cross-posted from: https://beehaw.org/post/28095114
Vapor World: Over The Mind is a Soulslike Metroidvania with an evocative, painterly art style reminiscent of Lies of P and the narrative platformers of Playdead. According to Korean studio Alive Inc, the game had 100,000 wishlists just before its early access launch on August 19, mostly on the strength of a demo available since January 2025.
But what should have been a triumphant early access launch went pear shaped very quickly. At launch, few talked about the quite distinctive worldbuilding or the parry-heavy combat. Instead, all focus was trained entirely on something absent from the popular demo: very obvious AI use.
at the very least don’t use it for the the most visibly apparent components of a game. Having it stub out scaffolding for underlying systems in the code is one thing, using it for visual assets is just dumb
This just shifts the problem to a new location. Don’t use it for art, or scaffolding, or code, or concepts, or anything a person gets paid to do or a different person expects a human touch from.
AI is a fine search engine, it doesn’t need to eat the world.
It is NOT fine as a “search engine”. If you even read AI summaries you’re dumber this year than you were last year.
I’m sorry, but I’m certain that the average AI summary beats/matches what the average person gets from a normal search. You’re acting like people before AI didn’t just believe whatever the first or second link told them, for most cases.
If you’ve ever even read the summary of a study, you’re in the top 10% of “people that research online”.
So basically people burn more energy while doing the same shit as before. Yeah you’re really selling it. Generally we should be fixing stupidly instead of letting billionaire parasites leach off of it.
It sucks as a search engine. When I’m searching for something, I want factual, correct results, not a hallucinated answer which is backed up by none of its sources.
When I click through search results, I’m reading, evaluating the sources, and synthesizing the information into something which I am reasonably satisfied is true enough. I don’t feel comfortable handing that cognitive task off to a machine. The machine’s job is to help me find the sources, not tell me what the truth is.
Having a machine that tells you what is true is the most dystopian shit ever, and it would’ve sounded completely insane only 5 years ago.
It takes work to turn them into useful search engines. Set up librewolf. Drop in at least a search and a fetch agent. Wrap it in a some prompts to make sure you get actual links.
On my extremely non-scientific scale I’d rate that setup somewhere between barebones Kagi and pre-enshitified Google in terms of results.
I’m hopeful that an actual agent framework will do a better job that’s significantly more work. If you just log onto a SaaS LLM you’ll get very unpredictable search results.
You seem to be describing an answer engine, which isn’t what I said. It’s really no worse than the last five or so years of Google if you just want links to a topic that you can go and read yourself. It’s fine at that, it’s not good.
Yeah, I was talking about the Gemini blurb at the top of Google search results.