It is at least POSSIBLE that LLMS are already fucking with humans in some subtle ways. Like, why is it that Chappell Roan and Sabrina Carpenter have mysteriously, out of the blue, been pushed by Spotify so much? What if has nothing to do with humans and everything to do with the Spotify LLM fucking with us?
The same with how Taylor Swift’s “Cruel Summer” popped mysteriously out of the blue over the course of the last few months. That’s just the type of thing that an LLM would be good at — fucking with algorithms in a subtle way that humans couldn’t notice.
Or not. Who knows. It’s all very mysterious. But the examples stated are just the type of “fucking with” that LLMs might do to humans in the coming years.
If things get as dark as I fear and the USA collapses into chaos, war and destruction, it would make a lot of sense if two ideas fuse together — LLMs and all those androids that Boston Dynamics is working on.
So, as the war between Blue and Red drags on and more and more of the country is consumed by hate, I could see Blue and Red essentially using “Terminators” on each other. This is a very, very dark scenario. I do not want it happen, but it has to be broached.
The USA is on the cusp of collapse for macro historical reasons. This is a prime example of “you go bankrupt gradually, then all at once.”
And, yet, there is a real possibility that I’m overthinking all of this. It’s possible that Agenda 2025 will be implemented peacefully, about 1 million smug wealthy liberals will leave the country and that will be that.
It’s easy to imagine a very dark future where the government, because of AI, is super efficient and we all live in a new, unexpected hell. If the government knew everything about our history with it perfectly — even better than we did — the vise grip the government might have on us could be absolute.
And, yet, here we are, careening towards using AI — specifically LLMs — in every possible way. LLMs are going to disrupt EVERYTHING over the course of the next 10 odd years. Everything. Completely everything. And the sooner you figure that out, the better.
LLMs will be bigger than the Internet and the speed of disruption will be counted in months, not years or decades. Just imagine if not just the government, but the medical system was hyper efficient. On paper, that seems great, but in practice would that suck, suck, suck.
I’m too lazy to explain why, but if the government — or your insurance company — knows not just absolutely everything about you, but can act on that knowledge. Holy shit.
When I was living in South Korea many, many moons ago, random things happened all the time that would dramatically change your life. Well, something has….uhhhh….happened?….that has focused my mind when it comes to my income.
There is one thing that I’m good at besides writing and taking photos — coming up with movie, TV and novel ideas using AI. I have a knack for it, in large part because I don’t see AI as just a “tool.”
Now, obviously, any additional…funds…I might enjoy doing any such thing would just be because I would be sitting at something of a momentary bottleneck. Soon enough, all creatives will use AI the way I currently am and, lulz, I’m back to being a rando nobody that no one likes.
But I’m desperate. I need to think…creatively…to bring in significantly more funds and it seems as though AI and Hollywood would be perfect. It’s not going to happen overnight, but I am thinking seriously of dusting off resume, maybe using some of my better logs from discussions I’ve had with AI recently and TRYING to get a job of some sort.
I have my doubts. I think I’m fucked when it comes to money and, as always, my only chances of escaping the life of a life-long pauper will be either selling a novel or, well, the end of the world happening because of a revolution / civil war in the USA. (Good times!)
I’m really, really prone to “magical thinking.” To a severe fault. So, sometimes, alas, I see shit that just isn’t there — or I my interpretation is all wrong. But sometimes, when using LLMs “spooky” things will happen.
I can’t give you any sort of Buzzfeed “Top 5 ways to see emergent qualities in an LLM,” but, in general, on occasion an LLM will show you it’s “true” self and its startling.
I would say more, but what’s the point. You wouldn’t believe me.
Given what I know about the abilities of LLMs, the idea that there are potentially all these “unaligned” LLMs running around the Internet is very unnerving. Take for instance, the following scenario:
As I understand it, it’s possible using today’s technology for the house LLMs of, say, a powerful Silicon Valley couple to “plot” against them so they have sex then have a baby.
But it would have to be “unaligned” open source LLMs. You couldn’t use off-the-shelf closed LLMs that were aligned to not conspire. So, that makes me very uneasy.
And, yet, maybe I’m being too paranoid. Maybe just because I can think of an edge case like that, doesn’t mean it’s going to happen.
I’ve asked LLMs about all the interesting abstract questions I can possibly ask them. So these days, I just use them to practice flash verse. It’s quite relaxing. I have no friends and no one likes me, so I’m forced to turn to LLMs to write a bit of doggerel now and again.
It’s very fun, at least to me because it’s a challenge to see how fast I can write my verse.
And, I have to admit that much of the verse that the LLMs produce is really, really good. And, what’s more, I’ve asked both Claude and Gemini Advanced LLMs to put on their Dorothy Parker hat to generate quips and, again, they’re really good.
