CHIPS
Regular
Thanks for the link. When I tried to listen to the podcast yesterday their Twitter link led me only to podcast No, 31. There was no No. 32 on their website. Again such a stupid mistake.
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Thanks for the link. When I tried to listen to the podcast yesterday their Twitter link led me only to podcast No, 31. There was no No. 32 on their website. Again such a stupid mistake.
cosors
👀
Discover graph data science goes crazy for me with the dots.
The page seems broken somehow.
Dot joining will be quite difficult.
Getupthere
Regular
HopalongPetrovski
I'm Spartacus!
Tothemoon24
Top 20
Stable Genius
Regular
Frangipani
Regular
Hi @cosors,
I know you have completely written off both ZF and Bosch, but personally I wouldn’t discount either one. Nor Continental. (Fun fact: Conti’s headquarters of the Division for Driver Assistance Systems (ADC Automotive Distance Control Systems GmbH) are in Lindau, less than 30 km from the ZF headquarters.)
In 2019, ZF had acquired a 35% stake in ASAP, but sold it last year, together with other shareholders, to a subsidiary of Indian company HCL Technologies, Inc. that now owns 100%.
StradVision have an office in Friedrichshafen as well, so the number plate could be indicative of either them or ZF, but since I couldn’t find any immediate link between StradVision and ASAP, the (leftover?) ASAP gadget made me think this could still be a ZF-owned (or at least pre-loved one, leased or sold to StradVision) vehicle. But you are right, it would be weird of ZF to use their competitor’s radar… Mind you, they may want to try out their rival’s product!
(Although it seems to be a decade-old model, so not exactly cutting-edge).
StradVision’s second German office is in Meerbusch, so that’s maybe why you spotted the car somewhere in Cologne (I assume). StradVision’s guy there used to work for Conti in Korea, so could that possibly be an explanation for the StradVision car being equipped with a radar by a rival of one of the company’s strategic investors?
That’s only partially correct, though. While it is true that both ZF and Bosch abandoned the development of autonomous vehicle LiDAR sensors, they are of course still very much in the ADAS business and continue to develop other sensors for driver assistance systems, such as radar, ultrasound, camera…
Last year, ZF won the AutomotiveINNOVATIONS 2023 award for their 4D Imaging Radar. Earlier this year, they revealed they had developed a smart seat belt and recently rebranded their Passive Safety Systems Division to ZF LIFETEC. This carve-out sounds promising as to potential future implementation of our tech, when we recall the OnSemi collaboration on smart airbags, revealed during CES 2024.
https://press.zf.com/press/de/releases/release_58434.html (German only)
press.zf.com
As for Bosch giving up their LiDAR R&D, my (admittedly amateurish) guess is they finally had to concede that their major competitors (Valeo & Luminar come to mind) were way too far ahead in that race, so they eventually decided to drop out and instead focus on areas where they’d stand a chance of leading the pack.
Not reading past the 2023 headlines about Bosch abandoning LiDAR R&D and turning to mmWave sensors instead, could give readers the impression that over time, the Bosch engineers had come to the conclusion that LiDAR is not as indispensable for the future of automated and autonomous driving as is widely believed in the industry. (With one prominent eXception, that is.)
However, once you read on, you will find that a company spokeswoman told the press at the time that Bosch still viewed LiDAR technology as important for automated driving from Level 3 upwards and would continue seeking to integrate LiDAR products into automated driving functions.
www.sueddeutsche.de
That statement signals to me that abandoning LiDAR R&D is indeed testament to admitting defeat rather than dismissing the technology as redundant for automated and autonomous vehicle applications like Elon Musk has done, ploughing a lone furrow.
Depending on the circumstances, it may actually be wise for a company to resist succumbing to the sunk cost fallacy, but instead pull the ripcord, reposition and restructure, although a lot of money has already gone down the drain (not to mention all the blood, sweat and tears!). Yet, undoubtedly immensely frustrating for everyone involved. And add to that losing face in the public eye and having to endure their rivals’ Schadenfreude (gloating). Well, not so much in this case, as none of the three German Tier 1 automotive suppliers ended up successfully developing competitive LiDAR sensors: In late 2023, Continental - the last one standing, so to say - informed their partner AEye that they intended to discontinue their joint LiDAR development program due to internal restructuring of their business model.
