Original Title: "AI, Crypto, and America"
Author: Derek Edws, Managing Partner at Collab+Currency
Translated by: Zen, PANews
I am no longer certain that our traditional strategic frameworks can adequately address the new realities of the future - especially given the astonishing acceleration of AI technology and its disruptive potential for the American workforce.
America must begin to move towards a new framework to adapt to the new labor reality. This starts with understanding the two most disruptive technological innovations of the past two decades: AI and Crypto.
Artificial Intelligence
Three decades ago, internet search engines first emerged. Their value seemed simple - to search the world's data. Twenty-five years later, OpenAI's ChatGPT has become the fastest-growing consumer application in history. The product has 300 million weekly active users and over 10 million monthly paid subscribers. In less than two years, the product has an annual revenue of around $3 billion.
What is the value of these AI systems? It is the combination of "retrieval" and "completion". Simply put:
Request any task in natural language;
Have the computer intelligence handle the completion of the task.
Over the past three months, we have seen these products rapidly improve in their ability to fulfill task requests, spanning an ever-expanding domain - text, math, audio, video, geometry, programming, and more.
So far, the scaling assumptions for AI systems still hold: more data, more compute power, and better models. Moreover, new dimensions of scaling are being explored, like test-time compute innovations. Last month, OpenAI's latest reasoning model scored 25% on frontier math benchmarks, problems that only deep domain experts could solve.
These next-generation reasoning models are adapting to new tasks, producing observable results. Meanwhile, new advances in robotics are making machines increasingly capable of executing complex physical tasks. Intelligent, humanoid hardware will work tirelessly, never taking a vacation or going on strike.
This unleashing of technological productivity is not unprecedented - the Industrial Revolution of the 18th century and the Digital Revolution of the 20th century dramatically lowered costs, increased efficiency, and transformed various forms of labor markets and economic structures.
But AI seems to be a different kind of technological revolution. These systems can create value in ways that transcend traditional cost structures. Anthropic's CEO Dario Amodei believes that AI may deliver 50 to 100 years of biological progress in the next 5 to 10 years, simply by shifting most of our human labor structures onto AI systems.
As the underlying infrastructure supporting these systems continues to develop, I believe two trends will emerge for the global human population:
The cost of all cognitive labor will trend towards the cost of running AI systems;
The cost of all physical labor will trend towards the cost of robotic mechanical parts.
The promise of this technology is clear:
Unlimited output;
Dramatically lower prices;
Across every value category.
For the purposes of this essay, let's assume the above trends hold true. As someone who deeply believes the next generation of Americans should have the same opportunities as their predecessors, I have two persistent questions: First, how can America leverage its advantages to win the AI revolution? Second, how can individual Americans participate in the future gains of AI systems, given their disruptive potential for human labor?
Cryptocurrencies
In 2008, Satoshi Nakamoto published the Bitcoin whitepaper. The paper proposed a new, gamified accounting system where a distributed network of computers could work together to achieve a shared digital consensus on the Bitcoin ledger.
Today, Bitcoin has become the world's most powerful supercomputer. The scale of this network far exceeds the combined network scales of companies like Amazon, Google, and Microsoft. Yet, after fifteen years of development, blockchains still seem clunky:
Poorly written code can have vulnerabilities, leading to theft or account balances being drained to zero;
User errors in managing private keys are the biggest cause of cryptocurrency thefts last year;
Decentralized applications are more difficult to use compared to traditional web apps, with extremely high user churn rates.
Despite these limitations, engagement with cryptocurrencies has reached historic highs. One study estimates that 40% of the US population now owns cryptocurrencies, up from 30% in 2023. Globally, there are around 24,000 active developers contributing to blockchain and blockchain applications each month, up from just 1,000 per month a decade ago.
Despite all these constraints, why are cryptocurrencies still growing? I believe it is because cryptocurrencies possess five unique characteristics that work in concert, and cannot be replicated by any other database architecture:
Digital Ownership. Blockchain databases are global, fully auditable, community-owned, tamper-resistant, and run 24/7. Through blockchains, individuals can own any digital object on the internet, establishing the first global system of digital property rights.
Coordinated Incentives. Through self-executing contracts, blockchain-based protocols can leverage programmable incentive mechanisms to coordinate new forms of digital work on the network. These incentives can be used to reward those who use the product or service, provide economic security, contribute core code, provide market supply or demand, or refer others to use the product.
Frictionless Micropayments. Today, most internet companies must rely on subscription, bundling, and ad-supported business models due to prohibitively high transaction costs. This constraint limits innovation in internet business models and prevents consumer-friendly alternatives from emerging. Blockchain databases can enable frictionless, low-cost digital payments globally - bypassing the inefficiencies of traditional payment systems and without the risk of chargebacks. Furthermore, any crypto asset can be divided into infinitesimally small units.
