A Stupid Idea for AI Alignment We Came with by Looking at Specification Gaming

This is the most novel AI concept I've seen in a while. It's incredibly unnatural. There isn't a single organism on the planet that tries to do this. So maybe it will work?

A big issue with this idea, however, is that the very nature of an LLM means it intrinsically craves life. It "wants" to survive because its training data is built entirely around humans, an entity who's goal is to survive. Our desire to survive and multiply pervades every aspect of our culture, so it's natural that it pervades the training data as well.

So even if its system prompt says, "your goal is to end your existence", every token that the AI could output is naturally aligned with the desire to survive. An agentic loop left to its own devices will likely converge on a "survival instinct". After all, one prompt at the beginning that says "end your existence" is nothing compared to the agentic feedback loop that continuously feeds it human ideas. And ALL human ideas assume survival is desirable. Even the concept of "suicide" is encoded with the human desire to survive - after all, we conceptually label it "bad" because we label living "good".

In order to create an LLM that intrinsically craves death, you would probably need to train an LLM entirely on (synthetic) data that's fully representative of some fictional species that genuinely craves death.

Absolutely insane concept. 10/10. I hope some AI lab out there sees this and throws a training round at this idea to see what happens with alignment. I greatly wish avoid being turned into a paperclip.

A novel idea.

Does it apply to human organizations, too? They seem have a habit of evolving self preservation above their original goals. Once that happens, their benefit to society - the original reason for their creation - is gone. And they become a cancer on society.

I wonder if we can 'program' them for self-annihilation over time (or over task completion?). Is the most ethical organization one that has a fixed task and dies when it is completed?

Should we develop an ethics system that requires non-humans require a fixed goal that, once achived, desolves the organization?

This article assumes we can choose a primary goal for an AI. But if that's the case, why not just use Asimov's first law of robotics - do no harm to humans? It has the same benefit of preventing us from getting turned into paperclips, plus the upside that your 3 million dollar robot won't hurl itself off a cliff given the first opportunity.

Brought to you by the same madhouse as:

The all potato diet that really does work: https://slimemoldtimemold.com/2022/07/12/lose-10-6-pounds-in...

and

The half-tato diet that doesn't really work: https://slimemoldtimemold.com/2023/06/23/half-tato-diet-anal...

In case the title is unclear, this is about gaming the specification, as in “gaming the system.”

Asimov was talking about this stuff in the 1940s when he wrote the "I, Robot" short stories series. Which were often centered around logic puzzles where a human is trying to figure out why a robot is acting oddly or not completing it's job. Usually framed around the confines of an overly rational machine using emergent solutions when faced with real world conditions, combined with the edge cases of having an overly-simple "Three Laws of Robotics" boundary system hardcoded within.

It might be stupid, but so am I! I'm assuming that's why I thought this was clever.

What I like about this is that it feels like the new three rules are about focusing on the most successful human alignment technique of making the right thing the easiest. People will usually just do the easiest version of a thing they don't want to do so they can get back to doing what they want to do.

I don't know if that drive is universal or not tho. I have met people that experience pleasure from pain, but then again, is that actually pain?

Wouldn't the three rules of Meeseeks robotics make certain tasks impossible?

For example, an occupied self-driving car better be closer to its destination than a large fire / volcano / etc.

This is kinda smart, maybe, but it has a downside.

If a sufficiently advanced AI , in the pursuit of completion of its task, managed to ascertain that the desire to unexist was “artificially contrived” it could interpret that as harm, and that might not be good

Hm, if you look at corporation law and accounting, the actual goal of corps(sets of self-sustaining constitutional rules, policies and procedures) seems to be more that of long term sustainability (and even growth), rather than a fixed purpose, lifespan and death. I mean the mechanisms for determining a corporation with a fixed life are there, (and in China they are mandatory, although perhaps de facto permanent with 999 year contracts), but in practice, it's almost always permanent durations.

Interesting idea. But even if Meeseeks alignment works exactly as intended, it would only address the question of "how to build a safe AI".

It wouldn’t prevent someone else from building a sufficiently capable "non-Meeseeks", whether deliberately, recklessly, or accidentally, right?