> > Using Google during rounds is technically cheating - I’m unsure about visiting domains you find during the rounds though. It certainly violates the spirit of the game, but it also shows the models are smart enough to use whatever information they can to win

Going off of the source article here, the author at least wasn't clear on whether the rules only prevent using google or if visiting any website is against the rules.

And either way, my point was that the person defining the rules to the LLM was ambiguous. The potential risk of misalignment isn't that a perfect set of rules can't be defined, its that the rules we do define will always be incomplete.

o3 already is an external source of information. It's an online service backed by an enormous model generated from an even more enormous corpus of text via an enormous amount of computing power.

o3 was the thing beating GeoGuessr. It isn't external to itself.

Sure, then o3 plus the World Wide Web was the thing playing the game, and also isn’t external to itself.

Right, and that’s indeed impressive! But still not what’s claimed in the headline.

That's debatable, given that searching the web is a standard feature of o3.

Fair enough, but searching the web is also a standard feature of humans, but explicitly prohibited when playing GG.

The GeoGuessr Community Rules and Terms and Service strongly imply that users must be people, so we are already conceding that exception to the rules when we want computer systems to compete.

I believed the rules were not explained to the model so it does use what it can.

Then you can 100% not claim it is “Playing” the game

That right there *is the alignment problem*.

If I task an AI with "peace on earth" and the solution the AI comes up with is ripped from The X-Files* and it kills everyone, it isn't good enough to say "that's cheating" or "that's not what I meant".

* https://en.wikipedia.org/wiki/Je_Souhaite

That's the alignment problem. We intended a certain set of rules but didn't define them completely, or there were conditions we didn't consider.

An AI wouldn't have to maliciously break a rule to go wrong. The point is that the system could so exactly what it was supposed to do, it plays within the given rules but the outcomes aren't what we wanted or intended.

It's playing a game in which the rules are a bit ambiguous if not explained.

And in reality the set of rules we would need can never be fully explained.

Alignment is the goal of having an AI system understand what we would want it to do even when the rules weren't predefined. That's an impossible task, or rather its seemingly impossible and we don't yet know how to do it.