But why should abandoning LiDAR R&D preclude a company from being or becoming a BrainChip customer? There are so many other types of sensors around that BrainChip can potentially make smart, in the automotive sector and elsewhere…
By the way: Just a gentle reminder that another division of Bosch, Bosch Sensortec, partnered with MegaChips in 2015. So if they got or were to get access to our tech via MegaChips, we may never actually get to see a direct IP license from them. As mentioned so often, we’ll have to resort to watching the financials.
I know you have completely written off both ZF and Bosch, but personally I wouldn’t discount either one. Nor Continental. (Fun fact: Conti’s headquarters of the Division for Driver Assistance Systems (ADC Automotive Distance Control Systems GmbH) are in Lindau, less than 30 km from the ZF headquarters.)
In 2019, ZF had acquired a 35% stake in ASAP, but sold it last year, together with other shareholders, to a subsidiary of Indian company HCL Technologies, Inc. that now owns 100%.
Gütt Olk Feldhaus advises ZF on the sale of its ASAP Holding stake - FYB Financial Yearbook
Munich — ZF Friedrichshafen AG has sold its stake in ASAP Holding GmbH together with other shareholders to a subsidiary of the Indian company HCL Technologies, Inc. sold. ZF was advised on this transaction by Gütt Olk Feldhaus. ZF Friedrichshafen AG acquired a 35% stake in ASAP Holding GmbH in...
www.fyb.de
StradVision have an office in Friedrichshafen as well, so the number plate could be indicative of either them or ZF, but since I couldn’t find any immediate link between StradVision and ASAP, the (leftover?) ASAP gadget made me think this could still be a ZF-owned (or at least pre-loved one, leased or sold to StradVision) vehicle. But you are right, it would be weird of ZF to use their competitor’s radar… Mind you, they may want to try out their rival’s product!
(Although it seems to be a decade-old model, so not exactly cutting-edge).StradVision’s second German office is in Meerbusch, so that’s maybe why you spotted the car somewhere in Cologne (I assume). StradVision’s guy there used to work for Conti in Korea, so could that possibly be an explanation for the StradVision car being equipped with a radar by a rival of one of the company’s strategic investors?
That’s only partially correct, though. While it is true that both ZF and Bosch abandoned the development of autonomous vehicle LiDAR sensors, they are of course still very much in the ADAS business and continue to develop other sensors for driver assistance systems, such as radar, ultrasound, camera…
Last year, ZF won the AutomotiveINNOVATIONS 2023 award for their 4D Imaging Radar. Earlier this year, they revealed they had developed a smart seat belt and recently rebranded their Passive Safety Systems Division to ZF LIFETEC. This carve-out sounds promising as to potential future implementation of our tech, when we recall the OnSemi collaboration on smart airbags, revealed during CES 2024.
https://press.zf.com/press/de/releases/release_58434.html (German only)
ZF rebrands its Passive Safety Systems division to ZF LIFETEC
Rebranding is a key milestone in ongoing carve-out process Sales reached around €4.7 billion in 2023 Business supported by automotive industry trends and increasing safety regulations ZF Group exploring further options to develop ZF LIFETEC Friedrichshafen, Germany. ZF Group has announced...
press.zf.com
As for Bosch giving up their LiDAR R&D, my (admittedly amateurish) guess is they finally had to concede that their major competitors (Valeo & Luminar come to mind) were way too far ahead in that race, so they eventually decided to drop out and instead focus on areas where they’d stand a chance of leading the pack.
Bosch exits LiDAR R&D and turns to mmWave sensors
Bosch, the global leader among Tier 1 automotive suppliers, recently announced its exit from the development of high-end autonomous vehicle LiDAR sensors and that it is reallocating its resources toward mmWave radar and other sensing technologies.
www.digitimes.com
Not reading past the 2023 headlines about Bosch abandoning LiDAR R&D and turning to mmWave sensors instead, could give readers the impression that over time, the Bosch engineers had come to the conclusion that LiDAR is not as indispensable for the future of automated and autonomous driving as is widely believed in the industry. (With one prominent eXception, that is.)
However, once you read on, you will find that a company spokeswoman told the press at the time that Bosch still viewed LiDAR technology as important for automated driving from Level 3 upwards and would continue seeking to integrate LiDAR products into automated driving functions.
