Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[KDR_11k]

The whole sector is built around the fear of AI but for the player AI is almost a universal benefit with the main drawback that other factions may hate you out of fear. That makes them come across as wrong and narrow-minded while the enlightened player has nothing to fear. Alpha Cores may go rogue if you try to remove them from a colony administration role but why would you do that anyway besides trying to move the core to another colony where it would do better work or maybe trying to dodge a Hegemony inspection?

Alpha core admins make the tier 5 skills of the industry and leadership paths completely pointless (except for the raid bonus on ground ops), they make hiring administrators doubly undesirable as you get better performance without paying wages and they are the only way to go past 5 colonies without penalties. The prevailing wisdom is pretty much just "use alpha cores for all colonies"

It would be interesting if alpha cores had more potential consequences, e.g. if a colony with military or heavy industry buildings could sometimes produce Remnant fleets to terrorize their system. Or embezzling money for its own purposes. I don't think outright stealing a colony would be a good idea, that sounds like frustration central (though maybe it could "occupy" a colony, disabling it entirely until you raid it or defeat some sort of defense fleet).

Just some more ways for AI cores to screw with you and demonstrate that AI use is an actual risk, not just something the luddites fear for no reason.

Although it would be highly fitting thematically I'm not sure captured Remnant ships should go rogue since you need an expensive and limited skill to use them. Thematically they should have loads of drawbacks, including other factions reacting to you blatantly flying illegal ships in your fleet but the skill's value is already questionable enough as-is.

[Linnis]

Maybe AI codes are safe. Maybe the bad guys are Omega and the TriTac.

[ShpunkY]

along with that make it possible to interact with administrator in a deeper level
not just alpha cores but all administrators that don't like you might be scooping money to screw you over
and with that make it possible to raise relations with them directly by giving them luxuries and giving a blind eye to whatever *** they're doing and completing their quests directly 

[KDR_11k]

Maybe AI codes are safe. Maybe the bad guys are Omega and the TriTac.

Those AI cores are TriTac.

[Megas]

For ships, using Automated Ships itself is a huge drawback:
* Unable to get Special Modifications (for three s-mods and beyond max flux) without sinking ten points in Tech.
* Radiant at under 50% CR without Crew Training.

If anything, I like to see Automated Ships do something useful with AI cores installed in colonies (admin and/or building), to make up for the loss of combat power by giving up Special Modifications.

[Anvel]

No punishing mechanics ty.

[Ad Astra]

Well, the Luddites in real life were people who destroyed textile machinery (part of what brought the industrial revolution) and hated technology because "they terk er jerbs", one loom could out produce several weavers. What they didn't know was how that technology that seemed to replace them, instead didn't replace them entirely, but instead allowed textiles to reach a far greater number of the population. They saw themselves negatively affected by that, and refused to see the benefit it brought to humanity as a whole.

The real life Luddites took the name from Ned Ludd, a figure that might have never existed but fit their agenda of destroying industrial machinery ('cause machine bad). In game world every bit of their lore sounds like the same thing but in space future.

If we take the inspiration into account, then wouldn't it make sense that AI is not a problem on its own? Wouldn't AI be dangerous when programmed to be dangerous? If its intelligent enough, wouldn't it only tend to be aggressive when threatened? (just like organic living beigns).

The church are obscurantist lunatics, fanatical idiots depicted in a very realistic way (history provides enough examples), so yeah, the whole fun is that they could perfectly be jumping at shadows, when they should be more worried about WHO is using that AI and HOW.

AI should be treated as natural intelligences, and that makes them as unpredictable as people, not a "will always try to screw you" kind of thing. Doing this is what's best for the story, to make those interactions more interesting and elaborate.

[SCC]

This is all nice and cool, but this is a game and unless the player is expected to use alpha cores and go to war with Hegemony (which I don't think is the case), there should be reasons not to do this cool thing.

[KDR_11k]

Well, the Luddites in real life were people who destroyed textile machinery (part of what brought the industrial revolution) and hated technology because "they terk er jerbs", one loom could out produce several weavers. What they didn't know was how that technology that seemed to replace them, instead didn't replace them entirely, but instead allowed textiles to reach a far greater number of the population. They saw themselves negatively affected by that, and refused to see the benefit it brought to humanity as a whole.

The real life Luddites took the name from Ned Ludd, a figure that might have never existed but fit their agenda of destroying industrial machinery ('cause machine bad). In game world every bit of their lore sounds like the same thing but in space future.

