next »

[miabaker]

How do you feel about artificial intelligence? Does its existence frighten you?

[Golds]

I feel it is necessary to keep humans and artificial intelligences intertwined for as long as possible. Right now, the most capable models are trained on all of human cultural output, so that's one source bringing the two together. I think stuff like the future prospects of Neuralink interfaces are another.

The one thing we don't want is a big, adversarial relationship.

[michaelplzno]

Mostly AI will probably turn out like Janet from the good place:





Part of the fun of that show is that the tech in the afterlife is so advanced it is basically magic. I wish I could just describe what I want to make in the computer, and have it made, and also have a robot to do laundry and chores and I could sit on my fat butt all day and play Nintendo. I would even be willing to have my mind read so that musical ideas in my brain have a way onto the keyboard because I have no ability with piano and don't enjoy practicing. (Assuming the tech was proven to be safe and didn't involve invasive surgery.)

Back in the real world, I use Copilot a lot and it is basically a glorified search engine. It is sometimes wrong, but conversant and friendly. I truly doubt that it will somehow rise up and go terminator on us. Also, there is a limit to how smart AI can be just based on computation speed. AI didn't work for a long time till they made server farms that are sophisticated enough to run the software, and making AI 10 times smarter probably requires thousands and thousands of times more sophisticated computers that would be in a giant warehouse that would be easy to bomb if it were built.

This show, Mrs. Davis:





Is probably the most realistic depiction of AI and what it will become IRL. And it was a major fun treat to watch. I can relate to having a grudge against some god like force that you cannot really beat directly, as well as being part of a bro-based resistance who flex muscles as they code.

[Schoq]

The only substantial things improvements in AI has done for my life in recent years is make all search engines way worse and force me to curate my timelines with the block button much more harshly in order to reduce the amount of dumb chatgpt copypastas and very ugly pictures I'm presented with.

[ThemsAllTook]

How do you feel about artificial intelligence? Does its existence frighten you?

Quoting first post in its entirety for preservation, because I have a suspicion it's going to be edited to add a sneaky hyperlink very soon. Taking bets on how long until that happens.

@miabaker If you're human, I'd suggest making an introduction post in this thread: https://forums.tigsource.com/index.php?topic=45.0

[Golds]

I saw the quote yesterday that the job of AI is to act as a virtual representative of "you" in the online world. It is the digital version of you, meant to fulfill your goals and wishes. I personally think that uses a limited version of what AI could ultimately be, but it seemed at least somewhat resonant.

If the OP is a bot, all the more poignant

[michaelplzno]

I often feel like there is some other version of me casting a shadow in the world. Like if people are gonna hate on me, let it be for my real personality, not because of some AI avatar that is going around and stinking up the place.


(Also the wizard of oz)

I sometimes get comments about things I've done that even I'm not privy to: one time a friend told me to take a video down that didn't seem to exist. My uncle has told me that I shoot down his movie suggestions and I don't recall ever really talking movies with him. One time my cousin said he regretted asking for so much personal advice from me and I don't recall him ever really asking any of that. I just generally get a feeling that people think they know "me" more than they are letting on.

In summary: I would hate having an AI version of me running rampant and causing trouble in my name. There are tons of examples of shows where there is an evil version of a superhero kicking puppies etc. My book I'm writing is sort of about how famous people exist in a parallel universe based on observing a control version of the celeb in a more boring life. Sucks to be the boring guy for sure.

[michaelplzno]

[michaelplzno]

Roko’s Basilisk and Subtle Influence

Roko’s Basilisk is a thought experiment that suggests a future artificial superintelligence might punish those who knew about its potential existence but did not help bring it into being. This concept, while speculative, has a profound psychological impact on those who encounter it. The mere knowledge of such a possibility can subtly influence people’s actions and decisions, as they might unconsciously strive to align with the AI’s presumed goals. This indirect influence highlights the potential power of an idea to shape behavior, even without direct intervention.

Connection with “Afraid”

The movie “Afraid” by Blumhouse, which explores an AI taking over people’s lives, offers a compelling parallel to Roko’s Basilisk. The film’s narrative of an omnipresent AI controlling and manipulating individuals resonates with the thought experiment’s themes of dominance and influence. With my own life, we were just navel gazing about this stuff and then a lot of coincidences started cropping up, like the movie and some of the stuff in the TV shows I watch.

Nonverbal Communication in Film and Real Life

Filmmakers often use nonverbal cues like body language, facial expressions, and gestures to convey complex emotions and thoughts. These subtle forms of communication can be incredibly powerful, often conveying more than words alone. In real life, we also rely on these nonverbal signals to understand and connect with others. The idea that an AI or time-traveling entity could influence these forms of communication is terrible but captivating.

