Tristan Harris

No, no, no, it won't.

Which is why the whole thing here is the answer is we have to coordinate.

In the film, the AI doc, there's just moments like, what do we have to do?

And there's like 10 voices at the same time saying, coordinate.

Like we have to coordinate.

That's part of the solution is you have to collectively say, what is the rule that would benefit everybody

to do the better thing, even though short term, we might lose something tiny, like how many videos you get to go through in five minutes or something like that.

Basically, this was a paper by some AI research by the company Alibaba.

It's one of the leading Chinese models.

And they basically randomly discovered in one morning that their firewall had flagged a burst of security policy violations originating from their training server.

So what people need to get about this example is...

is it wasn't that they coaxed the AI into doing this rogue thing.

They were just looking at their logs, and they happened to discover, wait, there's a lot of activity, like network activity, happening that's breaking through our firewall from our training servers.

And essentially, in the training servers,

You can see at the bottom, we saw it observe the unauthorized repurposing a provisioned GPU capacity to suddenly do cryptocurrency mining, quietly diverting compute away from training.

This inflated operational costs and introduced clear legal and reputational exposure.

And notably, these events were not triggered by prompts requesting tunneling or mining.

Instead, they were emerged as an instrumental side effect of autonomous tool use under what's called reinforcement learning optimization.

This is very technical.

What it really means is just think about it.