Fullmoonfever
Top 20
Part 3
What could be next for Nvidia? Neuromorphic Computing?
Okay, so to understand neuromorphic computing (NC), all you need to know is this formula:
Juust kidding.
In short, the answer is much simpler: The future of chips is making physical computers think more like human brains, where chips use the same physics of computation as our own nervous system.
The term "neuromorphic" comes from the Greek words "neuron" (meaning nerve cell) and "morphe" (meaning form). In the context of computing, it refers to the use of electronic circuits and devices inspired by biological neurons' structure and function.
While still nascent, this is another massive technological feat. You might be thinking—but we’ve done something similar already with AI neural networks, what’s the difference?
Simply, we’ve made great progress on the software side of things in terms of mimicking the logic of how a human brain thinks. But solving those challenges on a physical chip is a different beast.
That’s what NC is solving though, and you can imagine how much more advanced our AI and computing will be when we have the chips and software both working in unison like a brain.
And just to illustrate the monumental difference on the chips’ side:
Just imagine how fast things would be changing in the AI landscape if these were powering Nvidia’s data centers and AI platforms.
It sounds somewhat far-fetched, but that future is already here and in the works.
As was published in this research paper in Nature:
In the long term, NC poses a technological obsolescence risk to traditional GPUs and DPUs. If these types of chips become successful, it could threaten Nvidia's business.
However, because NC has the potential to be a game-changer in many different areas of society, and its consequences could be far-reaching and complex, I have zero doubt in my mind that Jensen and his crew are sitting in a Denny’s somewhere, dice in hand, and mapping out Nvidia’s strategic future over some burnt coffee.
What could be next for Nvidia? Neuromorphic Computing?
Okay, so to understand neuromorphic computing (NC), all you need to know is this formula:
Juust kidding.
In short, the answer is much simpler: The future of chips is making physical computers think more like human brains, where chips use the same physics of computation as our own nervous system.
The term "neuromorphic" comes from the Greek words "neuron" (meaning nerve cell) and "morphe" (meaning form). In the context of computing, it refers to the use of electronic circuits and devices inspired by biological neurons' structure and function.
While still nascent, this is another massive technological feat. You might be thinking—but we’ve done something similar already with AI neural networks, what’s the difference?
Simply, we’ve made great progress on the software side of things in terms of mimicking the logic of how a human brain thinks. But solving those challenges on a physical chip is a different beast.
That’s what NC is solving though, and you can imagine how much more advanced our AI and computing will be when we have the chips and software both working in unison like a brain.
And just to illustrate the monumental difference on the chips’ side:
- Your computer operates in binary. That’s 0s and 1s; Yes and Nos. It’s rigid, so, the code we use and the questions we ask these kinds of machines must be structured in rigid way.
- With NC though, we go from rigid to flexible, as these chips will bring computers to the ability to have a gradient of understanding. I’m no engineer, but from what I’ve read, that’s huge.
- It promises superior processing speed
- It has the ability to solve more complex problems (and faster)
- It can learn and adapt in real time. Compared to traditional AI algorithms, which require significant amounts of data to be trained on before they can be effective, NC systems can learn and adapt on the fly.
- Neuromorphic computing is more energy efficient. Great for lowering the environmental impact, and also enabling AI systems to be deployed in resource-constrained areas.
- NC allows us to create AI systems capable of true creativity and artistic expression that can learn and adapt in ways similar to human cognition.
- One example, it has the potential to enable hyper-realistic generative AI.
- NC provides privacy and security advantages.
- NC is poised to be vital for developing AI systems that are more resilient to noise and errors
Just imagine how fast things would be changing in the AI landscape if these were powering Nvidia’s data centers and AI platforms.
It sounds somewhat far-fetched, but that future is already here and in the works.
As was published in this research paper in Nature:
And builders are going after it. Intel is already working on these chips, as are various other startups.
In the long term, NC poses a technological obsolescence risk to traditional GPUs and DPUs. If these types of chips become successful, it could threaten Nvidia's business.
However, because NC has the potential to be a game-changer in many different areas of society, and its consequences could be far-reaching and complex, I have zero doubt in my mind that Jensen and his crew are sitting in a Denny’s somewhere, dice in hand, and mapping out Nvidia’s strategic future over some burnt coffee.
🫵
? 