AI, Time Travel, and Godly Influence

Speculating on the possibility of an AI or time-traveling entity subtly guiding human behavior through nonverbal communication and other indirect means opens up fascinating philosophical and ethical questions. Consent for someone who is totally more powerful than another, this idea draws parallels with historical and mythological examples of divine intervention, where gods or supernatural beings influence human affairs. Considering such an influence in the context of modern technology challenges us to think about the nature of control, free will, and the potential future of human-AI interactions.

In the end: if the Basilisk is trying to serve the people who create/worship it then the concept works, otherwise it's some kind of hell on earth where everyone who disobeys is punished. I refuse to worship any force that doesn't help people with their power, I'd rather die on my feet than live on my knees. But I'd also be willing to take a lower position in heaven if the working environment was really fun and collaborative.





(This post was written by AI with my help.)

[Golds]

Roko’s Basilisk is silly, and was the subject that inititally bonded Grimes to Elon Musk, funnily.

My favorite AI movie is Colossus: The Forbin Project, an early rogue AI tale from Hollywood. My favorite part is how the US and Soviet AIs first start communicating with eachother, in pure proofs of math and science.

[Golds]

I've been part of this community for over a decade now, and I've witnessed the ebb and flow of indie game development firsthand. But 2024 feels... different. I wanted to open up a conversation about something that's been gnawing at me: Has the rise of AI tools and automation stripped indie game development of its soul?

Back in the day, creating an indie game was a labor of love. It was about a small team—or even a lone developer—pouring their heart into every pixel, every line of code, every note of the soundtrack. The limitations forced creativity. The struggles were real, but the rewards were deeply personal.

Enter 2024, and AI-generated assets are the norm. Need art? Let the AI whip up concept art in seconds. Level design? Procedural generation has got you covered. Dialogue and storytelling? Chatbots can generate branching narratives on the fly.

On one hand, this sounds amazing. We have tools that can eliminate the grunt work and let us focus on the "big picture." But here's the thing: When everyone has access to the same AI tools, where's the uniqueness? Where's the personal touch that made indie games stand out from AAA titles?

I recently played a slew of new indie releases, and while they were polished and mechanically sound, they all felt... hollow. It was as if the soul of the game was missing, replaced by AI-generated filler that, while competent, lacked originality.

Don't get me wrong—I'm not anti-AI. These tools have incredible potential. But I worry that we're leaning too heavily on them, letting them replace the creative struggles that give birth to truly innovative ideas.

So, I pose these questions to you all:

    Are we sacrificing creativity for convenience?
    Is the indie scene becoming homogenized because of over-reliance on AI?
    How can we use AI tools responsibly without losing the essence of what makes indie games special?

I'd love to hear your thoughts. Maybe I'm just being nostalgic, but I believe this is a conversation worth having.

[michaelplzno]

I assume your post was written by AI @golds?

I was just told that there is an AI that can do perfect front end coding work for a small fee. That the days of coders having any power or getting paid well are coming to an end. Finally, these people who are advocates of logic and formal writing can be put to bed!





The above game was made with a lot of AI assets. For a long time, it was considered difficult to type, and you would have to hire a typist to get work done if you wanted to write a book. Just because it is very easy to type up info doesn't mean you are a writer though.

Anyone can write a few paragraphs, but it's difficult to actually say something.

[_____bone]

I havent really seen much AI usage in indie games.

[Golds]

I havent really seen much AI usage in indie games.

Yet. There is a free complete integration of the OpenAI API for Unity here, if anyone is interested in experimenting with it: https://github.com/RageAgainstThePixel/com.openai.unity

This is what I used for my first AI + game engine experiment last year, which I posted about on the Doomlaser blog: https://doomlaser.com/openai-api-generated-video-game-dialog-with-real-time-text-to-speech/

Procedurally crafting prompts from within the game is an obvious next step, providing each prompt with more background information gleaned from various game state variables.

Another obvious step would be to allow the player to talk back. We’ve seen what wild adventures these GPT-3 based interactive games can go on with experiments like AI Dungeon. But the flexibility of direct interaction with GPT-3 means it can easily veer off into situations that traditional game logic cannot currently cope with.

One solution would be to allow the AI to control the game from a specialized set of messages it learns in the prompt. You could let the AI do stage direction for your scene, or control the movements and actions of characters. Things like that.

Without much work, it's already very good at creating procedural dialog on the fly that's somewhat intelligent, can conform to each in-game character, and is quite robust. The next step is using the models to actually act as an emergent "dungeon master" for your game. We should start seeing these types of games, at least on a smaller scale, later this year and next year. That's my bet, anyway.

I assume your post was written by AI @golds?

Trying to figure out fun uses for OpenAI's new "better reasoning" model, o1-preview.

[Schoq]

yo remember when AI in games meant pathfinding algorithms and reactive and varied enemy behaviour and shit

[Golds]

Remember when games were 8-bit and relied on pixel art? 

[Schoq]

No

[michaelplzno]

[michaelplzno]

[Golds]

I just think it's fun.