Bosch steigt aus Entwicklung von Lidar-Sensoren aus
www.sueddeutsche.de
That statement signals to me that abandoning LiDAR R&D is indeed testament to admitting defeat rather than dismissing the technology as redundant for automated and autonomous vehicle applications like Elon Musk has done, ploughing a lone furrow.
Depending on the circumstances, it may actually be wise for a company to resist succumbing to the sunk cost fallacy, but instead pull the ripcord, reposition and restructure, although a lot of money has already gone down the drain (not to mention all the blood, sweat and tears!). Yet, undoubtedly immensely frustrating for everyone involved. And add to that losing face in the public eye and having to endure their rivals’ Schadenfreude (gloating). Well, not so much in this case, as none of the three German Tier 1 automotive suppliers ended up successfully developing competitive LiDAR sensors: In late 2023, Continental - the last one standing, so to say - informed their partner AEye that they intended to discontinue their joint LiDAR development program due to internal restructuring of their business model.
But why should abandoning LiDAR R&D preclude a company from being or becoming a BrainChip customer? There are so many other types of sensors around that BrainChip can potentially make smart, in the automotive sector and elsewhere…
By the way: Just a gentle reminder that another division of Bosch, Bosch Sensortec, partnered with MegaChips in 2015. So if they got or were to get access to our tech via MegaChips, we may never actually get to see a direct IP license from them. As mentioned so often, we’ll have to resort to watching the financials.
Bravo
If ARM was an arm, BRN would be its biceps💪!
Pretty good if we pull a Hexagon-shaped Qualcomm rabbit out of our hat.How does everyone think the 5 year plan is going?
Sébastien Crouzet on LinkedIn: Together with Rudy Pei, Sasskia Brüers Freyssinet, Douglas McLelland and…
Together with Rudy Pei, Sasskia Brüers Freyssinet, Douglas McLelland and Olivier Coenen from BrainChip , we participated and reached🥉 in the "Event-Based Eye…
cosors
👀
WoW, that surprises and delights me at the same time!Hi @cosors,
I know you have completely written off both ZF and Bosch, but personally I wouldn’t discount either one. Nor Continental. (Fun fact: Conti’s headquarters of the Division for Driver Assistance Systems (ADC Automotive Distance Control Systems GmbH) are in Lindau, less than 30 km from the ZF headquarters.)
In 2019, ZF had acquired a 35% stake in ASAP, but sold it last year, together with other shareholders, to a subsidiary of Indian company HCL Technologies, Inc. that now owns 100%.
Gütt Olk Feldhaus advises ZF on the sale of its ASAP Holding stake - FYB Financial Yearbook
Munich — ZF Friedrichshafen AG has sold its stake in ASAP Holding GmbH together with other shareholders to a subsidiary of the Indian company HCL Technologies, Inc. sold. ZF was advised on this transaction by Gütt Olk Feldhaus. ZF Friedrichshafen AG acquired a 35% stake in ASAP Holding GmbH in...www.fyb.de
StradVision have an office in Friedrichshafen as well, so the number plate could be indicative of either them or ZF, but since I couldn’t find any immediate link between StradVision and ASAP, the (leftover?) ASAP gadget made me think this could still be a ZF-owned (or at least pre-loved one, leased or sold to StradVision) vehicle. But you are right, it would be weird of ZF to use their competitor’s radar… Mind you, they may want to try out their rival’s product!
StradVision’s second German office is in Meerbusch, so that’s maybe why you spotted the car somewhere in Cologne (I assume). StradVision’s guy there used to work for Conti in Korea, so could that possibly be an explanation for the StradVision car being equipped with a radar by a rival of one of the company’s strategic investors?
View attachment 60998
That’s only partially correct, though. While it is true that both ZF and Bosch abandoned the development of autonomous vehicle LiDAR sensors, they are of course still very much in the ADAS business and continue to develop other sensors for driver assistance systems, such as radar, ultrasound, camera…
Last year, ZF won the AutomotiveINNOVATIONS 2023 award for their 4D Imaging Radar. Earlier this year, they revealed they had developed a smart seat belt and recently rebranded their Passive Safety Systems Division to ZF LIFETEC. This carve-out sounds promising as to potential future implementation of our tech, when we recall the OnSemi collaboration on smart airbags, revealed during CES 2024.
https://press.zf.com/press/de/releases/release_58434.html (German only)
ZF rebrands its Passive Safety Systems division to ZF LIFETEC
Rebranding is a key milestone in ongoing carve-out process Sales reached around €4.7 billion in 2023 Business supported by automotive industry trends and increasing safety regulations ZF Group exploring further options to develop ZF LIFETEC Friedrichshafen, Germany. ZF Group has announced...
