Concept

Continuing on ideating λC, i like the concept of self-funding swarm intelligence - that is collaborative, financially incentivized autonomous agents, working to various degrees of automations. The aim would be to gradually build towards type 1→4, with swarm types being loosely defined below:

• Automation as a Service (type 1) - agents that automate predefined tasks and generate money when the requested services are fullfilled. i.e. agents that research a given topic and output a pdf, available via service subscription.

• Human-AI Collaboration (type 2) - feedback loops with autonomous agents interjected. i.e. junior dev using GitHub Copilot to ship company code

• Partial Human Supervision (type 3) - swarm agents automating the lifecycle of a given task, but requiring human evaluation/confirmation to move between steps. i.e. swarm researching & writing a film script, generating audio+video, human validates and edits outputs of different phases

• Full Autonomy (type 4) - self-sustaining swarm that continuously works, monitors, improves and interacts with real world markets, generates gains without supervision. a close parallel to algorithmic trading, but beyond financial markets. small-scale AGI.

Related Dimensions

• Building machine-first infrastructure & frameworks for autonomous agents to take action in the real world with minimal human supervision.

• AI agents operational capabilities - ability to self-organize, model execution strategies, evaluate pertinent metrics in near real time.

• Blockchain & crypto for (1) Governance (DAO), within the swarm and for human-swarm interactions as well (2) fund management, both inter-operational (swarms & separate swarm groups exchanging money) and external (investment funds, fiat←→crypto to interact with the external world, non-crypto banking) (interesting : see capital.xyz)

Timing

As generative AI regulation gets progressively tight, doing live tests of your prototypes on the market is likely to get more complex. As far as innovating, market-testing and iterating fast on autonomous agents, the best time is likely right now.

Conclusion

These are the general ideas that came to mind. I'll come back and update this articles with more in-depth perspectives. Feel free to share any interesting perspective @n_raidenai