Opinion by: Daniil and David Liberman, Gonka Creator
Bitcoin taught us something important about developing infrastructure: when you reward efficiency, you get incompatible -believed change.
Fifteen years ago, Bitcoin Mining ran on the same graphics cards used by players. The network is slow, and the hashrate is low. But the bitcoin Proof-of-Work (Pow) System rewards can process blocks better.
It created a race to produce better hardware. Today, Bitcoin runs on machines (called Asics) literally a hundred -thousand times better than Nvidia’s best graphics cardics. Not 10% better, 100,000 times better. That is what happens when you reward the most useful work.
In just 15 years, the Bitcoin mining infrastructure exceeds a 16-GiLiwatt capacity. It represents sufficient power capacity to run 10 million of the strongest Blackwell Nvidia GPU – not only better, but larger in thin size than Openai, Microsoft Azure, Amazon Web Services, Google Cloud and Xai combined.
AI needs the same treatment
So far, AI is running in expensive, general-purpose chips because that’s what is available. But imagine if we built AI networks such as Bitcoin, where anyone can contribute to the power of computing and get paid for the most efficient implementation of benefits to AI.
Suddenly, you don’t have to have a team of sales bros with phonebooks that can use their C-suite clout to sell chips-no matter who can make, install and start “issue money.”
It is now made more meaning to produce chips that are specifically designed for AI activities.
Hardware manufacturers will compete to make the cheapest, most efficient AI processors possible. The same market forces that changed bitcoin (Btc) Mining is kicking. Only at this time with 10x more power because there are true blockchain believers with billions of dollars rich from the bitcoin revolution.
Do not be interrupted by proof-of-stake
Many early AI decentralized projects are the user Proof-of-Stake (POS) Instead – that simply means that the rewards come to anyone holding the most tokens, not anyone who builds the best infrastructure.
Related: US, UK to cooperate with AI, Quantum Computing, Nuclear Energy Development
For example, get a bittenor. The most powerful network compute capacity is concentrated in subnet 64, which actually operates large AI models. However its miners only get 5% of network rewards. The other 95% goes to token stakers or associated miners who contribute significantly or unemployed.
It is backward. We need networks to reward people who build better hardware, not people with the biggest frost stashes for yield hiring.
The real difference
Proof-of-work is not just about cryptocurrency. It is about using a competition to drive a change rather than accept any current hardware that currently exists. In just 10 years, the blockchain community can develop infrastructure that produces thousands of times more compute than the remaining centralized markets.
For AI, this can mean a difference between expensive, centralized computing and a intelligence that is as cheap and abundant as electricity. Within a few years, the operation of AI models may cost almost nothing.
What does this mean for you
Now for AI is like 2009 for Bitcoin. The networks are just beginning, and the first participants have the biggest opportunities.
Follow proof-of-work AI projects. Contribute to the power of computing, yourself or rented from the market. Start mining. People who build this infrastructure today will benefit.
Opinion by: Daniil and David Liberman, Gonka creator.
This article is for general information purposes and is not intended to be and should not be done as legal or investment advice. The views, attitudes, and opinions expressed here are unique and do not necessarily reflect or represent the views and opinions of the cointelegraph.
