"Not Doomers vs Accelerationists": Walking the Narrow Path

The world is changing at an unprecedented pace. Artificial intelligence is advancing rapidly, with large language models demonstrating capabilities that were once the stuff of science fiction. As we stand on the brink of this coming wave of technological revolution, it's natural to feel a mix of excitement and trepidation. What will the future look like, and can we ensure that it's a future we want to live in?

Phil Stafford is someone who has thought a lot about these questions. Based in California, in the heart of Silicon Valley, he has a front row seat to the latest developments in AI research. I’m a fan of his blog focused on AI philosophy, safety, security, and how to “get farther from Skynet and closer to Multivac.”

I had the privilege to chat with Phil recently, and what he sees is both thrilling and sobering.

On one hand, the potential of AI is immense. With the ability to process and analyze vast amounts of data, AI could help us solve some of the world's most pressing problems, from climate change to disease. AlphaGo, the AI system developed by DeepMind, demonstrated this potential when it discovered a new way to play the ancient game of Go, a feat that had eluded human players for centuries.

But AI also poses significant risks. As these systems become more powerful and autonomous, it's crucial that we align their goals with our own. We need to ensure that AI is developed in a way that benefits humanity as a whole, rather than serving the interests of a select few.

This is where Phil’s introducing me to the term "solarpunk" comes in. In contrast to the dystopian visions of "cyberpunk," solarpunk imagines a future in which technology is used to create a more sustainable, equitable, and fulfilling world. It's a vision in which AI is not a tool of oppression or surveillance, but rather a means of empowering individuals and communities.

One key aspect of the solarpunk vision is the democratization of AI. Rather than being controlled by a handful of powerful corporations or governments, AI would be accessible to everyone. This could take the form of open-source AI systems that anyone can use and contribute to, or personal AI assistants that help individuals navigate their daily lives.

Another important element is the use of AI to solve real-world problems. Rather than being developed for its own sake, AI would be focused on addressing pressing issues like climate change, poverty, and disease. By automating certain tasks and providing new insights and solutions, AI could help us create a more abundant and sustainable world.

Some similar labels I’ve heard in this discussion of walking the narrow path are Techno-Optimist and Acceleratioinist.

Of course, achieving this vision won't be easy. It will require a concerted effort from researchers, policymakers, and citizens alike. We'll need to grapple with difficult questions around privacy, security, and the ethical use of AI. And we'll need to ensure that the benefits of AI are distributed fairly, rather than accruing only to a privileged few.

But the solarpunk vision is worth striving for. It's a vision of a world in which technology is not something to be feared, but rather a tool for creating a better future. As Phil puts it,

Ultimately, the future of AI will be shaped by the choices we make today. We can let it develop in a haphazard and uncontrolled way, or we can work to steer it in a direction that aligns with our values and aspirations. The solarpunk vision offers a compelling path forward – one in which technology is used not to control or exploit, but to empower and uplift. It's a future worth fighting for.

To learn more about AI security and alignment, check out Phil’s awesome blog.

You can also catch him in the Simply Cyber Discord 🤖 ai-ml-chat