AI: DeepSeek accelerates 'Small AI', locally and on clouds. RTZ #616

...mainframe era of AI rapidly fusing with PC/Smartphone era of AI in few years vs decades

Another consequence of this AI Tech Wave’s ‘DeepSeek’ Moment of doing ‘far more with far less’, is their open-source innovations have immense applications and utility in what I’ve been discussing as ‘Small AI’ in these pages spanning two years now. And the benefits of that for the industry to companies. Including big ones like Nvidia, Apple, Google, Microsoft, and most others.

It’s about more AI Reinforcement Learning (RL) driven Inference, and Training of LLM AI models being done not just in large, multi-hundred billion dollar networks of AI data centers, but ALSO on LOCAL devices used by billions of mainstream users.

Axios highlights this in “DeepSeek erodes AI industry's "size is everything" faith”:

“The first big casualty of the stock market's DeepSeek scare — aside from a few hundred billion dollars in frothy Nvidia valuation — is the AI industry's religion of scale.”

One can tone down the ‘frothy’ characterization above, due to Nvidia’s secular position in this AI Tech Wave that I’ve discussed at length. Axios goes on to describe the industry focus on Scaling AI:

“State of play: Ever since the advent of ChatGPT two years ago, U.S. tech firms, led by OpenAI, have shared the belief that AI will keep improving as long as we keep throwing more chips, money, power and data at it.”

“OpenAI CEO Sam Altman preached this gospel last year, writing that machine learning gets "predictably better with scale."

I’ve discussed this as well, from Sam Altman and Anthropic founder/CEO Dario Amodei. Axios discussed the same:

• "To a shocking degree of precision," Altman wrote in his "The Intelligence Age" essay, "the more compute and data available, the better it gets at helping people solve hard problems. ... AI is going to get better with scale, and that will lead to meaningful improvements to the lives of people around the world."

  

The rest of the industry largely agreed.

• Nvidia's high-end chips became coveted and scarce as its stock price skyrocketed.

• OpenAI's backer Microsoft, its chief rival Google, and cloud giant Amazon all got busy with colossal infrastructure investments.

• “A macho race to build bigger, more energy-hungry data centers began. Elon Musk's xAI built a big computing cluster in Tennessee in what he claimed was record time.”

But the industry is of course debating how long can this go on before the financial benefits start accruing at scale:

“But over the past year, the payoffs from this race to scale up have grown elusive.”

• “Some of the most significant advances in pushing the boundaries of AI have been made not by making it bigger but by building it differently.”

• “That includes OpenAI's most recent leap forward in the form of its reasoning model, o1.”

• “This new generation of AI models can solve more complex problems than their predecessors but not because they're bigger or have been trained more. Instead, they strategically use extra time and computing resources while figuring out their answers to users' questions.”

Then came DeepSeek:

“Over the past months, DeepSeek, a China-based research lab, has been steadily matching many of OpenAI's achievements using what it says is a fraction of OpenAI's budget.”

• “The DeepSeek R1 model, released last week, performs comparably to OpenAI's o1. The company has publicly released all its models for free download and use.”

“Why it matters: Maybe we can get what we want from AI without spending hundreds of billions of dollars on infrastructure or choking the planet with CO2.”

And of course it’s raised a ton of doubts and questions on the underlying themes of AI Scaling:

“The bottom line: As businesses and investors rethink the future, they're squeezing some air out of the AI bubble — and no one likes to see their portfolios lose value.}

• “But it's not as though OpenAI, DeepSeek and everyone else aren't still going to need to buy Nvidia's chips and build more data centers.”

• “They just may not need as crazy much as everyone assumed until Monday.”

All of this is not say that this is a ‘Zero Sum’ game of ‘Big AI’ vs ‘Small AI’. But that both of them can help create mainstream, affordable, reliable and safe AI applications and services in this AI Tech Wave.

As I’ve said before, it’s like enjoying the spectrum of mainframe/minicomputer computing waves of the 1960s to the PC computing wave of the 1980s, to the Internet computing wave of the 1990s, to the Mobile and Cloud computing waves of the 2010s and 2020s. All happening together over the next five plus years.

DeepSeek is just another important accelerator of this trend. LLM AI is increasingly going to be scaled at both ends, in the super-sized cloud and locally on devices/robots/wearables and far more in the physical world.

And that is what is truly exciting about the DeepSeek innovations this time, despite the China overlays. Stay tuned.

(NOTE: The discussions here are for information purposes only, and not meant as investment advice at any time. Thanks for joining us here)