Skip to content Skip to footer

Why Is OpenAI Spending $51 Million on a Chip Startup?

Synthetic intelligence capabilities are advancing at a breakneck tempo due to improvements by main AI labs like OpenAI. Now, OpenAI plans to catapult its AI developments even additional into the longer term with a $51 million funding in Rain AI, a startup targeted on groundbreaking neuromorphic chip expertise.

With personalized {hardware} tailor-made to imitate neural pathways within the human mind, OpenAI may quickly smash by means of current limitations on how clever and responsive AI techniques will be. If the collaboration pans out as deliberate between these two trailblazing firms, new frontiers of synthetic intelligence may quickly unfold—from how seamlessly AI interfaces with on-line platforms to unlocking basic intelligence exceeding human cognition.

Why Is OpenAI Spending $51 Million?

OpenAI signed an settlement in 2019 to doubtlessly buy $51 million in neuromorphic chips from Rain AI, and paperwork from Rain AI said that Sam Altman had personally invested greater than $1 million within the firm. The paperwork obtained by Wired additionally talked about OpenAI’s letter of intent as an element that helped Rain AI acquire funding. The fund had led a $25 million fundraise introduced by Rain in early 2022, which was later pressured to promote by the Joe Biden administration. 

OpenAI can also be mentioned to be launching its personal synthetic intelligence chip firm, the reason is due to the excessive costs and shortage of GPU chips from Nvidia which might be at present limiting AI growth. There’s undoubtedly a marketplace for extra AI chips given Nvidia makes billions of {dollars} promoting AI GPUs, however they can not sustain with demand. Their excessive costs and lengthy waitlists are hindering the trade.

To offer you an concept, Nvidia made $14.5 billion in knowledge middle income final 12 months (principally from AI chips) in simply three months! That is a 206% improve. Large firms equivalent to Microsoft (which makes use of OpenAI techniques) and Meta bought over 150,000 GPUs from them. However to obviously perceive why OpenAI was prepared to take a position a lot in Rain AI, we first want to take a look at what Rain AI is and what it is able to doing.

What Is Rain AI?

Rain AI is a San Francisco-based startup situated lower than a mile from OpenAI’s headquarters. They’re engaged on neuromorphic processors (NPUs), that are laptop chips designed to imitate points of the neural structure of the human mind. When in comparison with conventional graphics processing items (GPUs), the purpose of those NPUs is to offer extra computing energy and power effectivity for AI functions.

Rain claims that their NPUs may produce 100 occasions extra computing energy whereas utilizing 10,000 occasions much less power than GPUs throughout AI mannequin coaching. Rain’s neuromorphic processors intention to carry AI capabilities extra in step with human cognition by way of effectivity and efficiency by mimicking key traits of organic neural networks. Their brain-inspired method is what separates them from conventional AI {hardware} within the race to construct higher AI.

How Do Neuromorphic Processors Work?

Neuromorphic processors (NPUs) are laptop chips that mimic points of organic neural networks like these present in human brains. They work in a different way than conventional laptop processors, to provide you a greater perspective right here is the way it works:

  • Enter knowledge comes into the NPU as electrical spikes, just like indicators between neurons. This enter reaches synthetic synapses, primary computational items that join the enter and output layers.  
  • Every synapse has a weight that determines the power of the connection from inputs to outputs. These synaptic weights are adjusted throughout coaching because the NPU learns to carry out duties.
  • When sufficient enter spikes attain a synapse, the subsequent neuron within the community is activated sufficient to fireside its personal output spike, propagating the sign.
  • Studying happens by means of synaptic plasticity – the variation of synaptic weights throughout neuronal connections. This course of permits the NPU to study and keep reminiscence like a mind.
  • Not like conventional processors that execute directions sequentially, NPUs carry out parallel processing throughout their neural community structure, enabling sooner processing for AI functions.

The tip consequence behaves equally to a neural community, able to studying duties by means of examples and environmental interplay. However this brain-inspired design permits environment friendly computation whereas requiring much less power than conventional {hardware}.

How Rain Chips May Change AI in The Future

Rain’s first chips depend on the RISC-V structure, which is actually an open-source chip blueprint backed by massive names like Google and Qualcomm. Rain is geared at edge units equivalent to smartphones, drones, and even vehicles. As a substitute of large knowledge facilities, these are digital units distributed all through the world.

The important thing level is that Rain needs to create chips able to coaching AI fashions and operating them after deployment. Most edge chips now solely concentrate on inference, which is why this is able to be helpful if OpenAI wished to push AI intelligence to units relatively than the cloud.

Nevertheless, it’s nonetheless being decided how OpenAI will use Rain’s expertise. Nonetheless, having chips carry out end-to-end AI from coaching to inference immediately on units can be vital. It has the potential to place superior intelligence on issues like robots and self-driving vehicles and doubtlessly change the way in which we use AI fully. 

Leave a comment