Wednesday, July 10th 2024

AMD to Acquire Silo AI to Expand Enterprise AI Solutions Globally

AMD today announced the signing of a definitive agreement to acquire Silo AI, the largest private AI lab in Europe, in an all-cash transaction valued at approximately $665 million. The agreement represents another significant step in the company's strategy to deliver end-to-end AI solutions based on open standards and in strong partnership with the global AI ecosystem. The Silo AI team consists of world-class AI scientists and engineers with extensive experience developing tailored AI models, platforms and solutions for leading enterprises spanning cloud, embedded and endpoint computing markets.

Silo AI CEO and co-founder Peter Sarlin will continue to lead the Silo AI team as part of the AMD Artificial Intelligence Group, reporting to AMD senior vice president Vamsi Boppana. The acquisition is expected to close in the second half of 2024.
"Across every industry, enterprises are looking for fast and effective ways to develop and deploy AI solutions for their unique business needs," said Vamsi Boppana, senior vice president of the Artificial Intelligence Group at AMD. "Silo AI's team of trusted AI experts and proven experience developing leadership AI models and solutions, including state-of-the-art LLMs built on AMD platforms, will further accelerate our AI strategy and advance the build-out and rapid implementation of AI solutions for our global customers."

Based in Helsinki, Finland, with operations in Europe and North America, Silo AI specializes in end-to-end AI-driven solutions that help customers integrate AI quickly and easily into their products, services and operations. Their work spans diverse markets, with customers including Allianz, Philips, Rolls-Royce and Unilever. Silo AI also creates state-of-the-art open source multilingual LLMs, such as Poro and Viking, on AMD platforms in addition to its SiloGen model platform.

"At Silo AI, our mission from the start has been to build an AI flagship company. Today's announcement is a logical next step in that pursuit as we join forces with AMD to shape the future of AI computing," said Peter Sarlin, CEO and co-founder of Silo AI. "We have a well-established history of building successful AI products and delivering value to our customers. We look forward to becoming part of AMD to further scale our impact and develop enterprise solutions and AI models that address the most complex challenges with deploying AI at scale today."

"As a leading European AI company, Silo AI has been a great long-term partner for us in many AI-related projects. We look forward to the enhanced capabilities the combination of AI technologies and innovative compute solutions from AMD will bring," said Nishant Batra, Chief Strategy and Technology Officer (CSTO), Nokia.

Silo AI marks the latest in a series of acquisitions and corporate investments to support the AMD AI strategy. AMD has invested over $125 million across a dozen AI companies in the last 12 months and also acquired Mipsology and Nod.ai to expand the AMD AI ecosystem, support partners and advance leadership AMD computing platforms.

"Silo AI has been a pioneer in scaling large language model training on LUMI, Europe's fastest supercomputer powered by over 12,000 AMD Instinct MI250X GPUs," said Dr. Pekka Manninen, Director of Science and Technology at CSC-IT Center for Science, Finland. "Together with university collaborators, they have trained state-of-the-art open-source models for EU languages, such as the Nordic Poro and Viking models. We have collaborated extensively with the team in optimizing the software layer, allowing for efficient training of AI models on LUMI."

"One year ago, Combient partnered with Silo AI, a leading AI lab with 300 AI scientists and engineers. Together with Marcus Wallenberg, we initiated Combient in 2015 to accelerate digitalization and AI adoption for 38 of the largest Nordic companies, with a total turnover of €270 billion, such as H&M, IKEA, Saab, KONE and Ericsson," said Mats Agervi, CEO of Combient. "Today's acquisition underscores the capabilities of Silo AI, opening up expanded opportunities for increased value creation in Europe and beyond."
Source: AMD
Add your own comment

22 Comments on AMD to Acquire Silo AI to Expand Enterprise AI Solutions Globally

#1
thesmokingman
Hmm, not sure what Lisa's goal is here. I guess they're going to provide turnkey solutions? Don't know that there is money in this, hell since AI hasn't turned anybody much of a profit. They need to produce moar gpus imo. GPU for AI market has a limited time as the giants in the field transition to dedicated silicon.
Posted on Reply
#2
nguyen
in an all-cash transaction valued at approximately $665 million
Could have spent that money on hiring competent software developers LOL
Posted on Reply
#3
Jomale
thesmokingmanHmm, not sure what Lisa's goal is here.
Let me guess: Automotive/self drive system, Banking/Insurance support, help communication/legal advice autonomy systems?
Posted on Reply
#4
thesmokingman
JomaleLet me guess: Automotive/self drive system, Banking/Insurance support, help communication/legal advice autonomy systems?
It's too late for Autos/ADAS at this point, only losers going to be contracting out for hw. Tesla has already shown the way to this and none of their hw is off the shelf. AMD however is a partner for Tesla's mediacenter chips. But I don't see how a good but snails pace moving company like AMD is going to compete for the dredges from Nvidia at this point.
Posted on Reply
#7
prtskg
nguyenCould have spent that money on hiring competent software developers LOL
Silo's developers are working for AMD now. So they've done exactly that. Hired developers with experience.
Posted on Reply
#8
Sound_Card
nguyenCould have spent that money on hiring competent software developers LOL
prtskgSilo's developers are working for AMD now. So they've done exactly that. Hired developers with experience.
I was going to say the same thing, I thought it was pretty obvious lol
Posted on Reply
#10
nguyen
prtskgSilo's developers are working for AMD now. So they've done exactly that. Hired developers with experience.
What developers and what experience LOL.

