DeepSeek could represent Nvidia CEO Jensen Huang's worst nightmare

Could someone explain as to why DeepSeek is bad for Nvidia?<p>The demand for Nvidia GPUs should now go up. Now anyone can run a GPT-like model by themselves. It&#x27;s a prime time for businesses to start investing and setting up on-prem infra for that. I know some have been avoiding ChatGPT due to legal concerns and sensitive data.

I don&#x27;t for a minute believe Deepseek v3 was built with a $6M rental.<p>Their paper (arxiv 2412:1947) explains they used 2048 H800s. A computer cluster based on 2048 GPUs would have cost around $400M about two years ago when they built it. (Give or take, feel free to post corrections.)<p>The point is they got it done cheaper than OpenAI&#x2F;Google&#x2F;Meta&#x2F;... etc.<p>But not cheaply.<p>I believe the markets are overreacting. Time to buy (tinfa).

All this media frenzy around DS V1 makes me feel sick to my stomach.<p>It increased the noise in the AI space by orders of magnitude. Every media outlet is bombarding you with a relentless torrent of half-true information, exaggered interpretation of single facts and speculation.<p>Sometimes I wonder whether this is amplified by a state actor?<p>Also curious, how Minimax-01, which is also an excellent model with impressive improvements, went by completely unnocited.<p>The only good thing is that this certainly put an end to OpenAIs price gauging - pretty sure that $200&#x2F;month individual plan is not the limit of their imagination.

It can&#x27;t get sillier than this. All that DeepSeek did were showing that NVIDIA products hold more value than believed, and that US dominance more illusory than thought. It has zero negative bearing on eventual path to AGI.<p>And that&#x27;s somehow supposed to be end of NVIDIA? Hello?

It is the end of neural language model scaling as described here:<p><a href="https:&#x2F;&#x2F;openai.com&#x2F;index&#x2F;scaling-laws-for-neural-language-models&#x2F;" rel="nofollow">https:&#x2F;&#x2F;openai.com&#x2F;index&#x2F;scaling-laws-for-neural-language-mo...</a><p>And a return to more normal world where progress comes from refinement of algorithms and approach rather than brute force compute.<p>Deepseek is just the poster child.

There was no market reaction when Deepseek r1 was released.<p>There was no market reaction when Deepseek v3 was released.<p>There was a massive market reaction when a Deepseek app was released.<p>There is too much damn irrationality in the AI and AI adjacent market right now, and pop articles like this are only making it worse.<p>Jevons Paradox absolutely holds for model development, and Deepseek should (and is - all my cybersecurity startup peers are investigating the feasibility of building their own models now) be viewed as an opportunity for smaller or medium stage companies to build their own competitive domain specific models, while not having to pay what is essentially protection money to OpenAI or Anthropic.<p>The only people at risk with the democratization of model development are the investors in foundational model companies<p>---------<p>Also, I hope Deepseek FINALLY reprioritizes distributed systems education in CS.<p>There are too many &quot;ML Engineers&quot; who do not know the basics of OS or Hardware Engineering (eg. Cannot optimize Mellanox hardware, really dig into workload optimization, etc) and are thus burning resources inefficiently. Most MLEs I meet now just import a package and glue code together WITHOUT also understanding the performance capabilities that their compute has.<p>Most universities in the US either don&#x27;t require OS or CompArch classes for CS majors (eg. Harvard) or dumb them down significantly (eg. Not going deep into Linux kernel implementation, scheduling, etc).<p>This is why the entire cybersecurity industry has largely shifted to Israel and India, and the skillset overlaps significantly with MLOps and MLEng (if you know how to implement spinlocks in C you can easily be taught CUDA)<p>&lt;&quot;Old Man yelling at clouds&quot; rant over&gt;

R1 and O1 points towards a world where training models will be a small bucket of the overall compute. That doesn&#x27;t mean the total amount of AI compute will stop accelerating, just that interconnected mega clusters is not the only or most efficient way to run a majority of future workloads. That should be negative news for the company that is currently the only one that is capable of making chips for these clusters, and positive news for the players that can run inference on a single chip, as they will be able to grab more parts of the compute pie.

Was the US sanctions on chip selling to China worth it? Yesssss. It was. It still is. (But I frown on this action.)<p>China is good on dumb stupid manufacturing. It does this with so much enthusiasm HNers are left scratching their heads why the excitement. The possibility of China manufacturing the next computing powerhouse was too high. Higher than it could ever be in the US.<p>The threat to US leadership was that the fastest supercomputer for AI in the world could be in China (the first sputnik moment per Barack Obama). So the chip embargo was to prevent that from happening.<p>So China did not spend&#x2F;waste $$ &amp; time building hardware. They spent it on software. Normally the Chinese version is cheap enough and good enough. But this time it was cheap enough and surprisingly terrific enough. Props to Liang the founder for having that game face on.

Nvidia AI Chip smuggling is a huge business in china. DeepSeek company on its own smuggled 50k h100.

I find it hard to believe that serious investors thought that vertical progress in ai had peaked, and it would be only horizontal expansion from now on. We know that it takes decades for this to happen in every industry.<p>It seems more like , investors were looking for a significant event to pop the obvious bubble if the last year

Time to give our family it&#x27;s own shared networked AI station, accessible from any of our devices. Then I&#x27;ll trust it with all of our personal, medical and legal docs. Seems to me the market is expanding, not contracting.

Is deepseek’s architecture not scalable anymore? I can see in a couple months some company with access to GPUs create a much larger and better performing model based on the deepseek architecture.

NVDA could go to zero and he’d still be a multi billionaire. His worst nightmare isn’t very bad.

&gt; “We didn’t think China was ever going to compete in AI,” said Dion Hinchcliffe, vice president of the CIO practice at Futurum Research last week.<p>What? Oh, come on now.

Does DeepSeek make it more possible to use AMD hardware? What chips are possible with DS that weren’t possible before?

Did they say which chips they used? And how many? I haven’t seen this in any reporting

Wait until some surprise new silicon suddenly appears

DeepSeek could represent a much larger amount of costumers to nvidia since now people can get an AI while spending less<p>ftfy