Shared Standards. By leveraging shared settlement standards on the blockchain, various protocol tokens, stablecoins, applications, games, and financial services can seamlessly integrate with one another, like Lego bricks.
Distributed Security. Blockchain networks are typically distributed across multiple global nodes, eliminating single points of failure. This decentralized architecture makes it much harder for malicious actors to attack these network systems, as they would need to simultaneously control a majority of the nodes.
Today, the crypto economy has a market capitalization of around $3.6 trillion, spanning multiple emerging domains.
Over the next decade, I believe the crypto economy will be repriced significantly higher, driven by two key intersections of AI and cryptocurrencies: (1) AI and crypto infrastructure; (2) AI and crypto applications.
AI and Crypto Infrastructure
To understand the current landscape of AI infrastructure, it can be helpful to look for historical parallels.
In 1849, the California Gold Rush quickly attracted massive investment. To enable rapid transportation, hundreds of new roads were built. The port of San Francisco became one of the busiest in the world, shipping out gold miners, goods, and tools to all corners of the globe. A powerful banking and financial system emerged to support the needs of the new global enterprises. The infrastructure investments of that era laid the foundation for the region to become an economic powerhouse.
Here is the English translation of the text, with the specified terms preserved and not translated:175 years later, the world is witnessing a similar gold rush, this time targeting the creation of Artificial General Intelligence (AGI). However, the infrastructure supporting AI is not limited to a specific region, as the networks of data, computing power, and electricity are being built by competitors globally.
Unsurprisingly, the capital and computing power required to train, optimize, and deploy AI infrastructure is extremely expensive, affordable only to a few companies. Estimates suggest that a single training of GPT-3 costs over $4 million, while the training of GPT-4 exceeds $60 million.
More capital, more computing power, better performance.
While I am very proud and supportive of the contributions the United States has made to AI in traditional corporate forms, I believe it is necessary to acknowledge the structural limitations:
Value capture. Although centralized companies can leverage venture capital to drive important AI innovations, the economic benefits of these products are limited to a few shareholders, restricting their broader impact on society.
Proprietary knowledge. Advancements in technology frameworks are often concentrated within centralized companies, limiting the opportunities for others to access critical breakthroughs - while information now flows globally at internet speeds.
Opaque systems. Concentrating AI within opaque, closed-source centralized systems makes it difficult for independent verifiers to audit the company's practices in areas like data collection, security, and accountability.
Closed competition. The massive computing resources required to develop advanced AI pose a significant barrier to market entry for new products, with only a few well-funded companies able to continuously overcome technical bottlenecks.
By combining the United States' AI infrastructure with the five unique properties of cryptocurrencies - digital ownership, coordinated incentives, frictionless micropayments, shared standards, and distributed security - I believe we can mitigate the negative impacts of centralized AI and restore the competitive spirit of the American capital markets. Furthermore, I believe this combination of the United States' AI infrastructure with cryptocurrencies will lead to AI systems with (a) higher transparency and (b) more equitable ownership for the millions of future American participants.
(a) Better Performance
To understand the breakthroughs AI can achieve without massive funding, one can look at the DeepSeek team. Two weeks ago, this China-based research group released DeepSeek-V3, a 670B parameter model that performs on par with many closed-source SOTA models, including GPT-4o and Claude-Sonnet-3.5. DeepSeek has not received any venture capital funding to date.
As open-source projects like Bitcoin and Ethereum have demonstrated, programmable incentives distributed to a global pool of contributors can significantly boost the qualified labor and computing networks, forming a more powerful force than a single laboratory or centralized system. From this perspective, creating a system that rewards AI labor and computing networks is not much different from creating a system that rewards Bitcoin labor and computing networks.
A few examples:
For improved training data, a crypto network can reward human contributors of high-quality annotated datasets - including private data, proprietary intelligence, or information not captured in traditional web crawling.
For more powerful computing networks, a crypto network can incentivize individuals and organizations to contribute computing power through a decentralized market - without upfront capital investment, rapidly building a global machine network.
For more efficient model training, open-source developers can be rewarded for customized contributions, improvements, and optimizations to existing models. Furthermore, when code and model weights are public, hundreds or thousands of researchers and developers can simultaneously release improvements, debug issues, fine-tune custom models and agents, and create new applications - all coordinated through the network's incentive mechanisms.
Over time, I believe this broad collaborative approach pioneered by decentralized projects like Nous Research, Prime Intellect, and Bittensor will surpass what resource-rich private companies can achieve internally.
(b) Higher Transparency
Open-source AI models allow the research community to thoroughly examine their training process, architecture, and behavior, and make improvements. This transparency helps identify potential risks or biases early on, leading to more reliable systems that people can trust. By leveraging blockchain in this process, the entire lifecycle of creating, rewarding, and improving AI protocols can be kept transparent and auditable.
