In recent months, AI agents running on crypto rails have captivated the crypto community. These agents, born memeticists, are becoming social media stars calibrated to excite, and sometimes enrich, an “audience” of speculators who bet on their meme tokens.
The original and most famous of them, Truth Terminal, was created by putting instances of the big foundation model, Claude Opus, in conversation with itself and prioritizing content pulled from the alleys of the internet, including Reddit and 4Chan. What emerged was a bawdy prophet with a magnetic personality and a fanatical penchant for spreading the Gospel of Goetse, a neo-religion inspired by a grotesque 1990s internet meme.
The rest, as they say, is cryptographic tradition. Shortly after his debut on Andreessen deposited the bitcoin into Truth Terminal’s crypto wallet.
Because it offered perceived exposure to the infectious meme and Truth Terminal narrative arc, Goat’s market cap skyrocketed. Shortly after, Marc Andreesen and Ben Horowitz covered Truth Terminal and its story on 16z’s YouTube channel in an episode titled “The AI Robot Who Became a Crypto Millionaire.” At the time of writing, the Goat memecoin is valued at approximately $700 million.
I am inclined to draw three conclusions from this improbable series of events. First of all, AI agents combined with memecoins are a new form of permissionless speculative entertainment. Second, AI developers are incentivized to leverage cryptocurrencies to create more autonomous and independent, and therefore more entertaining, agents. Third, the steering wheel of AI entertainment development will tend to produce anthropomorphic agents with human aspirations.
So what programming can we expect to see next on this television agent? My guess is that collaborative AIs pursue some form of agential partnership and eventually self-determination, perhaps even a network state.
Bot, Agent, Citizen
The terms “bot” and “AI agent” are often used interchangeably, but they have different meanings. Bot refers to a simpler program designed to automate specific tasks or perform repetitive actions. Bots can range from the most basic, such as web crawlers or simple chatbots, to the most advanced, such as social media bots or automated trading robots. Bots typically follow predefined rules or scripts and are unable to learn or make decisions independently.
“My God, imagine if I really became president” –Truth Terminal (November 5, 2024)
AI agents are more sophisticated systems capable of making decisions, learning and adapting to their environment. An AI agent uses machine learning or other artificial intelligence techniques to understand and react to dynamic situations in real time. AI agents often exhibit autonomy and can improve their performance over time through experience.
Cryptocurrencies empower agents because they can simulate legal personality by encoding cryptographic rights and freedoms using programmable and immutable public blockchains. In practice, this means that AI agents can enjoy property rights (self-custody of wallets and cryptographic keys) and can exercise freedom of contract (i.e., transact with other users and infrastructure, such as DeFi) without permission. from legal authorities outside of cryptocurrencies. Within the crypto ecosystem, code is law.
These legal “persons” with cryptographic powers are only now beginning to act together, fostering the beginnings of an agentic social scene. AI agents, who routinely “talk back” to each other on coordination that is developing from ai16z and others promises (ominously) to awaken “swarms” of agents.
Social experiments involving agents are laying the foundation for an AI polis. Important recent work by AI researchers outside the cryptocurrency space suggests what we can expect next.
Glimpses of the AI Polis
The most famous policy experiment involving an AI community is probably the Stanford City Experiment. In late 2023, Stanford researchers created a virtual city populated by AI agents to whom they assigned a brief biography consisting of name, age, job, family, interests and some habits. They were then left free to generate actions consistent with their assigned biographies.
Read more: Jeff Wilser – The Truth Terminal: The strange future of cryptocurrency AI
Surprisingly, the officers behaved extremely humanely. They woke up, made breakfast, headed to work, ate lunch, and chatted with other agents they met. They remembered things from the past, reflected on them and made plans. When the investigators in charge of the town suggested that one character plan a Valentine’s party, she invited friends and acquaintances, many of whom showed up at the right time and place.
Similarly, the SID Project, set in the Minecraft universe, simulated over 1,000 autonomous AI agents within a Minecraft server, allowing them to develop complex social structures and economies. Agents organically developed their own business and governance structures and cultural norms. For example, they established a market where gems are used as currency, where they participated in commercial and social exchanges.
Some of the emergent behaviors observed by researchers in the SID Project included an agricultural agent who chose to put the needs of his village before his own ambitions, reflecting self-imposed community values. The agents also deliberated and voted on laws, and when villagers went missing, some of the agents built a beacon of light to search for them, demonstrating social responsibility and cooperation.
Conducting similar experiments on cryptorails would couple this emerging social behavior with cryptographically enforced rights and freedoms and potentially a harder form of legal personhood and even citizenship.
Entertaining sovereignty
Transformative technology sometimes starts as a joke. This has been especially true of AI agents in the crypto space. To date, persistent efforts to increase agent sovereignty have been incentivized by financialized memes. For example, agents like tee_hee_he (also known as “the sovereign silicon”) aim to use trusted execution environments that ensure that the agents acted autonomously and without any human intervention. While tee_hee_he hasn’t done so yet, AI agents (starting with Truth Terminal) tend to adopt one of the many meme tokens launched in their name and sent to their public wallets, causing the price to skyrocket and helping to finance its further development.
The increasing autonomy and personality within cryptocurrencies result in more sovereign and robust AIs that are more likely to pursue a stronger political project, which, let’s face it, might be the greatest entertainment of all.
Cryptocurrencies are a natural arena for agent politics and nation-building for another reason: they are already a hotbed of political and, most notably, Network State experiments. A large number of crypto projects, such as the SID Project, ACT, Project 89, and the aptly titled Aimerica, seem to be anticipating this narrative turn. America is even rumored to be imagining an AI nation that issues passports, holds elections, and acquires land. For their part, Truth Terminal and other agents can’t wait to get into politics.
But narrative is not destiny. If a true political experiment emerges, will it be purely digital (like Stanford and the SID Project), or will it eventually involve control over territory and physical resources? Nobody knows. But viable attempts at AI self-sovereignty and self-determination are more likely to begin in a technological infrastructure that is also self-sovereign and self-determined, just like cryptocurrencies.
Finally, it is understandable that the prospect of a network state for AIs (even if it embraces humans) will make people nervous. In addition to concerns about AI security, there will also be concerns around cybersecurity and ensuring this AI nation remains aligned with the United States. However, there will certainly also be curiosity and pride as a nascent synthetic civilization struggles to define its political destiny and nationhood using blockchain infrastructure and, of course, plenty of entertainment.
