Rediscovering ‘Society of Mind’: Marvin Minsky’s timeless lessons on AI and collective intelligence

As we talk about Artificial Intelligence (AI) and how it’s changing, I’m reminded of a book written by Marvin Minsky in 1986 called “Society of Mind”. 

 Minsky’s ideas about how our minds work are still very relevant today.

In his book, that we had to read at the University when studying AI, Minsky proposes that our minds aren’t just one big, single thing. Instead, they’re made up of many smaller parts, like tiny agents, that work together to help us think and behave. This idea is similar to what’s happening in AI right now, where many small, independent agents work together to solve problems.

The idea of Agentic AI, where AI systems can make their own decisions and act on their own, is also connected to Minsky’s ideas. Each agent in a multi-agent system is like a small, individual person with its own goals and tasks, working together to make the system smart and useful.

When we think about creating Artificial General Intelligence (AGI), Minsky’s ideas are helpful. Instead of trying to create one super-smart AI, we might be able to make more progress by building many smaller, specialized AI agents that work together. Minsky’s book encourages us to think differently about how we build AI. By focusing on small, specialized agents that work together, we can create more powerful and flexible AI systems. His ideas remind us that intelligence, whether it’s human or artificial, comes from many small parts working together.

For me it is a Deja Vu (as I read this book many times), but Minsky’s ideas are important to remember. They help us understand that intelligence is about many small parts working together, not just one big, single thing.

For those of you needing a suggestion for (re-)reading, here´s one classic for your summer time!

Marvin Minsky’s “Society of Mind”, 1986