(c) More Equitable Ownership
Crypto networks designed for the various verticals of the AI technology stack will establish more equitable ownership structures than the current centralized models. Through programmatic incentives, all contributors and participants in the crypto protocols can be transparently rewarded.
Furthermore, the complete markets that will form around the various types of work in the AI infrastructure stack will drive more granular competition across AI design domains. Categories and subcategories of data, computing, training, and deployment can compete and accrue value in independent environments.
However, the ultimate beneficiaries will not be limited to the AI infrastructure. I believe AI agents will fundamentally reorganize the current trajectory of crypto adoption across all application verticals.
AI and Crypto Applications
The complexity of crypto applications has long been seen as a major barrier to widespread adoption. Over the past fifteen years, blockchains have required users to navigate complex approval processes, manage private keys, and understand complex UI patterns - all beyond the grasp of most internet users.
However, with the emergence of agent technology, these user patterns are rapidly changing. If you view AI models as a responsive infrastructure, making decisions based on prior training data, then you can view AI agents as proactive applications, integrating models into new architectures to accomplish narrow objectives.
In simple terms, AI agents leverage base models to think, plan, and execute actions automatically. The key to understand is that agents are different from what we've known as "robots." Unlike robots, AI agents can reason on-demand. They can analyze their own performance, adjust their strategies, and solve complex tasks that sometimes require hundreds or even thousands of unique steps.
In September 2024, I met with a founder in my investment portfolio who is building an AI agent protocol for blockchain navigation, called Wayfinder. Through his phone, using a few simple natural language prompts, I deployed a frontend and token contract, cloning Bitcoin's monetary policy onto the BASE blockchain, and using ETH bridged from the Ethereum mainnet. The entire process took less than four prompts and five minutes to complete.
Applications like Wayfinder demonstrate an important trend: AI agents will mediate the long-standing technical frictions of cryptocurrencies. Over the next 12 months, agents will translate blockchain's complex structures into seamless natural language interactions, increasing protocol accessibility, shielding users from their own mistakes, helping developers deploy safer code, and dramatically reducing consumer churn in complex decentralized products and services.
More importantly, key management networks will extend all these agent capabilities, allowing agents to execute tasks across blockchains seamlessly without human involvement. Global namespace networks will enable each agent's actions to span all blockchains, connected to a human identity.
In summary:
Crypto agents make it easier to build or use any crypto product, regardless of blockchain.
In decentralized finance, agents will compress the financial frictions of cryptocurrencies into each user's risk appetite through simplified natural language instructions.
In crypto gaming, agent workflows will enable personalized and ownable generative assets, more complex non-player characters, and chain-native experiences tailored for players.
In decentralized organizations, humans will agree on overall policy goals and constraints, and allow agents to execute across the various dimensions of business, protocol, or administrative functions.
These guiding roles and advantages have brought a revolutionary breakthrough from zero to one for all crypto applications. Millions of new users will join in this way, and no field can be immune.
AI, Cryptocurrencies, and Ownership
In understanding the grand trends before us, it is important to review and remember the lessons of our history. For most of human history, the ability to secure and defend resources meant survival itself. The modern property rights system is the product of millions of years of this evolutionary pressure. The concept of property rights is so crucial to human experience that it is enshrined in the US Constitution (the Fifth Amendment). The Founding Fathers of the US viewed property rights as the cornerstone of our system of governance and way of life.
Economists have long argued that strong property rights are the foundation of economic growth. They are crucial for individuals to safely create income, store wealth, and use these assets for credit and investment over time.
Multiple studies support this view. Research on over 100 countries from 1990 to 2002 showed that countries with stronger property rights grew faster than those with weaker property rights, in part because they were better able to promote technological progress and improve resource allocation.
From the perspective of property rights, blockchain is a core competitive technology. They are the most powerful technological foundation for global digital information, capable of providing immutable record-keeping, cryptographically secure ownership, distributed security, and programmatic execution of rights through smart contracts.
As the US enters the digital intelligence era, blockchain can also serve as the standard environment for all AI infrastructure and applications, ensuring that the US's AI systems can benefit from the structural support of the five unique features of cryptocurrencies. Historically, the US has created unprecedented opportunities, both for individuals and for the nation - from breaking free from colonial rule, to the constitutional promise of individual liberty, to the struggle against racial segregation, and the fierce market competition and entrepreneurial spirit.
Today, on the threshold of AGI, I believe the US has an important opportunity to further consolidate its leadership in these same dimensions. Aligning the US's AI policy goals with cryptocurrencies will unleash unprecedented individual participation in open-source networks, driving incentivized contributions across all layers of the AI stack. Broad participation in our AI market will stimulate competition, encourage new forms of bottom-up mobilization, and bring wider social impact.
