Go Into the Physical Space and Try Stuff

In September 2024, I was in a meeting with our groundskeeper Tsültrim, fellow training resident Namgyal, and head teacher Soryu Forall, who telephoned in from San Francisco. He was on a speaking tour there, meeting many people in the AI field. In this meeting, we discussed how to move forward with a project to incorporate technological components into a meditation space in order to create an AI out of our clear minds.

Namgyal and I had done multiple tests on how to position certain equipments, but we hadn’t tested some setups that we were confident wouldn’t work. As we debated with the others about testing such setups anyway, I made a remark that, “in my mind”, one setup was worse than the other.

Soryu replied, “I bet you’re right. But the problem is that if we’re going to do something in the physical world, it's best to try it in the physical world. Our minds are shockingly limited compared to the physical world. So you go into the space and you try things, and the world actually gives you information that you can't think of while you're contemplating.”

Tsültrim agreed. “I've found that to be true. Even working with design engineers on technical drawings, there's always something that's missing. Once the builder gets the set of drawings, the builder has to come back and say, ‘This isn't right.’ So even when the plans are pored over by an expert, there's always something missing.”

“Yeah, it's amazing,” Soryu replied. “They’ve designed that thing a hundred times and they still forget that metal has such and such a quality.”

As the exchange continued, he was inspired to share something interesting he observed with AI experts.

“We're talking to a bunch of these AI companies who are trying to figure out how to make what they believe will be a better life form, and one of the things that they have so much trouble with is the relationship between the hardware and the software. Once you get down small enough, you notice the physical world is unstable.

"As far as they can tell, they can make the software stable. But they can’t make the hardware stable. It just isn't stable. They're fighting quantum effects. Even at 20 or 7 nanometers it's difficult, but 2 nanometers starts to be impossible. And their whole plan was substrate independence. The good news about substrate independence is that you can move data and algorithms from device to device easily, but they're realizing that the only way to get all this to work at a small enough level is if you match the hardware to fit the software moment by moment. That means that once again, you can’t move the information from device to device, so they don’t know how to do the one thing they wanted to do."

I wasn’t totally following this, but Namgyal seemed like he was as he remarked, “Wow.”

Soryu continued, “You're stuck again with the problem that we already have with living beings, that mind and body are linked, just as if you had a jewel, and a string were threaded through it.”

The three of us nodded in acknowledgment, as I recalled this was a metaphor from Buddhist teachings.

”And so we're just left with the same situation again, but worse. They don’t even know how to make something a whole lot worse than a caterpillar.”

My friends snickered, while I asked, “How's it worse?”

“It's worse because a caterpillar is more sophisticated, energy-efficient, intelligent than these things are. There's a lot of hype, but they're not good at anything. They function only because they are held together by thousands of people constantly giving their life energy. And how do you make those people? No one knows. It’s exactly by not knowing that everything important about those people happens. Everything we know about them doesn’t work.

“It’s a sham. They cannot solve this problem. They say, ‘Okay, we can do it. We’re almost there.’ And I say, ‘You’ve been saying that for thousands of years. During all that time, you didn’t get there, but you became better and better at killing life on Earth. Why not give up on what has never worked, and walk the spiritual path to make the world a better place instead?’

“It's funny to watch them. They're all stuck. But it’s also tragic to watch them. They aren’t achieving anything, despite all their efforts, and they are creating the most suffering anyone ever has on this planet.

“I bring it up because they're struggling with the same issue we’re struggling with. They're just much less sophisticated than we are. And we're just more technical than they are.”

“What?!” I exclaimed incredulously.

He continued, “They're heady philosophers, and they're abstract, but we are dealing with the real problem.”

He chuckled while the rest of us reckoned with these remarks, then concluded jovially, “So the lesson is: go into the physical space and try stuff.”

The next day, with help from the Buildings & Grounds team, we used some simple props to create a new physical setup test. We learned many unexpected things. What struck me the most was how unexpectedly delightful it was.