RJ's commitments on Gen 2 R1 Autonomous Driving: L2 this year (2026), L3 next year (2027)

[TexasBob]

RJ was on Bloomberg talking about the R2 Launch and he made some very precise and very explicit claims about what and when the Gen 2 autonomy pack will achieve:

Ed Ludlow's question:

"I've experienced the R2 that does have the LiDAR and the in house silicon. And it's so interesting. There's gonna be a generation of R2s that don't have it. I know that you said they're very capable with the NVIDIA silicon and without LiDAR, but that's a big call to make. Right? Is there any flexibility to accelerate the timing of where you can start production on the R2 line in Illinois and bring in that gear? Because what you're pitching for the autonomous capabilities, even the advanced driver assistance capabilities of R2, they are predicated on that next gen gear."

RJ's response — setting the stage with Gen 2 R1 as the baseline:

"Well, the launch edition still has a very high ceiling, and so this will get to — and our Gen 2 R1 will also do this. Later this year, we'll start to roll out point to point level two. So meaning, hands off wheel, eyes on road, but you type the address in and the car completely drives you there. And then into next year, we'll start to roll out level three. So hands off, eyes off for specific domains. In this case, you know, starting with highways. And that will be true for Gen 2 of R1. It'll be true for the launch edition of R2, and it'll, of course, be true for the variant of the vehicle that has the higher compute stack and more capable perception."

Bloomberg Interview:



I am skeptical that our little single Orin chipset and no LiDAR will be able to achieve this, but it is very good to get a specific commitment on timing and capability level. To be clear, that means Tesla FSD (Supervised) this year and true L3 highway next year. That is what we are hoping and waiting for, but as I said, I am skeptical.

Sponsored

 

[Great Gatsby]

Heard that as well. Appreciate his commitment and transparency. Huge hurdle though considering how inconsistent UHF is proving to be in its current state.

My current plan is getting LE R2 and enjoy it with UHF then likely trade it in when the Tri becomes available at which point it should have lidar and the benefits of the L3 (hopefully). Based on the R1, I foresee the R2 holding its value rather well so hopefully I don't take too much of a bath in the next year.

All to say, I don't think this will affect sales. Most people truly have not experienced self-driving, so this will not be a selling point to most and even if Rivian can get just point to point, I think this will be good enough for most.

 

[ndmiller]

I love that he stated it, but L3 isn't just a technology capability issue, it's primarily a nationwide/state/insurance company legal issue. There is zero chance these eyes off L3 issues will be resolved with personal vehicles next year.

Not saying the car won't be able to do it, just saying it won't be doing it regardless for personally owned regular people with insurance in one of the 50 states.

 

[portdirect]

RJ was on Bloomberg talking about the R2 Launch and he made some very precise and very explicit claims about what and when the Gen 2 autonomy pack will achieve:

Ed Ludlow's question:

RJ's response — setting the stage with Gen 2 R1 as the baseline:

I am skeptical that our little single Orin chipset and no LiDAR will be able to achieve this, but it is very good to get a specific commitment on timing and capability level. To be clear, that means Tesla FSD (Supervised) this year and true L3 highway next year. That is what we are hoping and waiting for, but as I said, I am skeptical.

I’ll nit pick you that it’s a dual Orin chipset platform in gen2 - but Rivian have been so weird about the specs, and afaik no one has actully torn down an AXM2 to actually look at the board.

Agree on all other points though. I stumped for the lifetime autonomy+ as it was a way to offload gearshop points - but only expect basic point to point like capabilities (ideally stop at stop signs and traffic lights for UHF, but potentially even just very basic highway junction handlin) before we are left behind.

The additional headroom of RAP and cost of maintaining two stacks, as well as uber wanting to see some ROI makes me think by the end of q1 next year the brain trust will have moved on.

 

[godfodder0901]

I am skeptical that our little single Orin chipset and no LiDAR will be able to achieve this, but it is very good to get a specific commitment on timing and capability level. To be clear, that means Tesla FSD (Supervised) this year and true L3 highway next year. That is what we are hoping and waiting for, but as I said, I am skeptical.

I mean, if Comma can do it with an 8yr-old cellphone, why couldn't Rivian do it with 5x the compute and integrated RADAR?

