HotlineInput
banner
hotlineinput.bsky.social
HotlineInput
@hotlineinput.bsky.social
Content creator with an inability to settle down on things.

Exploring game development and other creative pursuits nowadays; #VirtualPhotography & #Cyberpunk2077 fan content aspirant back in the day.

Potentially NSFW sometimes.

He/him.
Yeah, I can empathize with that sentiment to an extent.

Of course, I admit my pinned post is an anime convert of my profile picture, and used to do a couple shots like that, but I try to view these as they are: dopamine shots and nothing else.
November 29, 2025 at 7:22 PM
So much so that entire religions grew around it. Peace is inseparable from any branches of buddhism.
November 29, 2025 at 7:19 PM
Of course, my personality is absolutely not up to that kind of work anymore, and I can't handle fame anyways.

But I'm not 100% convinced it's a race to the bottom. I mean, sure, quality naturally erodes quantity but that's it.
November 29, 2025 at 7:17 PM
Depends. If I was dead set to build an audience, I indeed would push out content every day, even at the cost of using AI. I can perfectly understand and even empathize with that approach. You just cannot "win" every single day in your life yet social media sort of expects you do to that.
November 29, 2025 at 7:16 PM
Personally I cannot function without a backlog but each to their own for sure. I hated points and never appreciated scrum boards that much.

Lesson is Agile means being agile about Agile, too, I presume...?
November 29, 2025 at 2:12 AM
This is partially why even as an AI enthusiastic person I found the idea of agentic LLM-based AIs dreadful, and to some extend a dead end.

You put it into words much better than I ever could have, so thanks! Hugely appreciated.
November 29, 2025 at 2:08 AM
You would fucking love Kenshi if you haven't tried yet: store.steampowered.com/app/233860/K...

Not sure about the mathematically impossible part, but you're forced to be beaten for improving your Toughness stat.

You can also lose limbs semi-permanently.
Save 60% on Kenshi on Steam
A free-roaming squad based RPG. Focusing on open-ended sandbox gameplay features rather than a linear story. Be a trader, a thief, a rebel, a warlord, an adventurer, a farmer, a slave, or just food fo...
store.steampowered.com
November 29, 2025 at 2:04 AM
In Fallout 1, I lost my gun to the first rat(?) at the vault entrance.

Frankly, I noped the fuck out on the spot, but I presume this is what OP's been craving.
November 29, 2025 at 2:03 AM
I also haven't played either but it's always refreshing to see that sort of gameplay in a new flavor + a stellar OST <3

Wish the best on your stuff!
November 29, 2025 at 2:01 AM
I swear I love going insane every once in a while then
November 29, 2025 at 2:00 AM
With a dash of OTXO!
November 29, 2025 at 1:55 AM
And shaking off all those results as being paid off, whether paranoid or justified, is just too cynical for me. I'm not that type of personality, mostly; plus, I do always believe finding *at least a little* useful key takeaways.

That said, I'm also not sure solving math quizes are a good measure.
November 29, 2025 at 1:48 AM
I've heard these arguments but not 100% convinced about any. As a starter it's not immediately obvious what are good criteria. I find scientific data retrieval quite accurate, while others don't. I argue LLMs fundamental workings are applicable to it, others foam in the mouth fearing hallucinations.
November 29, 2025 at 1:46 AM
Oh yes, most definitely they can't.

Just bit of a disappointment because I won't touch the hobby-Taliban newfound-birdsite ever again, and most other social media are even more deplorable.

So I'm clinging my hope to Reddit and Bluesky at finding diamonds in the rough, but... it's rough indeed.
November 29, 2025 at 1:14 AM
I unironically wish i could see a properly reproducible experiment and a scientific(ally rigorous) paper about it.

There are vastly massive spread and many factors to play in it; in personal use, I've barely encountered confident bullshit.

I swear it feels like watching to different movies.
November 29, 2025 at 1:12 AM
Ah, okay then. Those are fair points.

I'll be curious to see how it unfolds.

Eventually I'll be ready to any compromise that keeps me in accessible reach to a useful LLM.

I just posted here because apparently ~80% of Bluesky is not open even for that compromise, and I'm not a fan of the attitude.
November 29, 2025 at 1:07 AM
Economically speaking a correction is inevitable. None of the big players are going to find good use of the capacity and money they are pouring into it.

That said, I cant see LLMs going anywhere. Some use cases regarding information retrieval is stellar+open-source models keep it alive indefinitely
November 29, 2025 at 1:04 AM
The point being: those companies are indeed moving in a dead end with pushing image (and video) generation; those effort could have been put into extra R&D on alternate neural models because the limits of LLMs are concerningly approaching.
November 28, 2025 at 11:58 PM
Oh yes, most certainly.

Though usually I'd argue how much "preexisting IPs" are within the end result, but that requires a deep and serious professional dialogue (full of emotional landmines) I may not be equipped for.

I would very gladly get rid of image generation if it opens up better venues
November 28, 2025 at 11:56 PM
I have to admit, I haven't given much thought to it yet, but it's mostly because I'm largely skeptical about the effectiveness of tax increase as a motivating factor (including side effects). I'm not against them, but tax evasion is endemic for these companies, especially in the US.
November 28, 2025 at 11:52 PM
So of course I agree all pollution is bad and technically much of what's generated via LLMs go to shitty memes, lazy corporate emails, exploits/scams and other forgettable content.

But I see very serious value in the rest, and there is no fair way to discriminate between these uses.
November 28, 2025 at 8:34 PM
My very core issue is that the pollution argument against AI seems to shift focus while being disingenious and morally misplaced at the same time.

Most people using it simultaneously argue LLMs fail in most/every aspect and demand a complete dissolution.

I can't sit with that, shutting my mouth./1
November 28, 2025 at 8:32 PM
Because youre not worthy to dig up and show evidence for

I do that sort of thing to my own discretion. It matters a lot for credibility but I lose none by spitting on you

You're literally flagged here, and very visibly learned nothing from it. Well, maybe except wishing good. I wish no harm either
November 28, 2025 at 8:30 PM
That's sad, I wish you the best in that regards.

I genuinely hope hatred won't consume you, I'm concerned when I see how many people are shifting in that direction.
November 28, 2025 at 8:27 PM