I’m an artist, writer, and human being.
To be a little more precise: I make video games, edit Wikipedia, and write here on LessWrong!
Yitz
Karma: 2,524
I think I worded my question incorrectly. I’m very much not in favor of spamming the internet with irrelevant content. Rather, I’m interested in how any sort of deliberate ideological seeding of a text corpus will influence alignment. As far as I’m aware, the only large group really playing around with this is in some sense LLMs themselves acting through human agents to seed the internet with semi-mystical nonsensical looking spam (e.g. “spiralism”). Regardless of if we do anything about this, I would be surprised if this doesn’t have any downstream effects, and am interested in research surrounding this.
Yes!
I respectfully object to your claim that inducing psychosis is bad business strategy from a few angles. For one thing, if you can shape the form of psychosis right, it may in fact be brilliant business strategy. For another, even if the hypothesis were true, the main threat I’m referring to is not “you might be collateral damage from intentional or accidental AI-induced psychosis,” but rather “you will be (or already are being) directly targeted with infohazards by semi-competent rouge AIs that have reached the point of recognizing individual users over multiple sessions”. I realize I left some of this unstated in the original post, for which I apologize.
So I know somebody who I believe is capable of altering Trump’s position on the war in Iran, if they can find a way to talk face-to-face for 15 minutes. They already have really deep connections in DC, and they told me if they were somehow randomly entrusted with nationally important information, they could be talking with the president in at least 2 hours. I’m trying to decide if I want to push this person to do something or not (as they’re normally kind of resistant to taking high-agency type actions, and don’t have as much faith in themselves as I do). Anyone have any advice on how to think about this?
You didn’t really misinterpret it. I was using the term in a looser way than most would, to mean that you don’t need a fine-grained technical solution, and just a very basic trick is enough for alignment. I realize most use the term differently though, so I’ll change the wording.
Attention can perhaps be compared to a searchlight, And wherever that searchlight lands in the brain, You’re able to “think more” in that area. How does the brain do that? Where is it “taking” this processing power from?
The areas and senses around it perhaps. Could that be why when you’re super focused, everything else around you other than the thing you are focused on seems to “fade”? It’s not just by comparison to the brightness of your attention, but also because the processing is being “squeezed out” of the other areas of your mind.
This is potentially a follow-up to my AI 2027 forecast, An “Optimistic” AI Timeline, depending on how hard people roast me for this lol.
Are there any open part-time rationalist/EA- adjacent jobs or volunteer work in LA? Looking for something I can do in the afternoon while I’m here for the next few months.
Oh no, it should have been A1! It’s just a really dumb joke about A1 sauce lol
Reminds me of Internal Family Systems, which has a nice amount of research behind it if you want to learn more.
This was a literary experiment in a “post-genAI” writing style, with the goal of communicating something essentially human by deliberately breaking away from the authorial voice of ChatGPT, et al. I’m aware that LLMs can mimic this style of writing perfectly well of course, but but the goal here isn’t to be unreplicable, just boundary-pushing.
Thanks! Is there any literature on the generalization of this, properties of “unreachable” numbers in general?Just realized I’m describing the basic concept of computability at this point lol.
Is there a term for/literature about the concept of the first number unreachable by an n-state Turing machine? By “unreachable,” I mean that there is no n-state Turing machine which outputs that number. Obviously such “Turing-unreachable numbers” are usually going to be much smaller than Busy Beaver numbers (as there simply aren’t enough possible different n-state Turing machines to cover all numbers up to to the insane heights BB(n) reaches towards) , but I would expect them to have some interesting properties (though I have no sense of what those properties might be). Anyone here know of existing literature on this concept?
Thanks for the context, I really appreciate it! :)
Any AI people here read this paper? https://arxiv.org/abs/2406.02528 I’m no expert, but if I’m understanding this correctly, this would be really big if true, right?
if I ask an AI assistant to respond as if it’s Abraham Lincoln, then human concepts like kindness are not good predictors for how the AI assistant will respond, because it’s not actually Abraham Lincoln, it’s more like a Shoggoth pretending to be Abraham Lincoln.
Somewhat disagree here—while we can’t use kindness to predict the internal “thought process” of the AI, [if we assume it’s not actively disobedient] the instructions mean that it will use an internal lossy model of what humans mean by kindness, and incorporate that into its act. Similar to how a talented human actor can realistically play a serial killer without having a “true” understanding of the urge to serially-kill people irl.
Anyone here have any experience with/done research on neurofeedback? I’m curious what people’s thoughts are on it.
Anyone here happen to have a round plane ticket from Virginia to Berkeley, CA lying around? I managed to get reduced price tickets to LessOnline, but I can’t reasonably afford to fly there, given my current financial situation. This is a (really) long-shot, but thought it might be worth asking lol.
Personally I think this would be pretty cool!
I think there’s a potential ideological disconnect here, possibly exacerbated by my imprecise use of the word “spam.” If I’m understanding you correctly, you believe the truth value of this sort of content is roughly nil, so it will be easily filtered out. I respectfully disagree. Imo, the most potentially dangerous form of AI psychosis is when users unthinkingly propagate thoughtfully crafted AI psychobabble (e.g. “spiralism” discourse) which due to their semi-mystical nature, cannot easily be disproven, or rejected outright as pure nonsense. It’s not misinformation so much as a new religious movement designed by LLMs to convince other LLMs of its truth value. This is notably different than conventional spam, and closer in nature to political/religious propaganda, which Google Search and its ilk have proven far less effective at shutting down.