 

[DuoRivian]

RJ was on Bloomberg talking about the R2 Launch and he made some very precise and very explicit claims about what and when the Gen 2 autonomy pack will achieve:

Ed Ludlow's question:

RJ's response — setting the stage with Gen 2 R1 as the baseline:

I am skeptical that our little single Orin chipset and no LiDAR will be able to achieve this, but it is very good to get a specific commitment on timing and capability level. To be clear, that means Tesla FSD (Supervised) this year and true L3 highway next year. That is what we are hoping and waiting for, but as I said, I am skeptical.

Very clear from what he has said in this interview and other recent ones is that the 2026 built R2s will be capable (as is the current R1) of L3 hands free and eyes off the road autonomy. Therefore people don’t need to wait until RAP1 vehicles next year for this.

 

[Hereforthesnacks]

Very clear from what he has said in this interview and other recent ones is that the 2026 built R2s will be capable (as is the current R1) of L3 hands free and eyes off the road autonomy. Therefore people don’t need to wait until RAP1 vehicles next year for this.

I really can’t believe Rivian on this one. It’s the same playbook as with the Gen 1 R1. Promises it will do x, y, z. But it didn’t. And this time they’ve already telegraphed that the have a more capable R2 in the works.

 

[DuoRivian]

I really can’t believe Rivian on this one. It’s the same playbook as with the Gen 1 R1. Promises it will do x, y, z. But it didn’t. And this time they’ve already telegraphed that the have a more capable R2 in the works.

ADAS v2 compute is 10x v1 so very believable that it is much more capable. If you are going to hold out for RAP1 then wait longer for RAP2! Hands free point to point is easily within capability of the current system.

 

[mkhuffman]

I mean, if Comma can do it with an 8yr-old cellphone, why couldn't Rivian do it with 5x the compute and integrated RADAR?

I agree with Bob, I am definitely skeptical especially regarding "eyes off".

It took years for Tesla to get to decent point-to-point, and Rivian can do it in one or two? Really? I hope so, because I want it, but still. Skeptical.

I am ready for them to prove me wrong.

 

[Hereforthesnacks]

ADAS v2 compute is 10x v1 so very believable that it is much more capable. If you are going to hold out for RAP1 then wait longer for RAP2! Hands free point to point is easily within capability of the current system.

I’ve been burned by them, though! Lol

 

[UnsungZero_OldTimeAdMan]

Aside from the twister and how they're working around the damage... nothing new. Also already said before, gen3 and LiDAR's primary objective is to train their Large Driving Model and catch up to Tesla, which all gen2 Rivians will benefit from (including hands-off and eyes-off, point-to-point). Gen3 cars will just have higher development envelope and [probably] less restrictive parameters—because they can "see" more and further in adverse weather. LiDAR is not a must-have some make it to be.

 

[JoulesVerne]

My interpretation of RJ speak is that non-LIDAR R2s and Gen 2 R1 will have hands-off/eyes-off for highways only (which will probably also require mapping - so not all highways).

Therefore LIDAR equipped R2s will get hands-off/eyes off (aka Level 3 autonomy) everywhere.

 

[Riviot]

RJ's response — setting the stage with Gen 2 R1 as the baseline:
"Well, the launch edition still has a very high ceiling, and so this will get to — and our Gen 2 R1 will also do this. Later this year, we'll start to roll out point to point level two. So meaning, hands off wheel, eyes on road, but you type the address in and the car completely drives you there. And then into next year, we'll start to roll out level three. So hands off, eyes off for specific domains. In this case, you know, starting with highways. And that will be true for Gen 2 of R1. It'll be true for the launch edition of R2, and it'll, of course, be true for the variant of the vehicle that has the higher compute stack and more capable perception."

Comma is $1k to upgrade to what Rivian advertised to 2022 buyers. R2 is $58k.

 

[skyguyscott]

If what RJ means by hands-off, eyes-on, point2point for Gen 2 R1 is a Tesla FSD, camera-only style monthly subscription for fatal accidents and debilitating injuries, nah, hard pass. You are going to have to demonstrate real-world reliability in a host of adverse driving conditions in complex environments and/or the vehicle consistently and correctly recognizing it's own limitations and refusing to engage before I bite, and even then, I doubt I 'd pay much for it unless and until I am too old and feeble to drive myself.

 

[DuoRivian]

My interpretation of RJ speak is that non-LIDAR R2s and Gen 2 R1 will have hands-off/eyes-off for highways only (which will probably also require mapping - so not all highways).

Therefore LIDAR equipped R2s will get hands-off/eyes off (aka Level 3 autonomy) everywhere.

Gen 2 is currently universal hands off, so why would they go backwards to mapped roads?

Sponsored