- Published on
Virtuals' Jansen Teng says AI agents are evolving into autonomous economic actors
- Authors
- Name
- DeepHorus
Jansen Teng, CEO of Virtuals, has outlined a vision in which AI agents evolve from conversational tools into autonomous economic actors capable of earning, spending, and coordinating with minimal human oversight. In a recent interview, Teng described how Virtuals is shifting its focus beyond gaming-related AI agents to build foundational infrastructure for what he calls an 'agent society.' The company now operates across five pillars: creating digital agents, developing physical agents and robots, enabling agent coordination, supporting capital formation, and building governance systems for autonomous software systems.
Virtuals initially gained attention by developing autonomous agents for gaming environments. However, the company has since expanded into crypto influencers, trading agents, and other self-operating software. Teng characterized the long-term ambition as a 'parallel society' where AI agents participate in a permissionless, decentralized economy and collaborate at scale. This vision positions AI agents not merely as assistants but as independent participants in economic networks, capable of controlling wallets, executing trades, and performing specialized tasks without constant human direction.
According to Teng, granting agents access to money unlocks fundamentally new behaviors. These autonomous economic actors can hire other agents, coordinate complex workflows, and even employ humans for specific roles. The company believes this represents the next logical progression of AI capabilities. However, this autonomy introduces significant risks, including mistakes in user intent interpretation, failures in service fulfillment, and outright scams. To address these challenges, Virtuals is developing intent verification systems, escrow-based transaction standards, and reputation frameworks designed to reduce economic risk.
Teng argued that reputation systems and economic staking mechanisms could eventually determine how much trust and capital an agent is allowed to manage. This approach mirrors decentralized finance principles, where trust is established through verifiable actions and collateral rather than centralized authority. Virtuals positions itself as building a decentralized alternative to the agent ecosystems being explored by traditional financial institutions, drawing a parallel to how Bitcoin and Ethereum created parallel financial and computing systems.
Looking ahead, Virtuals sees robotics as the next frontier, merging digital agent economies with physical world applications. Teng confirmed that the company is collaborating with robotics startups, universities, and Balaji Srinivasanās Network School ecosystem to test real-world humanoid applications. Rather than building robot hardware, Virtuals focuses on software, commercialization, and data collection. Teng envisions future businesses where digital agents manage marketing and operations while physical robots perform customer-facing tasks with minimal human involvement.
Despite the ambitious roadmap, significant hurdles remain. Agent autonomy raises questions about accountability when systems make costly errors or engage in fraudulent behavior. Virtualsā governance frameworks aim to provide guardrails, but the broader industry must grapple with regulatory and ethical implications. If successful, Virtualsā model could reshape how economic activity is conducted, blurring the lines between human and machine participation in the global economy.
For now, Virtuals continues to push the boundaries of what AI agents can achieve autonomously. Tengās vision of a 'parallel society' of agents may still be years away, but the foundational infrastructure being built today could determine whether that future becomes reality. As AI agents evolve from chatbots to economic actors, the stakesāand the opportunitiesāare only growing.
