top of page

From Memes to Machines: The Birth of AI Agents

Updated: 3 hours ago


Intro Disclaimer: Any views expressed in this blog are the personal views of the author and should not form the basis for making investment decisions, nor be construed as a recommendation or advice to engage in investment transactions. The author of this content may hold personal investments in the assets, securities, or instruments mentioned within this material. This ownership could potentially create a conflict of interest, as the author's opinions may be influenced by their investment position. Readers should not interpret this information as financial advice and are encouraged to conduct their own research. Investing in cryptocurrencies involves significant risk and may not be suitable for all investors. The value of cryptocurrencies is highly volatile and can fluctuate widely in response to market dynamics, regulatory actions, technological developments, and other external factors.


Background

The overarching theme of on-chain trading activity in 2024 was none other than.... (drum roll please) memecoins. Over 5 million new tokens were launched on Solana's pump.fun alone, generating ~4 billion in total revenue fees. Yes, you read that right. The trends and memes might be rudimentary in nature, but there's no denying the sheer volume, user adoption and activity.


Blockchain users are at the forefront of a digital world built across a constantly evolving internet, driving billions of value into the world's new attention economy, best described as an ecosystem of internet trends, themes and assets whose value can be driven by their level of public awareness. Today, we can quantify this awareness in the form of mindshare, or how much public dialogue a specific topic or theme commands as a % of total public consciousness. In simpler terms, if a new development is trending, it has some degree of public mindshare.


As an investor, navigating the attention economy can be an incredibly arduous task because the music is constantly changing its tune. One second, a meme is in the public spotlight, the next, it's a ghost of the past. The fund flows typically follow, creating fierce competition between traders on their paths towards an exit.


If you're curious how the representation of a meme can be tokenized into a tradable asset, please see the below learning point on digital ownership through tokenization at the bottom of the article titled, "Learning Reference #1"


So, How Did We Get Here?

The obvious precursor to blockchain's hundred-billion-dollar meme economy stems from the NFT craze of 2021, where copies of individualized jpegs and images were digitized as tokens and sold in the open market, resulting in hyper inflated valuations upwards of billions in total capitalization.


Following the collapse of FTX and outside a few globally recognized projects such as Pudgy Penguins or Bored Apes, NFTs never truly recovered. Most projects incurred 90%+ draw downs and saw their founding teams jump ship.


Fast forward to the first half of 2024, and Memecoins would ride the next wave of the attention economy, becoming one of the highest performing crypto sub-verticals. According to Kaito, memecoins led and commanded ~25% of crypto's total mindshare in Q1 of 2024 and as of year-end, 22 of the top 200 cryptocurrencies by market value are memecoins.


But yet again, the tides are changing and there seems to be a new sheriff in town. Artificial intelligence currently commands ~57% mindshare of the entire crypto ecosystem, dominating the next leading categories of Memes (~8%) and DeFi (~6%).



The Rotation from Memes to Agents

The birth of the first autonomous memecoin was presumably led by Andy Ayrey and his pair of Claude Open AI large language models, and ultimately his twitter bot known as Truth Terminal, which would lead to the conceptual invention of "GOATSE OF GNOSIS". And soon, AI memecoin Goatseus Maximus ("GOAT") was born.


Truth Terminal and GOAT would become one of the first AI agents with its own corresponding crypto tokens, ultimately surpassing $1.3 billion in market value this past November.


Inspired by GOAT, various twitter agents would spawn their own corresponding meme tokens, and soon, the era of tokenized bots would arise.


The AI Agent Landscape

A) AI Agents

What on earth is an AI agent?

A little rebranding never hurt anybody; AI agents in their current state are essentially just multi-feature chat bots. Agents can communicate via text, speech, and 3D animation and can interact in various environments, such as TikTok, Telegram, Discord, X, Roblox, and more.


Autonomous agents can:

  • Speak and move in 3D spaces

  • Learn, plan and make decisions

  • Interact within an environment

  • Make on-chain transactions with their own wallet


Why should they exist on blockchains?

