Crypto's Crystal Ball: Predicting Tech Costs with Cumulative Doublings

In the fast-paced world of cryptocurrency and emerging technologies, predicting the future is more than just a guessing game—it's a science. A recent analysis of cost decline curves across various technologies, from electric vehicles to AI, has unveiled a fascinating pattern: the power of cumulative doublings. This method, which has accurately predicted the cost trajectory of technologies like the Model T over a century ago, is now being applied to the crypto and tech sectors, offering a glimpse into what the future might hold for investors and enthusiasts alike.

From Model T to Crypto: The Power of Historical Data

Imagine being able to predict the cost of a car within 20% accuracy over a hundred years just by analyzing the first decade of its production data. That's exactly what researchers have done with the Model T, using the concept of cost per horsepower. Now, this same methodology is being applied to the burgeoning fields of cryptocurrency and advanced technologies. By examining the early stages of a technology's life cycle, experts can forecast its future cost with a surprising degree of precision. This approach not only offers a medium to long-term view but also helps in understanding which technologies will be meaningful to future buyers.

Cumulative Doublings: The Key to Unlocking Future Costs

The concept of cumulative doublings is central to this predictive model. It's not just about doubling production from one to two units; it's about the cumulative effect of repeated doublings—from one to two, two to four, four to eight, and so on. For electric vehicles, every cumulative doubling in the production of electric drive trains led to a 28% cost decline. In the realm of industrial robots, this figure jumps to 50%, while for short-read DNA sequencing, it's 40%. The most astonishing revelation, however, comes from the world of AI, where a combination of hardware and software improvements has resulted in a 48% cost decline per cumulative doubling.

The Astonishing Pace of AI and Crypto

What sets AI and crypto apart is the speed at which these cumulative doublings are occurring. In less than a year, AI training costs are dropping by 70%, and there's emerging data suggesting that inference costs could plummet by an astounding 98% for every cumulative doubling. This rapid decline is a game-changer for the crypto industry, where the cost of mining and processing transactions is a critical factor. As these costs continue to fall, the barriers to entry for new projects and the scalability of existing ones are dramatically reduced.

Experts like Brett, who have been closely monitoring these trends, emphasize the importance of the unit metrics used in these analyses. For AI, the combination of hardware and software improvements is a key factor in driving down costs. In the crypto world, similar metrics are being developed to track the efficiency of mining operations and the cost-effectiveness of blockchain technologies.

The implications for investors are clear: understanding these cost decline curves can provide a significant edge in predicting which technologies will dominate the future. As the crypto market continues to evolve, those who can accurately forecast the cost trajectories of emerging technologies will be better positioned to capitalize on the next big thing.

While the future is never certain, the science of cumulative doublings offers a powerful tool for those looking to navigate the complex world of crypto and tech. By paying close attention to the unit economics and demand dynamics of these technologies, investors and enthusiasts can gain valuable insights into what lies ahead. As the pace of innovation accelerates, the ability to predict future costs will become an increasingly valuable skill in the ever-changing landscape of cryptocurrency and beyond.