The Thin, Renewable Layer of Gold Dust Covering the Earth

Imagine that there is a thin layer of Gold dust that has settled on the entire land surface of the Earth. It's too thin to notice, mixed in with other dirt and dust, but because of the high price of Gold and the huge land surface area, it all adds up to 120 Billion Dollars worth of Gold. Just sitting there. And if you could somehow sweep it up and collect it, you could just wait a year and a new thin layer of Gold dust would settle down, and you could sweep up another 120 Billion Dollars worth of Gold every year.

But there's lots of challenges. First, what kind of incredible Elon-Musk-like operation would actually work to sweep up all that Gold dust from everywhere? And if you did sweep it up, it would be mixed in with all the other dirt and dust, so it would be useless unless you cleaned and purified it. And then you'd need a huge shipping network to send all these small shipments of clean Gold dust back to headquarters, melt it down, pour it into Gold bars, the convenient form that some customer would actually buy from you. And then you would need a Marketing operation to let the world know you had some new Gold bars at a particular quality level, and a Sales operation to go find customers for these Gold bars, and start booking sales.

That whole operation isn't cheap. It costs 60 Billion Dollars to do all of those things: sweep up the Gold dust, clean it, ship to to headquarters, melt it down into Gold bars, Market it, and Sell it. So, your revenue would be $120B/year, your Cost of Goods Sold would be $60B/year, Gross Margins would be 50%, and Profit would be $60B/year.

This little thought experiment is directly analogous to Neocortix.

The dirty Gold Dust particles covering the Earth represent the computational abilities of the #Arm processors in the 2.5 Billion Android cellphones all over the world. The best 10% of the phones (Samsung Galaxy S9-level or better) have processors that you could charge 5.5 cents/hour for. So that's 250 Million good phones * $0.055/hr * 24 hr/day * 365 day/year = $120.45 Billion/year. We pay the phone owners half what we charge, so we gladly pay them $60.225 Billion/year. The sweeper and the shipping network is the Internet, which is free (well, already paid for by someone else). The cleaner/purifier is our Secure Linux Container technology, which turns a useless Android phone into a very useful Linux server. The smelter is Neocortix Cloud Services, which aggregates all this compute capacity and sells it.

How would an entrepreneur get an operation like this started? You don't start with 250 Million phones, spending $60B/year, and then look for first customers. You start small, with 3,000 phones, spending about $200,000/year, and keep that supply-side operation going while building up the first useful products that customers can buy. Then finally you get a few key customers signed up, start to get some revenue, and investors can start to see that there is an enormous amount of money to be made in this new category: Shared-Economy Cloud Services. It scales unbelievably well on the supply-side, because 99% of the world's population needs more money and they have phones, and those phones keep getting better and better every year. The limiting factor is on the demand side. What are the great applications for this distributed network of mobile computing devices? The early great applications are Global Last-Mile Load Testing, Network Telemetry for 5G Rollouts, and cost-sensitive compute-intensive Academic Research, as we have demonstrated by running Folding@home and Rosetta@home in our PocketScience app. Those are enough to get us to early profitability with our initial Scalable Compute and BatchRunner products. But those applications and products are just our opening acts. We have already integrated 3D image rendering with Blender and POV-Ray, and Machine Learning with TensorFlow onto our platform, and we have built a Web Crawler running on phones. We have an extensive Product Roadmap with Docker and Kubernetes support, Intel and NVidia support, Storage, MicroFunctions, BlockChain, and much more. Cloud Services cannot exist on simple compute instances alone, it requires a stack of interconnected products, functions, and features to fully support the needs of sophisticated commercial customers, and we are working hard to provide those products in our overall platform.

There's $120B/year worth of Gold, just sitting there for the taking. Neocortix is the first company to credibly go after it.