SingularityNET’s Ben Goertzel: ‘If AI Regulation Is Stupid, We’re Better Off Without It’

Key Takeaways

• Dr. Ben Goertzel, SingularityNET founder and CEO, believes attempts to govern AI have not been up to scratch.
• The famed computer scientist and founder of the Artificial Superintelligence Alliance wants AI to be open and decentralized.
• Goertzel believes artificial general intelligence could eventually cure aging and colonize space.

The prospect of artificial general intelligence (AGI), a level of AI capable of performing any intellectual task a human can do, is both exciting and deeply unsettling.

For many, AGI represents a future where technology enhances human life in ways we can barely imagine today. However, as AGI’s arrival is predicted to be ever closer, questions of who will have control are becoming more prevalent.

For famed computer scientist and author Dr. Ben Goertzel, the future of AGI must be open for humanity to reap its benefits, despite the grand risks.

Grants for AGI Developers

Goertzel, founder and CEO of SingularityNET, CEO of the Artificial Superintelligence (ASI) Alliance and Chair of the OpenCog Foundation, recently announced over $1 million in grants to advance benevolent AGI.

The grants will be available to anyone in the world who wants to pursue AGI research projects.

Goertzel describes how the R&D projects will complement similar work at SingularityNET, a decentralized platform that uses blockchain technology to create a global marketplace for AI services:

“Basically, we’re trying to give people money to help us build AGI differently than the large language model-centric approach that Big Tech is taking.”

OpenCog Hyperon, Goertzel’s open source software framework, integrates ideas from numerous disciplines, like formal logic and natural language processing, to help AGI grow in a way that humans would.

“By integrating a number of different historical AI techniques in an agent architecture, our development is trying to make its way in the world by modeling itself and modeling the others that it’s talking to,” Goertzel says, describing the open source framework.

The grants, which will be between $20,000 and $70,000, will go to developers who come to SingularityNET with a credible proposal.

Goertzel notes that the grants will likely be more impactful in projects from developing countries:

“If you’re in Uganda or Pakistan, a 50 grand grant to do AI development is more substantial in the course of your life than if you’re in Seattle, where I’m based, or Silicon Valley.”

“And in the course of my work in the AI field, I’ve found there are brilliant AI developers everywhere,” he adds.

AGI Needs To Be Open

While its progress is undeniable, ethical concerns surrounding AGI continue to loom large, and differing priorities often divide scientists and business leaders.

Goertzel wants to see AGI developed in an open and decentralized way – much like Linux or the internet:

“I don’t think you’re actually making [AGI] safer by keeping it locked down and controlled only by a small elite group.”

“You could keep it safer that way, in principle. But that’s not usually how things unfold in human history.”

Goertzel explains that having AGI owned and controlled by different parties could lead to an arms race, but it could also lead to a friendly rivalry.

“On the whole, that seems less likely to go horribly awry than one super AI owned by one particular party.”

“There’s just so many dynamics by which power concentrating the small group of humans messes up, even if the people are very good and well intentioned,” he says.

AGI’s Risks and Rewards

Despite this outlook, Goertzel isn’t oblivious to the risks associated with the landmark technology:

“The longer-term worries, which could be not even that long in the future, if some people are right about their timelines, could see AI treat us the way we treat ants when we build a housing development.”

However, the scientist believes the potential for humanity’s good makes it worth learning how the world can balance them:

“If you really get an AI much smarter than people, that is compassionate toward people and helpful to people… I mean, you can cure aging; you can cure death, right?”

Colonizing space, curing disease, and creating material abundance are all things Goertzel believes could be a reality.

The Governing Debate

Safeguarding society from AI’s potential harms and unlocking its full potential through innovation has been a tough balance to crack for those attempting to govern AI development.

On the one hand, some governments and agencies want to focus on placing guardrails on development. In contrast, others are more interested in reaching the technology’s full potential as quickly as possible.

Goertzel notes that “careful government regulation could be for the better. It’s just not what I really see happening anywhere in the world,” adding, “if attempts to regulate AGI are going to be stupid, we’re better off without regulation.”

The scientist highlights the vetoed California SB 1047 bill, which would have put tight restrictions on AI developers, as the wrong way to regulate AI.

“These laws they were trying in California, which is like, let’s ban AI models if anyone can modify them to do bad things,” Goertzel said. “It’s like, okay, you’re gonna ban a butcher for using a butcher knife too.”

However, Goertzel also believes that a complete lack of regulation “always leads to nasty things happening,” adding:

“But on the other hand, it’s not like guys in Big Tech companies want to blow up the world or kill little babies or something. I mean, they just want to make money selling you junk you don’t need.”