I actually can envision the eventual benevolent AI winning out over anything more nefarious.
Can’t say much about work, but it’s not hard to infer if you dig into my LinkedIn profile.
I actually can envision the eventual benevolent AI winning out over anything more nefarious.
Can’t say much about work, but it’s not hard to infer if you dig into my LinkedIn profile.
But why is that? That’s what we are trying to discuss… wait, are you an AI? Holy shit!
Okay, write me a response that explains your position to an audience that believes the opposite point of view.
And we’re the ones falling victim to TV tropes re:Evil AI?
I understand well that DavidC is not allowed (and doesn’t want) to tell details of his job.
But without telling job details I think there are not so much different areas for AI:
This is probably not a complete list, but I guess it boils down to these.
@Pebblerubble For sure. I think AI will revolutionize many industries for the better. AI surgery and medical diagnosis will improve our recovery and health immensely. That sort of stuff excites me.
I do worry about social harm to our society, which I feel will be the “first contact” so to speak. If you can agree that social media has done some harm to us… imagine a personalized AI feeding you your media constantly, tracking your reaction to content before you can even process how you feel about it, adjusting the flow to keep you watching so that advertisers can maximize their profits and platforms can retain users.
This is what is already happening with YouTube and it’s recommendation algorithm. It’s not hard to go down a rabbit hole on that platform. There have been a few times where I just had to stop because it was feeding me too much of one kind of opinion… and I felt that I wasn’t thinking for myself. That kind of scared me. This is how I see AI harming us… because I’m witnessing it.
Imagine when it gets competent enough that we don’t realize we’re being conditioned. It’s like having a parent that never says no. That’s not healthy, but that’s what’s happening.
Yes, and the sad thing is that all people I know and who listen to YT or Spotify etc. don’t care about being manipulated.
And let me quote from a book I’m reading:
Someday workpower will be obsolete. Then only one thing will decide whether you belong to the slaves or the masters: It’s the fact who owns the machines.
6 posts were merged into an existing topic: Pulled message from ARG topic
Here’s GPT 4o:
Compare counterfeit monkey and a mind forever voyaging.
“Counterfeit Monkey” and “A Mind Forever Voyaging” are both notable works in the interactive fiction (IF) genre, though they differ significantly in their themes, mechanics, and narrative approaches.
Developer: Emily Short
Release Year: 2012
Overview: “Counterfeit Monkey” is an interactive fiction game set in a world where language manipulation is literal and central to gameplay. The game takes place in Anglophone Atlantis, a fictional island where the protagonist uses a device that can alter words and their corresponding physical objects.
Themes:
Mechanics:
Narrative Style:
Developer: Infocom (written by Steve Meretzky)
Release Year: 1985
Overview: “A Mind Forever Voyaging” is an interactive fiction game set in a dystopian future where the protagonist, a sentient AI named PRISM, is tasked with testing a political simulation. The game focuses on exploring different simulated futures based on a proposed government plan.
Themes:
Mechanics:
Narrative Style:
Overall, “Counterfeit Monkey” and “A Mind Forever Voyaging” offer distinct experiences within the interactive fiction genre, each showcasing the versatility of IF in exploring complex themes through unique gameplay mechanics and narrative styles.
What are some games by Andrew Plotkin?
Andrew Plotkin, also known as “Zarf” in the interactive fiction community, is a well-known author of interactive fiction games. Some of his notable works include:
Andrew Plotkin has also contributed to the development of interactive fiction tools and platforms, including the Glulx and Glk systems, which are used to create and play modern interactive fiction games. His works are celebrated for their innovative use of the medium and their engaging, often thought-provoking narratives.
What about System’s Twilight?
“System’s Twilight” is a notable game created by Andrew Plotkin. Here are some details about the game:
Overview: “System’s Twilight” is a hybrid puzzle and interactive fiction game that combines logic puzzles with a narrative set in a unique, abstract computer world. It was one of Plotkin’s earlier works and is celebrated for its innovative design and challenging puzzles.
Themes:
Gameplay:
Legacy:
“System’s Twilight” is a testament to Andrew Plotkin’s creativity and his ability to craft compelling and intellectually rewarding experiences in the realm of interactive fiction and puzzle games.
I’m going to upload a transcript of a playthrough of a parser-based IF game.
Sure, please go ahead and upload the transcript. I’ll take a look at it and help you with any analysis, feedback, or other insights you might need.
(Uploaded a transcript of the original competition version of Whom the Telling Changed with the following prompt.)
Consider the file ‘tell-scr.txt’. What game is it?
The game in the transcript is “Whom the Telling Changed” by Aaron A. Reed. This interactive fiction game was first released in March 2005, and it blends storytelling with cultural and social themes, focusing on the interplay of narrative and decision-making within a tribal community. The player takes part in a storytelling ritual that affects the decisions and fate of their people, exploring themes of leadership, conflict, and cultural values.
I’m even more convinced now that we’re nowhere near AGI given the existing technology.
Vector databases can only give you what exists. They can’t accurately connect things or see large concepts without the glue of human interaction. The tech that can do that hasn’t been invented yet and LLMs are not it.
There are threats to some industries though. Any information-based business where their guidance is public knowledge can be trained into an effective model and replace that business’s activities. Tax preparation comes to mind. The government publishes all of the tax code. It will be trivial to train an LLM model to do even the most complex tax returns.
The current AI craze is the dumbest bubble ever and maybe even dumber than web3/blockchain.
Good take.
There are threats to some industries though. Any information-based business where their guidance is public knowledge can be trained into an effective model and replace that business’s activities. Tax preparation comes to mind.
They’ll find some way to make it illegal to use the technology that way, gimping easily accessible LLM implementations so that they won’t work for much of anything economically viable, and making use of LLMs through approved systems part of expensive software packages that will practically just turn “AI-powered” into a marketing label and blocking practical use of this technology from ordinary customers except for an enhanced inline chatbot.
That’s why I want to use it now before it gets too gimped and commercialized. I mean, it’s already just about there. Of course, making our own LLMs would be a lot of fun. Maybe it won’t be illegal to ask GPT or Gemini to teach me how to do that before I get around to trying it on a Linux box at home.
And I still like blockchain.
There are a crazy number of open source models on huggingface. The cat is out of the bag. Pandora’s box is irrevocably open.
Our wacky first amendment will protect anyone using an open source LLM model to do taxes.