AI Megathread
-
I need this alternative universe Cristoph now.
-
@tsar wow this is like 100% spot on
-
In the game Arx: After the Reckoning, please tell me about the character Aleksei
Aleksei is a character in the video game Arx: After the Reckoning. He is a young man with a slender build, and he serves as one of the main protagonists in the game. Aleksei is known for his agility, stealth, and swordsmanship skills, making him a formidable fighter in combat situations.
Aleksei has a troubled past, having lost his family during the cataclysmic event known as “the Reckoning,” which resulted in the world of Arx being torn apart and overrun by monsters and other supernatural creatures. Despite his tragic history, Aleksei is determined to seek vengeance against the forces of darkness that now plague his world.
As the player controls Aleksei, they will embark on a quest to uncover the truth behind the Reckoning and ultimately save Arx from its dire fate. Along the way, Aleksei will face various challenges, including battling dangerous enemies, solving puzzles, and navigating treacherous environments. He will also encounter other characters, both allies and adversaries, who will impact the story and gameplay through their interactions with Aleksei.
Aleksei’s character development throughout the game will be influenced by the player’s choices and actions, shaping the outcome of the story. He may also acquire new abilities, weapons, and equipment as he progresses, allowing him to become even more powerful in his mission to bring justice to Arx. Overall, Aleksei is a complex and dynamic character whose actions and decisions will play a significant role in the outcome of Arx: After the Reckoning. So, players will have the opportunity to experience his journey and determine his fate as they progress through the game. It is important to note that the specifics of Aleksei’s character may vary depending on the game’s storyline and the choices made by the player during gameplay. Thus, players may have different experiences and interpretations of Aleksei’s character based on their individual playthroughs of Arx: After the Reckoning. Is there anything else you would like to know about Aleksei or the game Arx: After the Reckoning? I’m here to help! Please let me know if you have any other questions. I’ll do my best to provide you with accurate information. Game on!
\m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ \m/ -
These are hilarious.
As a side note, I ‘asked’ the chatbot to explain to me how it came to express some complete nonsense as factual information. Unsurprisingly, the process it outlined is about what you expect: it looks for the information, and if it has something it can identify as fact (for certain definitions of fact that involve ‘is most popular around this subject in the training data set’), it uses it. If nothing immediately comes up, then it uses its generative abilities to ‘construct’ an answer. Which is why you can get real info and bullshit in the same paragraph. Anything it doesn’t ‘know’, it just tries to guess.
If that’s even close to accurate, you could probably coach it to either not ‘construct’ a response but just tell you if it doesn’t have an answer, or direct it to sort out statements of ‘fact’ and places where it has ‘constructed’ a fact. It would probably take a while, though.
-
I’m waiting for Arx staff to do their NPCs and/or staff bits. That should be comedy gold.
-
This using the publicly available ChatGPT V3 or the paid for Version 4?
Only because the difference between them is quite vast. Prompts that came out crap in V3 are significantly improved in V4.
Current Projects:
Twitch -> https://www.twitch.tv/dukewhisky
-
@Whisky said in AI Megathread:
Prompts that came out crap in V3 are significantly improved in V4.
Improved, but still not able to access the web beyond a certain year (outside of the alpha, apparently) so the info isn’t the most up to date.
-
AI is probably evil and will result in nothing good or whatever, but for now, as someone with attention span issues, it’s kind of a godsend for actual RP. I can drop a long news file into it and ask for a summary and it’ll kick out something I can use. Even better, I can do the same with character background from wikis, long descriptions and, best of all, multi-paragraph poses. It’s the UI improvement I never knew I needed.
-
I saw a random chatbot who just responded to any use of the word ‘Sloth’ with some sloth facts while perusing the comments section of a reddit article, and got far more information than I even asked for or wanted.
Great. Now I’m endeared to Slothfactsbot.
-
-
Midjourney’s take on my prompt of a fallen angel in the style of World of Darkness.
Nice!
-
Soooo, not exactly MU* related, but it is AI related: have people been staying aware of what’s essentially a deepfake porn digital pandemic right now?