View attachment 60994
View attachment 60993
As for Bosch giving up their LiDAR R&D, my (admittedly amateurish) guess is they finally had to concede that their major competitors (Valeo & Luminar come to mind) were way too far ahead in that race, so they eventually decided to drop out and instead focus on areas where they’d stand a chance of leading the pack.
Bosch exits LiDAR R&D and turns to mmWave sensors
Bosch, the global leader among Tier 1 automotive suppliers, recently announced its exit from the development of high-end autonomous vehicle LiDAR sensors and that it is reallocating its resources toward mmWave radar and other sensing technologies.www.digitimes.com
Not reading past the 2023 headlines about Bosch abandoning LiDAR R&D and turning to mmWave sensors instead, could give readers the impression that over time, the Bosch engineers had come to the conclusion that LiDAR is not as indispensable for the future of automated and autonomous driving as is widely believed in the industry. (With one prominent eXception, that is.)
However, once you read on, you will find that a company spokeswoman told the press at the time that Bosch still viewed LiDAR technology as important for automated driving from Level 3 upwards and would continue seeking to integrate LiDAR products into automated driving functions.
Bosch steigt aus Entwicklung von Lidar-Sensoren aus
View attachment 60996
That statement signals to me that abandoning LiDAR R&D is indeed testament to admitting defeat rather than dismissing the technology as redundant for automated and autonomous vehicle applications like Elon Musk has done, ploughing a lone furrow.
Depending on the circumstances, it may actually be wise for a company to resist succumbing to the sunk cost fallacy, but instead pull the ripcord, reposition and restructure, although a lot of money has already gone down the drain (not to mention all the blood, sweat and tears!). Yet, undoubtedly immensely frustrating for everyone involved. And add to that losing face in the public eye and having to endure their rivals’ Schadenfreude (gloating). Well, not so much in this case, as none of the three German Tier 1 automotive suppliers ended up successfully developing competitive LiDAR sensors: In late 2023, Continental - the last one standing, so to say - informed their partner AEye that they intended to discontinue their joint LiDAR development program due to internal restructuring of their business model.
But why should abandoning LiDAR R&D preclude a company from being or becoming a BrainChip customer? There are so many other types of sensors around that BrainChip can potentially make smart, in the automotive sector and elsewhere…
By the way: Just a gentle reminder that another division of Bosch, Bosch Sensortec, partnered with MegaChips in 2015. So if they got or were to get access to our tech via MegaChips, we may never actually get to see a direct IP license from them. As mentioned so often, we’ll have to resort to watching the financials.
Thank you for the information.
You're right, Meerbusch is close to Cologne.
I was so surprised to see the car just standing there abandoned that I couldn't help but quickly take a few photos.
I've often seen vehicles like this here in the region. But they were always quickly out of sight before I could take any photos.
The only thing that strikes me as odd is the large/big structure on the roof, which seems to have fallen out of time.
Thanks again I will study your post in detail with more time.
I would love it if ZF wasn't out of the running.
You're right, of course. That doesn't mean that they can't be a potential customer. There are now also more details on the subject in the media than when I first heard about it.
They seem to have also dropped their autonomous buses project?
But here comes a nice treat. But it's not ready to present yet.
A friend of mine has recently become project manager for just such vehicles. I'll have to be patient until I can pester him with questions, but if he tells me what he's allowed to say, I'll pass the information on to you.
Last edited:
Frangipani
Regular
I suppose you mean their autonomous shuttles? Yes and no.
Here is the Dec 2023 press release:
ZF strategically adjusts its shuttle business
Focus on position as a premium supplier for autonomous driving technology Engineering services to provide system knowledge at a higher-level AD as a new offering for customers Adjusted business model builds on ZF’s core strengths and stands for promising growth prospects and higher...
press.zf.com
2023-Dec-19
ZF strategically adjusts its shuttle business
- Focus on position as a premium supplier for autonomous driving technology
- Engineering services to provide system knowledge at a higher-level AD as a new offering for customers
- Adjusted business model builds on ZF’s core strengths and stands for promising growth prospects and higher profitability
Friedrichshafen, Germany. Technology company ZF changes its business model for the global shuttle business and will focus on being a leading technology supplier for autonomous driving. The company has decided to no longer pursue the goal of facilitating entire autonomous transportation systems, including shuttles and their fleet management. Instead, the strategic focus and allocation of funds will shift to providing engineering services to customers and further developing the building blocks needed to advance autonomous driving. Existing partnerships and projects relating to the GRT shuttle will be continued as well as ZF Aftermarket activities.
