on.
So not to say that that might be worth a lot
but there is some value there from group level.
Kushal Fernandez, a partner at Carney.
Thank you so much for joining us.
Thank you, Molly.
Kushal Fernandez spoke with our own Molly Boygon.
You can hear the first part of that conversation
on Wednesday's episode of Daily Drive if you missed it.
Thanks for joining me for this bonus episode
of the show.
We'll be back on Monday
with a brand new full episode of Daily Drive.
About this episode
AI-driven demand for computer chips—especially DRAM—is reshaping the auto supply chain and changing who pays, who adapts, and who gets squeezed. DRAM prices have surged as AI infrastructure buyers outbid other sectors, leaving automakers facing “hundreds of millions of dollars in losses.” The discussion weighs automakers’ relative position versus EV makers’ “structural advantage” in securing supply, and explores how memory costs can drive decontenting of infotainment and ADAS, with tier ones absorbing much of the pressure.
Kearney partner Kushal Fernandes breaks down which industries are best — and worst — positioned to survive the DRAM memory chip shortage, why the auto industry lands in a surprising middle ground and what structural advantages electric vehicle makers have over traditional automakers when it comes to securing supply.