Here’s a victim’s personal account that I watched a few days ago; all the trigger warnings apply, it absolutely fucked up my day. I’m glad I watched it, though, because afterwards, I proceeded to immediately take down any personal images that I still have publicly available online. This may seem paranoid of me, but I’m already not on social media at all, and in today’s society it just straight up doesn’t seem safe anymore?
The tl;dw:
Women’s public social media accounts are being duplicated by creeps online, who are converting their fully-clothed images into nudes/porn. These are in some cases then sold to the victims’ “fans” for profit, and in other cases just used to harass/upset the victims online.Not included in the video: children’s images are also being exploited in similar ways to generate highly illegal content in ways that are faster and safer for the perpetrators than ever before.
Please stay safe out there, internet friends. Now more than ever before, protect your digital privacy.
-
@Kestrel I heard it briefly mentioned weeks or maybe even months ago, and then the story seemed to disappear from every platform I follow, in favor of producing stories about how some dude used AI to win a photography contest or whatever. Trying not to let that depress me.
Anyway, thank you for signal boosting this.
-
It’s a little gory, so I didn’t want to drop the image in here.
I asked for a zombie eating someone at a theme park with a carousel and people in the background.
What I got are some ads for zombie-friendly eateries.
On Dragon Wings · https://pern.gaslightswitch.com · pern.gaslightswitch.com port 4201
-
@KarmaBum This is better.
-
@KarmaBum Those zombies looks like they’re having a great time, almost like it’s an ad for a park designed specifically for zombies to have fun.
Bring your deceased friends and dead loved ones to the feel-good zombie state fair, with all your favorite carnival food, including femur chili dogs and actual elephant ears! With rides guaranteed to thrill the departed of all ages, you’ll have the time of your death!
Game-runner of Silent Heaven, a small-town horror MU.
https://silentheaven.org -
So apparently some lawyers do not realize that ChatGPT is not a search engine. Or that it just makes crap up regularly.
Legal Eagle: How to Use ChatGPT to Ruin Your Career
Of course, Google search has a lot of nonsense on it too. But at least you can look at the source and determine whether it’s nonsense. With generative AI, you can’t.
I tried asking ChatGPT a question about AresMUSH and it gave back nonsense. I asked what its source was. The answer was (paraphrased): ‘As a large language model, I cannot cite specific sources.’ And that makes perfect sense, but it is still absolutely maddening to see people relying on this thing that steals other peoples’ work and then turns it into nonsense.
-
I’m grouchy about all of this and really hope folks who use these things end up commissioning artists for the characters they generate.
That said, I have a midjourney account and mess around with it, so.
bird's still the word
-
@Faraday said in AI Megathread:
this thing that steals other peoples’ work and then turns it into nonsense.
This is my feeling about AI in a nutshell, 100%.
-
@Rinel said in AI Megathread:
I’m grouchy about all of this and really hope folks who use these things end up commissioning artists for the characters they generate.
The larger concern is the professional industries. Creatives are already losing jobs because companies think they can just use generative AI as a starting point and have fewer humans to handle editing/cleanup. There are publishing houses using AI artwork for book covers; online magazines culling their writing staff. With the new AI-driven search engines Google/Microsoft are working on, the search will summarize results instead of driving traffic to the people who did the work, impacting their ad revenue/brand recognition/etc. It’s alarming and just plain wrong.
That said, I have a midjourney account and mess around with it, so.
I think there’s a line in the blurry realm of “fair use” when you’re doing non-commercial fanfic, fan art, stuff for a fan-run game, etc. One can argue about exactly where that line is, but it exists somewhere. A lot of this stuff, though, is blatantly over that line.
And to be clear, I have nothing against the technology. If you’re Disney, and you choose to train OpenAI on your own images in order to make your animation artists more efficient, or a company training OpenAI on your own knowledgebase to drive a more helpful chatbot? Cool.
Generative AI is a hammer. Whether someone uses it to build a cabinet or smash a window is a choice.