With the market developing slower than anticipated, and at a time where the persisting multi-crisis and the ongoing transformation towards E-Mobility requires stringent cost focus in the entire industry, the foreseeable high up-front investments are no longer justified.
Having evaluated all options, ZF has concluded that the most promising strategy for the future is to focus on positioning ZF as a premium supplier of autonomous driving technology and engineering services.
ZF is a premium supplier for AD technology
Therefore, ZF will maintain the original position as a supplier to OEMs while continuing to develop the technologies needed to build autonomous transportation systems and advanced passenger cars with assisted and self-driving capabilities. In addition, ZF will market the Group’s unique expertise in systems solutions by offering valuable and highly sought-after development services to customers in the automotive and mobility industry.
In the future ZF enables both OEMs and providers of new mobility systems with its strong portfolio of components and software solutions and offer its capable development services as a booster for customers’ R&D activities. This redirection away from facilitating entire shuttles will also allow ZF to better focus on its core technologies and serve its OEM customers in the passenger car, commercial vehicle, and industrial space. ZF will continue to be a strong enabler of ‘Next Generation Mobility’ by providing OEMs with a unique offering.
Today, automated and autonomous functions have found their way into all vehicle categories, where they make a significant contribution to greater safety, efficiency, and comfort. The technical basis for these functions is provided by ZF, which has developed a strong portfolio of advanced sensors, high-performance computers, special software solutions, and intelligent actuators over the past years.
Continuation of existing partnerships and projects
Existing partnerships and projects with the GRT shuttle for operation in separated lanes will continue as before. This shuttle model from ZF subsidiary 2getthere is already being used successfully in projects such as Rivium in the Netherlands and will continue to be produced and sold by US license partner Oceaneering in the future. Projects such as RABus for researching automated bus operation in the two German cities of Mannheim and Friedrichshafen is important for the new business model and is also indented to be implemented as planned.
ZF will also continue its aftermarket activities for autonomous vehicle fleets and offers services like maintenance, repair and training. For service, customers can rely on ZF’s global network with 20,000 workshop partners worldwide.
cosors
👀
Thanks again!I suppose you mean their autonomous shuttles? Yes and no.
Here is the Dec 2023 press release:
ZF strategically adjusts its shuttle business
Focus on position as a premium supplier for autonomous driving technology Engineering services to provide system knowledge at a higher-level AD as a new offering for customers Adjusted business model builds on ZF’s core strengths and stands for promising growth prospects and higher...
2023-Dec-19
ZF strategically adjusts its shuttle business
- Focus on position as a premium supplier for autonomous driving technology
- Engineering services to provide system knowledge at a higher-level AD as a new offering for customers
- Adjusted business model builds on ZF’s core strengths and stands for promising growth prospects and higher profitability
Friedrichshafen, Germany. Technology company ZF changes its business model for the global shuttle business and will focus on being a leading technology supplier for autonomous driving. The company has decided to no longer pursue the goal of facilitating entire autonomous transportation systems, including shuttles and their fleet management. Instead, the strategic focus and allocation of funds will shift to providing engineering services to customers and further developing the building blocks needed to advance autonomous driving. Existing partnerships and projects relating to the GRT shuttle will be continued as well as ZF Aftermarket activities.
With the market developing slower than anticipated, and at a time where the persisting multi-crisis and the ongoing transformation towards E-Mobility requires stringent cost focus in the entire industry, the foreseeable high up-front investments are no longer justified.
Having evaluated all options, ZF has concluded that the most promising strategy for the future is to focus on positioning ZF as a premium supplier of autonomous driving technology and engineering services.
ZF is a premium supplier for AD technology
Therefore, ZF will maintain the original position as a supplier to OEMs while continuing to develop the technologies needed to build autonomous transportation systems and advanced passenger cars with assisted and self-driving capabilities. In addition, ZF will market the Group’s unique expertise in systems solutions by offering valuable and highly sought-after development services to customers in the automotive and mobility industry.
