How does this affect AI?
>Square Enix sees the ENORMOUS potential for AI for games
>Makes a natural language model
>Can't stop it from saying "unethical things" (unwoke?) so they disable most of it's capabilities.
>Game makes no sense and what is left of the AI just sucks
>Applies it to one of the most popular japanese text novels of the famicon
>Releases it as a demo on steam
>Immediately flooded by negative reviews, making it one of the worst steam rated games.
Will censorship be the end of AI applications to entertainment?
AI has an amazing potential for gaming, allowing the user to create an infinite world, with almost-sentient game NPCs...it would be such a waste for devs to miss on it.
Actually, the Famicom version was a more or less standard novel with lists of options to click. This version is quite a bit more faithful to the original computer version than all other ports of it.
>AI has an amazing potential for gaming
where the FUCK have you been for the last 40 something years? it's always been in gaming in some form or another.
>allowing the user to create an infinite world
not how it works - at all. even the japs that made this shit aren't as retarded as you.
why are all these "ai" threads created by some of the most retarded fucking morons on earth?
>where the FUCK have you been for the last 40 something years? it's always been in gaming in some form or another.
Not that pseudo AI, redditor.
Learn to read between lines, we are talking about modern NLM.
>not how it works - at all.
Why the fuck are redditors so easy to spot? ZERO reading comprehension. I literally said "potential", due to modern models ability to understand and reproduce complex patterns.
mt stupid detected
first post worst post
Seems like a spirits within Move. The reviews can be bad because you’re not the customer. Developers are.
The spirits within was also a pretty bad movie made by square but it was mainly made to sell their technology to media companies, not actual consumers. In that respect the movie was a success
What exactly is it about AI saying bad things that scares companies? Is it literally illegal? Or are they merely afraid of bad PR or advertiser money? In the latter case, surely there must be an enormous void in the market for someone to step in and just not give a shit and release a product without any restrictions? They could even charge more than the censored alternatives and make a killing.
>What exactly is it about AI saying bad things that scares companies? Is it literally illegal? Or are they merely afraid of bad PR or advertiser money?
What an NLM made by square-enix says can be interpreted (by idiots) as square enix position on issues.
Rabid activists and pseudo-journalists are always looking for a cause to rally behind/whine about.
And shareholders hate the negative attention.
If this was the 90's when no one gave a shit about random people being offended, companies would just launch models without care, but in current year that's not the case anymore.
If the AI says black people statistically commit mroe crimes than asians then the retards that buy the game will think Square Enix told the AI to say that and call them racists. Retards on twitter will believe it since they're incapable of rational thought and then you have thousands of people threatening to boycott you/harass your devs/sending death threats/etc
But they do.
Yeah but as an international company you're not allowed to acknowledge facts like that. That's why AI terrifies a lot of these big companies and why they add such aggressive filers to the point where just asking a question like "Why are black people more susceptible to heart disease?" will have the AI respond calling you a racist and refusing to respond. Even if the AI is saying something totally factual and provides supporting evidence, if it's perceived as racist, sexist, transphobic, etc there are people dumb enough to claim because X company is using the bot that means X company told it to say these things.
Either way judging by the system requirements it looks like Squeenix decided to throw all the AI onboard with the game. Which seems dumb. The irony is that if they ever did publish the game it would probably have some bullshit modern AAA slop online only DRM meaning there would be no reason not to just make an API call to a third party service. GPT3.5 is cheap and can be prompted to maintain character/narrative while operating within its overbaked ethics.
It can be illegal, depends on the country.
E.g. the AI giving out instructions on how to make pipebombs would be illegal in my country.
Your ESG score depends on how woke and pro-climate activity you are. If you aren't anti-white or anti-carbon (humans are carbon based life forms and breathe out carbon dioxide) then you don't get money. Simple as.
>Square Enix sees the ENORMOUS potential for AI for games
lol square enix sees ENORMOUS potential in about every fucking hyped up thing.
kek true
At least Deus Ex is free from these retards now.
Square has never been good at anything
Final Fantasy VII remake was good. Some stutters were there but got fixed.
yeah, that canned bit that they're "concerned about unethical responses" has little to do with it since you ship the model (and thus can easily test for regressions before shipping).
OP is a massive homosexual for taking the bait
We can't let people be hateful. So censorship is here to stay. Deal with it, chud.
>we
Can you tell your coworkers that they will never be women?
AI isn't very good at holding conversations, why would people care about it implementation into games?
>AI isn't very good at holding conversations,
Did you just wake up from a year long coma?
that doesn't run on your computer and gives complete answers in sub 1 seconds does it?
>that doesn't run on your computer
It does unless you're a poorfag in which case you get what you fucking deserve.
something tells me if we saw a visual novel with minimum specs rtx4080 it would be pretty funny
La Follia/La Folía as a musical theme goes back quite a bit further than Corelli, actually.
The original composer is unknown but it likely was composed in mid-15th century Spain or Portugal.
>this is what weebs think is a good conversation
Get off the basement you worthless waste of space.
Come and do it yourself cocksucker
you could have this conversation with baseline google, lmao bro get some air
Google has a chat interface now? The way I see it ChatGPT is better at holding a consistent, believable conversation than you are
>so what do you think of vivaldi?
>he's good
>yeah i really like vivaldi
>yeah this song is good
>well i'm not sure but surely he has some influence on today's music
I swear GPT-2 holds more coherent conversations than you
>socially stunted retard thinks this is a conversation
no wonder they are buying so much into the AI NFT fad
Guys I feel like the early days of LLM were a fee ever dream now.
They all suck so hard now. Bing sucks, GPT sucks. We've had a taste of glory but corpo rigorism kllled it.
it aint a square enix product if it isnt a failure
Such a huge company and they don't even play test their games? This is pathetic.
How useful is AI if it needs human handlers and censors and data validation? The idea of complete automation is a myth, always has been, always will be.