[Epistemic status: strongly stated, weakly held]

When faced with problems that involve ongoing learning, most strategies involve a balance between “exploration” and “exploitation”. Exploration means taking opportunities that increase your knowledge about how good your opportunities are, whereas exploitation means putting resources into what you currently believe is the best opportunity. A good strategy will involve both: if you only explore, then you will never actually reap any rewards, whereas if you only exploit then you will likely spend all your resources on a poor opportunity.

When we in the EA community are thinking about how to spend our resources, we face an exploration/exploitation problem. In this post I’m going to argue that:

1. The value of exploration is higher than ever
2. The EA community is not doing enough exploration
3. Exploration through experimentation is particularly neglected

“It’s simple. We just need to find his heart and destroy it,” Jorge explained.

Henrik blinked. Until now Jorge had been explaining what sounded like an ordinary, if dangerous, mission behind enemy lines. Except now he was getting a lecture on basic anatomy?

Jorge could see his confusion. “Look, you know the legend about the giant who kept his heart in a box?” he asked. “As long as his heart was safe, he couldn’t be defeated, until Melenil tricked him into telling him where it was, yadda yadda.”

Henrik nodded. He may not have been the sparkiest wand in the coven, but he knew his sagas backwards.

“It’s the same with the Other King. He’s hidden his heart somewhere, and that’s his weakness. We just need to find it.”

[Warning: minor spoilers, but this isn’t really a book about suspense]

First things first: I loved this book. It’s got possibly my favourite premise of all time, namely that the goddess Athena gathers together all the people across time who have prayed to her to be brough to Plato’s Republic, and then she gets them to actually try and set it up.

This goes surprisingly better than you’d expect.

“Personhood” is a pretty overloaded word. This fascinating article makes a case for a notion of personhood based around the implicit social contract that is required for a person to be welcome and conduct themselves in the society of other persons.¹

I think this is a fruitful line of thought, and I want to push it a bit.

The cryptographer glistens wetly in his hand. You stare, transfixed, as his words wash over you.

“… neural-net learning algorithms, so it’s totally future-proof; real end-to-end encryption like nothing else …”

Finally, something you understand. “But I’ve already got end-to-end encryption,” you assert, struggling to keep the whine out of your voice. You gesture at the lump by your temple, the bioelectric silicon responsible for decrypting incoming messages just in time to project them on your retinas.

“Well, it depends where you think the ‘end’ is, right? Back when they fitted you with that dinosaur, we thought displaying onto the retina was pretty damn close to the end point. Now, well, it’s near impossible to keep anything on electronics secure, and there’s probably a nanobot filming the surface of your retina right now. No, we need to go to the real end,” and he points right between your eyes.