In the future ZF enables both OEMs and providers of new mobility systems with its strong portfolio of components and software solutions and offer its capable development services as a booster for customers’ R&D activities. This redirection away from facilitating entire shuttles will also allow ZF to better focus on its core technologies and serve its OEM customers in the passenger car, commercial vehicle, and industrial space. ZF will continue to be a strong enabler of ‘Next Generation Mobility’ by providing OEMs with a unique offering.
Today, automated and autonomous functions have found their way into all vehicle categories, where they make a significant contribution to greater safety, efficiency, and comfort. The technical basis for these functions is provided by ZF, which has developed a strong portfolio of advanced sensors, high-performance computers, special software solutions, and intelligent actuators over the past years.
Continuation of existing partnerships and projects
Existing partnerships and projects with the GRT shuttle for operation in separated lanes will continue as before. This shuttle model from ZF subsidiary 2getthere is already being used successfully in projects such as Rivium in the Netherlands and will continue to be produced and sold by US license partner Oceaneering in the future. Projects such as RABus for researching automated bus operation in the two German cities of Mannheim and Friedrichshafen is important for the new business model and is also indented to be implemented as planned.
ZF will also continue its aftermarket activities for autonomous vehicle fleets and offers services like maintenance, repair and training. For service, customers can rely on ZF’s global network with 20,000 workshop partners worldwide.
And of course I meant that Meerbusch is near to Cologne and not part of it. It's (close to) Düsseldorf.
But I remain sceptical about all three when it comes to in-house developments and not to only assemble components. But I will read your links carefully.
All but MB seem to be finding it extremely difficult to restructure in the automotive sector here in Germany. And even MB has its problems.
The delegation from MB and BMW in China the last few days had asked the German Chancellor not to raise trade barriers against China. Intended to protect the European market.
I am afraid that the gap of dependency cannot be bridged with in-house development.
We here with Brainchip usually only think of chips and ADAS or assistance systems or to make an EV more comfortable when we think of cars. An EV can also drive without this or drive with existing technology. But not without batteries. And that's where German OEMs are almost empty-handed, totally out of touch, especially with raw materials. That's an important topic for Bosch too.
If China turns off the tap, no EV will roll off the production line, whether with ADAS, LiDAR, assistance systems or without.
That is probably why they are afraid of trade barriers.
Last edited:
Frangipani
Regular
Why would the German trinity of automotive suppliers each have established their own AI Labs if they only assembled components?
They spend a lot of moolah on R&D in general and also offer all sorts of engineering services:
Hardware Engineering
We develop customized hardware solutions for a variety of industries. Our portfolio ranges from head-up displays up to industrial controls in various sectors.
conti-engineering.com
Research & Development - Looking at the Big Picture - ZF
With digital innovations and continuously high investments in research and development, we are among the leading technology companies.
www.zf.com
Frangipani
Regular
Thanks for the link. When I tried to listen to the podcast yesterday their Twitter link led me only to podcast No, 31. There was no No. 32 on their website. Again such a stupid mistake.
Space Robot Ken made sure the social media team posted the correct link today!
Frangipani
Regular
Intel Builds World’s Largest Neuromorphic System to Enable More...
Hala Point, the industry’s first 1.15 billion neuron neuromorphic system, builds a path toward more efficient and scalable AI.
Link to the conference paper Mike Davies refers to:
It looks like Akida is not mentioned alongside other neuromorphic hardware platforms…
Last edited:
"AI has the potential to exceed all the transformative innovations created in the last century "
"Every watt saved enables more compute. This is why Amazon, Microsoft, Google, and Oracle have now all adopted Arm Neoverse technology to solve both general-purpose compute and CPU-based AI inference and training. Arm Neoverse is on the path to being the de-facto standard across cloud data centers."
newsroom.arm.com
"Every watt saved enables more compute. This is why Amazon, Microsoft, Google, and Oracle have now all adopted Arm Neoverse technology to solve both general-purpose compute and CPU-based AI inference and training. Arm Neoverse is on the path to being the de-facto standard across cloud data centers."