AI developer with 1 year experience?
Posted on Reply
#11
thesmokingman
JomaleYou are right, look how teslas "self driving" is learning:
www.businessinsider.com/tesla-prioritizes-musk-vip-data-self-driving-2024-7
Lmao, you got jack so you resort to trolling, using a BI article lmao? Shorts are losing their shirts down over 6B since June.
prtskgSilo's developers are working for AMD now. So they've done exactly that. Hired developers with experience.
Still think AMD is way behind. Horowitz of A16Z is amassing 20K GPUs and leasing them out to startups. AMD still hasn't gotten the memo on this.
Posted on Reply
#12
dragontamer5788
thesmokingmanHmm, not sure what Lisa's goal is here. I guess they're going to provide turnkey solutions? Don't know that there is money in this, hell since AI hasn't turned anybody much of a profit. They need to produce moar gpus imo. GPU for AI market has a limited time as the giants in the field transition to dedicated silicon.
Acquihire.

Its not the "AI" that Lisa is interested in. She's interested in a large team of developers who know how to use AMD's MI250x GPUs to make something.

That "something" happened to be AI over the last few years (possibly as that team raised money), but moving forward the software engineering skills are the focus. The developers are always the focus in these kinds of acqui-hires, as people like to call them.
Posted on Reply
#13
Daven
nguyenCould have spent that money on hiring competent software developers LOL
AMD did exactly that with this acquisition yet still…you critique.
Posted on Reply
#14
sLowEnd
nguyenWhat developers and what experience LOL.

AI developer with 1 year experience?
Their website has some information

www.silo.ai/work
Posted on Reply
#15
wolf
Better Than Native
I am interested to see how this move pans out for AMD. They evidently want the biggest slice of the AI pie, but I fear by the time they can do a whole lot that is meaningful with this acquisition, the bubble may have burst - the sooner the better, who else is sick of hearing/reading about "AI"? It would also take quite a while of kicking goals for them to alter the mindshare, time will tell if it's money well spent but good luck to them, they'll need a bit of that too.
Posted on Reply
#16
stimpy88
AMD has plenty of things it would be better spending this kind of money on. Like competent driver/software devs, more capacity at TSMC, better GPU engineers etc... All of which would benefit AMD more.
Posted on Reply
#17
Vayra86
JomaleNo, in Europe only Bosch and Mercedes (S-Klasse) have a Class 4 autonomy driving permission:
www.automotiveit.eu/technology/autonomes-fahren/welcher-autobauer-hat-beim-autonomen-fahren-die-nase-vorn-124.html

And: www.roedl.de/themen/kompass-mobilitaet/2023/20/tesla-autopilot-kein-autonomes-fahren

Myeah.... its all bullshit because the key issues remain unsolved, mainly insurance/responsibility, the uncanny valley issue, and the general fact people really don't want half the road filled with autonomous and the other with regular vehicles. Its going to be an immense shitshow and it will never happen until we can let everything drive autonomously. I'm smelling a Google Glass / VR / Metaverse (or insert any other hype tech of the last 10 years) failure here. Its never really gonna happen.

Class 4 is perfectly useless. Class 5 is where we should be at, and nobody has it. Of course Germany is trying their bestest to show the world they're still leading in automotive, but we all know they're not. They were good at ICEs. They, so far, SUCK, at EVs compared to the competition abroad. They move too slowly. Even the French make better EVs now (drove a Renault Megane the other day. VW's platform feels immature in comparison and VAG generally just sucks when it comes to overall cost/options packages). There's a strong chance my next lease EV will be a BYD Seal (U).... I kid you not. Or a Tesla Model Y long range. They're just better in every metric including price.
Posted on Reply
#18
dragontamer5788
stimpy88AMD has plenty of things it would be better spending this kind of money on. Like competent driver/software devs, more capacity at TSMC, better GPU engineers etc... All of which would benefit AMD more.
AMD's GPUs are already rivaling those of NVidia's. The only problem is the software.

And this IS how companies acquire driver/software devs. You go out, find companies who make software, and buy out the whole company. As I stated earlier: Acqui-hires are a tradition in Silicon Valley.

Anyone who builds a team of 300+ developers gets bought out at hundreds-of-millions, because the team is worthwhile together and not as piecemeal hires. Sure, some will leave because many preferred working for a small company but the vast majority of the team stays and continues to write software for the larger company.
Posted on Reply
#19
64K
dragontamer5788AMD's GPUs are already rivaling those of NVidia's. The only problem is the software.
Not just the software. AMD needs to improve the hardware side of their ray tracing. It's not critical yet but it will be more and more important as the years go by. Not trying to start a value of RT debate. The real answer to that is looking at what both AMD and Intel are claiming are important in their next GPU generations to see the answer. AMD has said in an article here that they will be improving that considerably in the next gen. Have to wait for reviews to see but I don't doubt it. If they are focused on the hardware and software side then it could be an upset for Nvidia. Probably not like the AMD delivered wake up call to Intel level event but we'll see soon.
Posted on Reply
#20
InVasMani
RT is getting to the point where it should become modestly acceptable especially at the mid range and high end. Upscale is certainly helping with that a bit. This coming generation should be a nice general improvement though the one after is where I see RT really thriving more readily and being a nice feature to insert into games a bit more actively w/o it feeling a bit over compromising. It's still going to be a performance hit, but the generation after this one I suspect RT will start to look and feel rather impressive at least on better mid range to top end SKU's in particular.

I think we'll see lots of changes in between to different things like upscale, but also post process in general will be better optimized, refined, and improved which will help.
Posted on Reply
Add your own comment
Jul 15th, 2024 23:17 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts