Opinion: Gonca creators Danil and David Liberman
Bitcoin has taught us important things about building infrastructure. Rewarding efficiency can bring about incredible innovation.
Fifteen years ago, Bitcoin mining was run on the same graphics card used by gamers. The network was slow and the hashrate was low. But the Bitcoin Proof of Job (POW) system rewards miners who can handle blocks most efficiently.
This has created a race to build better hardware. Today, Bitcoin runs on machines (called ASICs) that are literally hundreds of thousands of times more efficient than the best Nvidia graphics cards. 10% not good, 100,000 times better. That’s what happens when you reward the most useful work.
In just 15 years, Bitcoin mining infrastructure has exceeded a whopping 16 gigawatt capacity. This not only represents the most efficient, but also has a much larger shell scale than the combination of Openai, Microsoft Azure, Amazon Web Services, Google Cloud and Xai, as well as enough capacity to run the most powerful Blackwell Nvidia GPU.
AI requires the same treatment
Today, AI runs expensive, general purpose chips. But imagine that if you build an AI network like Bitcoin, anyone can contribute to computing power and get paid for the most efficient execution of useful AI tasks.
Suddenly, you don’t need to have a phonebook and a team of Sales Brothers who can leverage the influence of C-Suite to sell chips.
Now it makes much more sense to build a chip specifically designed for AI tasks.
Hardware manufacturers compete to enable the cheapest and most efficient AI processors. The same market power that transformed Bitcoin (BTC) mining is only 10 times stronger this time, as there are true multi-billion dollar blockchain fans who have become rich from the Bitcoin Revolution.
Don’t be distracted by the proof
Many early distributed AI projects use Proof-of-Stake (POS) instead. This means rewarding those who hold the most tokens, not those who build the best infrastructure.
Related: Cooperate with the US, UK AI, quantum computing and nuclear development
For example, consider Bittensor. The most powerful computational capacity of the network is concentrated on the Subnet 64, which actually runs large AI models. However, the miner has earned just 5% of the network’s rewards. The other 95% go to token stakers or related miners who have contributed significantly less or none of the work.
This is behind. We need a network that rewards people who build better hardware, not people with the biggest stash they’ve frozen for yield extraction.
The real difference
Cryptocurrency is not the only proof of work. Instead of embracing the hardware that exists now, it’s about using competition to drive innovation. In just 10 years, the blockchain community can build infrastructure that produces thousands of times more computation than other regions of the centralized market.
For AI, this could mean the difference between expensive, centralized computing and intelligence that is as cheap and abundant as electricity. Running AI models within a few years costs little to do.
What does this mean to you?
Today’s AI is like Bitcoin in 2009. The network is just beginning and early participants have the greatest opportunity.
Follow the Proof of Work AI Project. Contributes computing power or rents from the market. Start mining. Those who build this infrastructure will benefit most now.
Opinion: Gonca creators Danil and David Liberman.
This article is for general informational purposes and is not intended to be considered legal or investment advice, and should not be done. The views, thoughts and opinions expressed here are the authors alone and do not necessarily reflect or express Cointregraph’s views and opinions.