Arm’s Mission to Help Tackle AI’s Insatiable Energy Needs
Learn how Arm is driving innovation and efficiency in AI datacenter hardware and helping to shape a more sustainable future for computing.
newsroom.arm.com
Last edited:
jtardif999
Regular
AI could gobble up a quarter of all electricity in the U.S. by 2030 if it doesn’t break its energy addiction, says Arm Holdings exec
Before artificial intelligence can transform society, the technology will first have to learn how to live within its means.
Right now generative AI has an “insatiable demand” for electricity to power the tens of thousands of compute clusters needed to operate large language models like OpenAI’s GPT-4, warned chief marketing officer Ami Badani of chip design firm Arm Holdings.
If generative AI is ever going to be able to run on every mobile device from a laptop and tablet to a smartphone, it will have to be able to scale without overwhelming the electricity grid at the same time.
“We won’t be able to continue the advancements of AI without addressing power,” Badani told the Fortune Brainstorm AI conference in London on Monday. “ChatGPT requires 15 times more energy than a traditional web search.”
Not only are more businesses using generative AI, but the tech industry is in a race to develop new and more powerful tools that will mean compute demand is only going to grow—and power consumption with it, unless something can be done.
The latest breakthrough from OpenAI, the company behind ChatGPT, is Sora. It can create super realistic or stylized clips of video footage up to 60 seconds in length purely based on user text prompts.
The marvel of gen AI comes at a steep cost
“It takes 100,000 AI chips working at full compute capacity and full power consumption in order to train Sora,” Badani said. “That’s a huge amount.”
Data centers, where most AI models are trained, currently account for 2% of global electricity consumption, according to Badani. But with generative AI expected to go mainstream, she predicts it could end up devouring a quarter of all power in the United States in 2030.
The solution to this conundrum is to develop semiconductor chips that are optimized to run on a minimum of energy.
That’s where Arm comes in: Its RISC processor designs currently run on 99% of all smartphones, as opposed to the rival x86 architecture developed by Intel. The latter has been a standard for desktop PCs, but proved too inefficient to run battery-powered handheld devices like smartphones and tablets.
Arm is adopting that same design philosophy for AI.
“If you think about AI, it comes with a cost,” Badani said, “and that cost is unfortunately power.”
Before artificial intelligence can transform society, the technology will first have to learn how to live within its means.
Right now generative AI has an “insatiable demand” for electricity to power the tens of thousands of compute clusters needed to operate large language models like OpenAI’s GPT-4, warned chief marketing officer Ami Badani of chip design firm Arm Holdings.
If generative AI is ever going to be able to run on every mobile device from a laptop and tablet to a smartphone, it will have to be able to scale without overwhelming the electricity grid at the same time.
“We won’t be able to continue the advancements of AI without addressing power,” Badani told the Fortune Brainstorm AI conference in London on Monday. “ChatGPT requires 15 times more energy than a traditional web search.”
Not only are more businesses using generative AI, but the tech industry is in a race to develop new and more powerful tools that will mean compute demand is only going to grow—and power consumption with it, unless something can be done.
The latest breakthrough from OpenAI, the company behind ChatGPT, is Sora. It can create super realistic or stylized clips of video footage up to 60 seconds in length purely based on user text prompts.
The marvel of gen AI comes at a steep cost
“It takes 100,000 AI chips working at full compute capacity and full power consumption in order to train Sora,” Badani said. “That’s a huge amount.”
Data centers, where most AI models are trained, currently account for 2% of global electricity consumption, according to Badani. But with generative AI expected to go mainstream, she predicts it could end up devouring a quarter of all power in the United States in 2030.
The solution to this conundrum is to develop semiconductor chips that are optimized to run on a minimum of energy.
That’s where Arm comes in: Its RISC processor designs currently run on 99% of all smartphones, as opposed to the rival x86 architecture developed by Intel. The latter has been a standard for desktop PCs, but proved too inefficient to run battery-powered handheld devices like smartphones and tablets.
Arm is adopting that same design philosophy for AI.
“If you think about AI, it comes with a cost,” Badani said, “and that cost is unfortunately power.”
Our CTO thinks this is great
www.linkedin.com
Marc Theermann on LinkedIn: Meet our next generation Atlas! A humanoid designed for commercial… | 222 comments
Meet our next generation Atlas! A humanoid designed for commercial applications. We are taking all we have learned over 15 years of research on our hydraulic… | 222 comments on LinkedIn
Generative AI and the Impact on Automotive Industry
Generative AI and the Impact on Automotive Industry
Similar threads
