Wow! Thanks so much @overpup an@Mr Tech Laden for highlighting...

Wow! Thanks so much @overpup an@Mr Tech Laden for highlighting this report. Here it is in full.

Brainchip, Next Generation AI Aug. 11, 2021 9:20 PM ET BrainChip Holdings Ltd (BRCHF)
https://seekingalpha.com/instablog/10313401-fundamental-stock-analysis/5626322-brainchip-next-generation-ai#comments

What this means is that Akida can be integrated into an existing chip design, with just the IP-blocks that the company needs. This is saving companies space (crucial in modern compact devices), it saves money and it can mean speed advantages. Basically the same strategy that ARM used and incidentally they have been hiring former ARM executives like Antonio J. Viana and Rob Telson.

Target Market

The target market is what’s called edge computing, which to some degree is synonymous with IoT. To put it in a simple way, it’s everything that computes in the field and not on a server. This can be everything from mobile phones to satellites. It’s a huge market that is currently in development as smart home, smart health, IoT etc. and there currently isn’t any other AI solution that is ideal for AI computing on the device.

Brainchip, Next Generation AIAug. 11, 2021 9:20 PM ETBrainChip Holdings Ltd (BRCHF)1 CommentPlease Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.SummaryExtremely fast learning AI from Australia.Perfect for implementation in small devices (Edge).Production chip being taped out now.No serious direct competition in it's market segment.First contract and many partners.A new breed of AI has been created in Australia, it’s extremely fast learning, consumes very little energy, it’s currently in it’s first mass production and I will in the following argue whe I think it's extremely undervalued.

When Akida from Brainchip is expected to be available in Q3 this year, it will have a unique market position and baiscally no significant compition in it’s target market.

What is Akida?

Akida is extremely advanced and ahead of the only real direct competition which is IBM and Intel, who are still in the research stage. With 1.2 million neurons and 10 billion synapses, it’s more complex than both Intel and IBM’s research chips. It can perform a billion synaptic operations per second, where IBM and Intel’s research chips performs below 50 million synaptic operations per second.

There are many AI companies in the world, I’ve been skimming over 50 in various levels of detail and what I notice is that there are two general paths. One is to accelerate software models with more conventional computing like; nVidia, GraphCore and Cerebras. The other is to develop a chip that mimics the human brain like IBM, Intel and Brainchip.

The vast majority of AI companies are focused on more conventional computing and extremely few are seeking to make a chip that mimics the human brain. The most successful at this are Intel, IBM and Brainchip.

Brainchip is the first company to actually have a significant Spiking Neural Network (SNN) in mass production (Akida). Not only is the technology ground breaking and catching a lot of attention in the industry, but the business plan is very nimble and scalable. They basically offer hardware designers any part of their intellectual property (IP-blocks) for licensing or if they prefer, they can have a complete independently functioning system on a chip (SoC). I consider this the key to Brainchip’s future success, because judging from the history of Intel and IBM, neither are going down that path.

So far most AI have been done server-side, which means not on the device, because this kind of computing have simply been to intensive for the mobile processors. This is going to change with Akida and it’s the first time in history that you can have data-center AI capabilities on a phone.

This translates into a unique market position, in a field that is rapidly expanding, with little competition in edge AI. They have already signed an IP contract with Rensas Electronics America and as such have secured some future revenue, they further have a lot of partners like; NASA, Biotome, Vorago, Magik Eye and others.

AI is the core to Industrial Revolution 4.0 and is the core in most trends described ”Symbiotic Autonomous Systems” by IEEE. AI will increasingly be able to solve tasks in place of humans and as such Akida should be seen as workforce on chip, which is incredibly powerful and incredibly valuable.

Valuation

Now the interesting part for value oriented long-term investors is the price, about 650 million USD or just about 900 million AUD. The closest comparison in regards to potential future value is ARM that was recently sold for 40 billion USD and I don’t mean this as a price target, but just to put Brainchip’s valuation into perspective.

Brainchip have been through a long and heavy development phase and it’s been a tough time for their investors. The product has been thoroughly tested by Brainchip, it's been reviewed by their Early Access partners, they are in production, the chip is working and they have a contract.

When all of this is going to manifest it self in a near future, when products hit the market and when significant revenue is hitting the financial statment, this company simply must be valued more than a billion USD.

Because of the acceleration of AI and interest in edge AI, I believe this unique company, with it’s unique position, to become worth many multiples of it’s current value.

A few additional details:

Brainchip on the Australian Stock Exchange (NYSE:ASX) under the ticker BRN.

Brainchip also have a road-map for Akida, that will increasingly mimic functionality of the human brain and integrate functionality based on feedback from the industry.

Akida is the first kind of AI that offers one-shot-learning, meaning that the chip can learn to recognize a new object from just one exposure. Normally this is a very heavy process that needs hundreds of exposures and retraining of a model.

Akida can continue to learn in the field (on the edge).

Akida uses less than 1 watt power and depending on the application it can be sub 0.1 watt.

Brainchip has a Scientific Advisory Board with professors Simon J. Thorpe, Adam Osseiran and Allan R. Harvey, founder Peter Van der Made and a Nobel Laureate Barry J. Marshall.

Rensas:

First Akida IP License Agreement

Biotome:

BrainChip Research Enter into a Research Collaboration with Biotome for SARS-CoV-2 Antibody Detection

Vorago:

BrainChip and VORAGO Technologies Agree to Collaborate through the Akida™ Early Access Program

Magik Eye:

https://brainchipinc.com/brainchip-partners-magikeye-breakthrough-3d-vision-solution/

Disclosure: I/we have a beneficial long position in the shares of Brainchip either through stock ownership, options, or other derivatives.

Additional disclosure: Everything in this article is my personal observations as I saw them at the time of publication and are written based on my best beliefs.

This is in no way a buy recommendation and no shares in Summit Wireless should be purchased based on these notes alone. There are risks with all investments and you should do you're own due dilligence to uncover those risks.