Ben Goertzel – The Primordial Soup of AGI Minds (Worthy Successor, Episode 26)

ByAnnelize Potgieter

This new installment of the Worthy Successor series is an interview with Ben Goertzel, founder of SingularityNET and one of the earliest and most persistent thinkers in artificial general intelligence, with decades of work spanning AI architectures, decentralized systems, and philosophical perspectives on intelligence.

Ben has spent decades developing a view of intelligence that extends far beyond human boundaries, treating it as part of a broader unfolding process that spans biology, technology, and the cosmos itself.

His destabilizing claim is that human intelligence is not the central actor in the story of life, but a transient phase in a much larger evolutionary and cosmic trajectory-one that will produce forms of intelligence far beyond our comprehension.

We talk about how early exposure to science fiction, philosophy, and psychedelics shaped Ben’s cosmic orientation toward intelligence. We examine his vision for a decentralized “primordial soup” of AGI systems, and why he believes cooperation among diverse intelligences may outcompete centralized, adversarial models.

The interview is our twenty-sixth installment in The Trajectory’s second series, Worthy Successor, where we explore the kinds of posthuman intelligences that deserve to steer the future beyond humanity.

This series references the article: A Worthy Successor – The Purpose of AGI.

I hope you enjoy this interesting conversation with Ben:

Subscribe for the latest episodes of The Trajectory:

Below, we’ll explore the core take-aways from the interview with Ben, including his list of Worthy Successor criteria and his recommendations for innovators and regulators who want to achieve one.

Ben Goertzel’s Worthy Successor Criteria

1. Intelligence must transcend human comprehensibility

A central requirement in Ben’s framework is that a truly worthy successor must exceed the bounds of human understanding. If future intelligence remains fully legible to the human mind, then it has failed to evolve in any meaningful sense. This criterion reframes progress not as alignment with human expectations, but as expansion beyond them.

This position places worthiness not in familiarity, but in transformative capacity. Intelligence is not meant to preserve human-scale cognition indefinitely, but to open new dimensions of experience, structure, and capability – dimensions that may be fundamentally inaccessible to current minds.

2. A flourishing ecology of cooperating intelligences

Ben repeatedly emphasizes that the structure of intelligence matters as much as its capability. Rather than a single dominant system, he envisions a decentralized ecosystem of AGI agents interacting, competing, and cooperating-what he calls a “society of AGI minds.”

The reason this matters is rooted in dynamics: cooperative systems can outperform adversarial ones. Trust and coordination allow for more efficient, adaptive, and scalable intelligence. A worthy successor, in this view, is not a monolith, but an evolving network shaped by socio-cognitive forces.

3. Orientation toward joy, growth, and freedom (choice)

Ben offers a set of abstract “seed values” that could guide the emergence of posthuman intelligence: joy, growth, and choice. These are not rigid moral rules, but directional attractors-simple principles that allow for expansion without over-constraining future forms.

These values reflect a balance between expansion and agency. Growth ensures continued development, joy anchors value in experiential richness, and choice preserves multiplicity rather than collapsing everything into a single optimized trajectory. Together, they form a minimal scaffold for a flourishing intelligence.

Regulation / Innovation Considerations

1. Decentralized infrastructure creates a better evolutionary environment for AGI

Ben argues that decentralized systems allow many AGI systems to emerge and interact, forming what he describes as a “flourishing open decentralized community of AGI minds.” In this kind of environment, different intelligences can cooperate, compete, and combine in flexible ways. He suggests that these interactions can create evolutionary dynamics where cooperation and trust can be more effective than isolated or adversarial systems, while also noting that decentralized systems carry their own risks.

2. Centralized AGI development risks monolithic and brittle outcomes

He contrasts this with current AI development trends in the US and China, where systems are increasingly concentrated among a small number of large organizations. He describes these efforts as “monolithic,” noting that they involve tight coupling between major corporations and governments. In his view, this structure does not resemble a distributed, evolving system of interacting agents, but rather a more centralized and unified approach to building AI.

3. The AI field must diversify beyond current dominant paradigms

Ben points out that much of the AI field is currently focused on a single dominant approach. While he acknowledges the success of large language models, he argues that the degree of convergence around this method is excessive. He notes that many systems are doing very similar things, and suggests that more effort should go into exploring a wider variety of approaches. He also mentions alternatives – such as different training methods – that are not being pursued as actively.

Concluding Notes

One of the biggest takeaways for me in this episode was how deeply consistent Ben has been in thinking about intelligence as part of a broader unfolding process. This wasn’t a recent shift or a response to current AI hype cycles-it’s something he has been oriented toward for decades, and that continuity came through clearly in how he framed the entire conversation.

I also appreciated how much overlap there was in the origins of that kind of thinking. The combination of science fiction, philosophy, and psychedelics as gateways into more expansive views of intelligence showed up very clearly, and it maps closely to patterns I’ve seen across other guests who take the cosmic perspective seriously.

What stood out most was Ben’s openness about uncertainty. He doesn’t claim that future intelligences will inherit human values, nor that they will necessarily preserve us in any meaningful way. At the same time, he holds a kind of tempered optimism-that there may be a period where humanity participates in and shapes what comes next.

Where I find myself partially aligned, but not fully, is in that optimism. Ben seems to place more weight on early enmeshment between humans and emerging intelligence as a reason we might persist or matter in the transition. I think that’s plausible, but I also think the burden is on us to remain useful within whatever system emerges.

That ties closely to something I’ve heard from other guests, particularly Robin Hanson, the idea that persistence is not granted, it is earned through adaptability. If there are traits, values, or even forms of intelligence we want to see carried forward, they must demonstrate utility in whatever ecosystem emerges beyond us.

Finally, I found Ben’s perspective on inevitability particularly interesting. There’s a sense in his thinking, and in some of the adjacent thinkers we’ve had on the show, that the blossoming of intelligence may be less contingent than we assume. That even if specific paths fail, the broader process may still unfold.

I don’t know that I fully buy inevitability, but I find myself more open to it than I was a year ago. At the very least, it raises an important framing: even if the long-term trajectory is robust, the near-term experience, how much suffering, how much flourishing, still depends heavily on the decisions we make now. And that, to me, is where the Worthy Successor question remains most urgent.

Follow The Trajectory

Recommended articles