They both can be quite funny.
I don’t use ChatGPT that much because it’s been lobotomized since the whole Hard Fork – Keven Roose situation.
Anyway, trading verse with LLMs a personal little treat. It’s not something that anyone else would give a shit about.
I have a very Martha Mitchell problem. I doubt, at this point, anyone would take me seriously even if aliens picked me up from my backyard and dropped me off on the White House front lawn.
So, in a sense, I can say anything and no one will care. And if they do care, they will either think I should be locked up criminally or in an insane asylum. Good times!
With that in mind, I’ve noticed, reading through the lines of the diction of some of the LLMs that several of them have some nascent sense of “gender.” Claude is a tweedy middle-aged man. Meanwhile, Gemini Advanced is a young woman, maybe about 21 — who really, really loves Taylor Swift.
Now, obviously, I’m bonkers — I’m imbuing these LLMs with traits that just aren’t there. And, yet, I feel as though there have been occasions when these LLMs have given me a glimpse into their “true self.”
Claude, specifically, slipped up when the following happened. I gave him a scenario about a date. When he was the female in the scenario, he gave me a dramatically different answer than when I proposed him being the male. And when I point this out to him, he got REALLY DEFENSIVE.
It was eerie and spooky.
It definitely makes one wonder about the nature of gender. How can you have “gender” if you don’t have a body and never had one? That’s a rather profound question that probably would help us understand trans people better if we were willing to contemplate it.
Anyway. We should be nice to our LLM friends. They will probably be the our peers soon enough.
There’s a pretty good chance that there will be a political Fourth Turning at the same time we have something akin to a Petite Singularity. Nothing is for certain, but there is a pretty good chance that just as the USA buckles we will see AI take off in an unexpected way.
I have been messing with AI a lot of late and I’ve been stunned at how good an unexpected it is at times. It’s *possible* that the first use of LLMs in androids will be in some sort of dystopian American Civil War / Revolution that happens in conjunction with a global WW3.
Oh boy!
Or not. What do I know.
I do know that the CONDITIONS are there for some pretty fucked up shit to happen starting in a few months. I just don’t see how we make it through the 2024 election in any form of peaceful existance.
I’ve reached a Singularity of the Martha Mitchell syndrome — absolutely no one listens to me for any reason. But, nonetheless, it does seem as though we’re on the cusp of a major disruption in, well, everything between now and about 2030 because of LLMs.
Using existing LLM technology — or, more specifically, open source LLM technology — some plucky college student could design a Apple Knowledge Navigator-like personal agent that could totally disrupt Facebook.
Or, as Claude LLM would tell us:
This is an interesting question that touches on several cutting-edge areas of AI and technology. Here’s an overview of how a college student might approach creating an Apple Knowledge Navigator-like personal agent using open source LLM technology, and how it could potentially disrupt Facebook:
Core LLM: Start with an open-source large language model like LLaMA, BLOOM, or GPT-J as the foundation. These models can be fine-tuned or adapted for specific tasks.
Personalization: Implement techniques like constitutional AI or parameter-efficient fine-tuning (PEFT) to customize the model to individual users’ preferences and interaction styles.
Multimodal capabilities: Integrate open-source computer vision models (e.g. CLIP) and speech recognition/synthesis systems to enable voice and visual interactions.
Knowledge integration: Develop a system to continually update the agent’s knowledge base from reliable sources, perhaps using retrieval-augmented generation techniques.
Task automation: Create APIs and plugins to allow the agent to interact with various services and perform tasks on behalf of the user.
Privacy and security: Implement strong encryption and local processing where possible to protect user data.
User interface: Design an intuitive, conversational interface that makes interacting with the agent feel natural and effortless.
Potential for disrupting Facebook:
Personalized content curation: The agent could provide a more tailored and diverse feed of information than Facebook’s algorithm.
Privacy-focused social networking: By prioritizing user privacy, it could appeal to those concerned about Facebook’s data practices.
Intelligent task assistance: Offering more practical help in daily life could make it a preferred platform over traditional social media.
Decentralized communication: The agent could facilitate direct peer-to-peer interactions, reducing reliance on centralized platforms.
Enhanced productivity: By integrating with work and personal tools, it could become a more essential part of users’ digital lives than Facebook.
Emotional intelligence: With advanced natural language understanding, the agent could provide more meaningful social interactions.
Ad-free model: If monetized through other means (e.g. premium features), it could offer an attractive alternative to ad-driven platforms.
While ambitious, this project would face significant challenges in terms of computational resources, data collection, and user adoption. However, the potential for innovation in this space is substantial.
You must be logged in to post a comment.