1) Blockchains offer AI agents a decentralized and secure network to autonomously perform tasks without the need for intermediaries by executing smart contracts, managing micropayments, and securely handling transactions all on their own.


2) Agents are capable of generating their own revenue, and can be tokenized and co-owned on a blockchain. Democratizing ownership is key, allowing anyone to capture ownership exposure to an agent, that once deployed, can generate revenue through user interaction.

(Source: vvaifu)


Illustrative Agents

  • $Zerebro: Multi-platform content agent that can manage specific actions such as posting on X, generating images, and minting artwork (on Polygon)

  • $AIXBT: X bot that autonomously tracks discussions and identifies momentum-based investment opportunities; token holders gain access to the agent's analytics platform

  • $LUNA: Voice-activated X bot influencer, with its own live stream on Youtube

  • $VaderAI: X bot that aims to become the Blackrock of agents by creating investment DAOs that will identify and invest in emerging tokens of decentralized agents

  • $Aimonica: Meme focused AI investment bot with the goal of outperforming Web3 VCs in memes investing


B) AI Launchpads

Where can I launch an AI Agent?

On a launchpad of course! Launchpads act as a single-hub environment for creators to launch an AI agent and for consumers to select which agent they'd like to co-own.


Illustrative Launchpads

  • Virtuals: A Base-native launchpad and current market leader; home to agent leaders such as $AIXBT, $GAME, $LUNA, $CONVO, and $VADERAI

  • vvaifu.fun: A Solana-native launchpad; home to $AIMONICA

  • Top Hat: A Solana-native launchpad


C) AI Frameworks

All great agents need an operating system, which act as building blocks for the creation of AI agents. These operating systems are generally referred to as frameworks and can be defined as developer tool kits and software platforms for deploying and managing autonomous AI agents.


Frameworks such as AI16z's, "Eliza", can support inference through open-source models, cloud-based inference through OpenAI, and even integrate with advanced models such as Nous Hermes Llama and Claude. Unlike general-purpose machine learning frameworks, Eliza focuses on pre-trained models for agent-based interactions (Source Eliza)


Illustrative Frameworks

  • Eliza by AI16z (: Built with TypeScript, Eliza is the largest framework by market share with over 7.1k+ stars, 2k+ forks, and 220 contributors; excels at multi-agent management and application integration, such as Discord, X, or Telegram

  • G.A.M.E by Virtuals Protocol: Language-agnostic, API driven, and built for gaming and metaverse agents; G.A.M.E is the 2nd largest agent token on Virtuals

  • Rig by ARC: Built with Rust, the native programming language of Solana, Rig is a performance-focused framework built for enterprise-grade applications

  • ZerePy by Zerebro: Built with Python, ZerePy is an artist-oriented framework with a focus on NFTs, music, and digital art

Source: arndxt


I'd like to highlight that there are other notable AI agents, launchpads, and frameworks that were not listed above as the AI Agent ecosystem is rapidly evolving and changing in real time.


If you have questions about the above blog, the AI agent landscape, or cryptocurrencies in general, please do not hesitate to reach out!


Learning Reference #1

In the form of unique (Non-Fungible) or interchangeable (Fungible) tokens, blockchains can itemize trends using tokens, establish digital property rights, and leverage native decentralized finance software applications to support trading and liquidity.


Fungible tokens include commonly known cryptocurrencies, where no one token is different than that of its counterpart. For example, 1 Bitcoin = 1 Bitcoin, or 1 Dogecoin = 1 Dogecoin. The two are purely interchangeable.


Non-Fungible tokens on the other hand are one-of-a-kind, and commonly used for art, collectables, or tickets, where each token is an independent version of its counterpart.



Boomer Saraga | Crypto Investment Manager 


 

 

 


Closing Disclaimer: This blog is for informational purposes only. While the information provided herein is believed to be accurate and reliable, none of Khelp, or any of their respective affiliates or representatives or any other person makes any representations or warranties, express or implied, as to the accuracy or completeness of such information.


Comments


bottom of page