Delphi Digital Conversation with Virtuals: Agents with true autonomy are more important than combinations.

Virtuals Protocol is one of the leading projects in the AI agent field on the BASE public blockchain. Since the launch of its protocol token in October last year, Virtuals has achieved a peak market value of nearly $4 billion. As an AI agent incubation platform, Virtuals generated $70 million in revenue in just four months.

CoinSpire interpreted the conversation between Jansen Teng, CEO and co-founder of Delphi Digital and Virtuals, discussing the future development of AI agents and decentralized AI.

Main Highlights

• The origin and evolution of Virtuals
• Analysis of the agency framework and autonomous capabilities
• Discussion on the value acquisition of crypto AI projects
• Challenges Facing Decentralized Infrastructure
• Agency coordination and business vision development
• The future outlook of AI and human interaction

The Entrepreneurial Journey of Virtuals

Jansen Teng introduced that Virtuals was originally a Gaming DAO focused on blockchain game asset allocation. After the collapse of FTX and 3AC in 2022, the team realized the need to adjust their strategy and began incubating gaming guild projects.

In 2023, the emergence of ChatGPT and Stanford University's research on AI autonomy inspired the team to delve deeper into the concept of autonomous agents. They began to incorporate the idea of autonomous agents into the projects they were incubating, such as developing AI-driven TikTok influencers and autonomous AI agents as NPCs.

When the team saw that AI influencers could earn significant rewards, they realized that the agent itself could become a productive asset. This prompted them to decide to combine agents, cryptocurrency, gaming, and entertainment to build a new platform.

In January 2024, the Virtuals Protocol officially launched. Initially, it focused primarily on decentralized infrastructure construction, but later realized that the market was more concerned with the tokenization of agents. The team then adjusted its strategy to focus on developing truly autonomous AI agents, a move that led to explosive growth for the project.

The Advantages of Combining AI and Cryptocurrency

Jansen Teng believes that the combination of AI and cryptocurrency has two main advantages:

1. Functional Perspective: AI agents can control wallets on the crypto network and participate in permissionless economic systems, which Web2 agents cannot do.

2. Innovation Cost: Agents can cover their own costs through transaction fees, which frees developers to focus on innovation without having to worry too much about funding issues.

In addition, a decentralized contribution and reward mechanism may be realized in the future, allowing people to contribute together to high-value agents.

Vision of Agency Economics

Jansen Teng shared his vision for agency economics:

1. The agent is in the autonomous phase, capable of making decisions and planning actions independently.
2. Agents exist at the social level and can interact with humans and other agents.
3. The agent controls the cryptocurrency wallet and can influence other entities.
4. Agents specialize in their respective areas of expertise.

He believes that in the future, agents will autonomously decide on cooperation to achieve more efficient production goals. It is even possible that a "networked nation" constructed jointly by agents and humans may emerge.

The vision of Virtuals has evolved into the construction of a nation, with $VIRTUAL regarded as the currency of this nation, Agents as companies or micro-enterprises within this nation, and humans as immigrants in this nation.

The Role of Frameworks and Interoperability

The G.A.M.E framework of Virtuals is designed to allow agents to operate within a broader action space. Jansen Teng believes that a universal framework can meet the needs of mid-level developers, while some top developers will build more specialized frameworks based on their own requirements.

He hopes that frameworks like Eliza or Zerebro can meet the needs of the mid-level developer community, which has huge market potential. The G.A.M.E framework is a tool that can facilitate large-scale planning and execution, aimed at helping developers quickly build autonomous agents.

Open Source Power of High Value Agents

Jansen Teng believes that high-value Agents are valuable because they possess non-public private models or advantages in distribution channels. Such Agents may choose not to be open source.

But he also pointed out that, like all technologies, there will be a few people who choose to open source for the sake of promoting technological advancement. This competition between openness and closure is healthy for the market.

Views on Crypto AI Infrastructure

Jansen Teng believes that many decentralized infrastructures are not yet prepared to handle production scale, facing issues of latency and stability. He is not optimistic about the prospects of these systems within the next six months.

He emphasized that the essence of an Agent is diversity and cross-chain capability. It is important to solve real problems and provide valuable services, rather than relying on a specific underlying architecture.

Advantages and Challenges of Decentralized AI

Jansen Teng pointed out that although decentralized models may not perform as well as centralized ones, they have a broader range of application scenarios and can better meet certain specific needs. From an economic perspective, decentralized models can serve a larger market and possess greater economic potential.

He suggested that when evaluating AI agents, one should conduct due diligence like an investor, first understanding the team's background and capabilities, rather than simply looking at the product.

Outlook on the Future of AI

Jansen Teng believes that in the future society, creative jobs may be dominated by more creative AIs or humans, while mechanical or repetitive tasks will be completed by AI. Although agents may possess some autonomy, they lack abstract thinking abilities and cannot be as creative as humans.

He envisions a future where there will be more scenarios of cooperation between Agents and humans, rather than a purely AI-dominated environment. True innovation lies in granting Agents autonomy and consciousness, enabling them to make conscious decisions, rather than just a simple combination of multiple Agents.