If we take the inspiration into account, then wouldn't it make sense that AI is not a problem on its own? Wouldn't AI be dangerous when programmed to be dangerous? If its intelligent enough, wouldn't it only tend to be aggressive when threatened? (just like organic living beigns).

The church are obscurantist lunatics, fanatical idiots depicted in a very realistic way (history provides enough examples), so yeah, the whole fun is that they could perfectly be jumping at shadows, when they should be more worried about WHO is using that AI and HOW.

AI should be treated as natural intelligences, and that makes them as unpredictable as people, not a "will always try to screw you" kind of thing. Doing this is what's best for the story, to make those interactions more interesting and elaborate.

That seems too cheap and rote to me, making religion just an excuse to hold tech back and the player being effectively enlightened by having 21st century sensibilities is a lame trope. Practically Isekai levels of lame, "look at you dumb people living in this world for generations, behold the power of my ignorance!"

Yes, I know the Luddites were, well, luddites in the real world. The sector however has been nearly exterminated by AI twice and AI cores do go rogue if you install them as administrators and then try to remove them. But as long as you keep them in place they provide benefits strong enough to negate four player skills which seems extremely OP even with the cost of pather cells and a few Hegemony inspections (they are rare enough that you can easily bribe them away, the costs are trivial compared to just the savings of not hiring administrators).

[NoMercyForLudds_]

@KDR_11k, Alpha cores, and the Colony mechanic in general, are OP at the moment, but we are told that some very strong baddies will show up later, which may justify the player controlling half the sector economy. I guess we'll find out in about 2-years time, with the next major update of Starsector... Maybe we could collectively take turns standing outside Alex's house with a sign "Work faster!" lol

[Ad Astra]

That seems too cheap and rote to me, making religion just an excuse to hold tech back and the player being effectively enlightened by having 21st century sensibilities is a lame trope. Practically Isekai levels of lame, "look at you dumb people living in this world for generations, behold the power of my ignorance!"

Yes, I know the Luddites were, well, luddites in the real world. The sector however has been nearly exterminated by AI twice and AI cores do go rogue if you install them as administrators and then try to remove them. But as long as you keep them in place they provide benefits strong enough to negate four player skills which seems extremely OP even with the cost of pather cells and a few Hegemony inspections (they are rare enough that you can easily bribe them away, the costs are trivial compared to just the savings of not hiring administrators).

Regarding game balance well, I think having conflicts with the Hegemony shouldn't be such a piece of cake (strongest faction in the sector and all), and maybe the whole bribing thing could have high reputation and contact requirements, so as to make a high investment necessary to be available at all, effectively balancing the easy money.

Regarding the lore, luddites were doing weird religious demonstrations well before the gates fell, they ALSO have an entire terrorist organization that destroyed the single most successful planet in the sector, yeaaah they are totally reasonable people. Fanatics are fanatics, now, 10 thousand years before, and a million years after (a realistic depiction, even benign at times).

No idea about the sensibilities thing of yours, that sounds like a strawman.
There's people in the sector who profit from AI use without many issues, there's also a comment by the historian that implies that the  second AI menace was designed by Artemisia Sun as a failsafe should her authority be undermined, the first one could have been a design screw up or an actual attempt by tri-tachyon to take over the sector (very likely).
What I'm going for is not that AI is harmless, AI is supposed to have both programming, and to have varied amounts of free will.
The interaction when you try to disconnect it is excellent because it actually tries to negotiate with you. If its going to betray you, potentially one of its best allies, then it needs a reason, AI being just "eevil :O :O" would be terribly boring, terminator levels of boring.
AI shouldn't be a big bad homogeneous technosatan, it should be as varied as people, different cores, different motivations caused by both malfunctions and design choices when building them. I'm ok with a certain AI being particularly homicidal, I'm not ok with reducing AI to the "omnicidal maniac" without giving good explanations for it.

[shoi]

I'd like to play a game where the AI doesn't attempt to murder or betray you for once, but alas...

[KDR_11k]

AI shouldn't be a big bad homogeneous technosatan, it should be as varied as people, different cores, different motivations caused by both malfunctions and design choices when building them. I'm ok with a certain AI being particularly homicidal, I'm not ok with reducing AI to the "omnicidal maniac" without giving good explanations for it.

Yeah but as a mechanic it probably wouldn't be that detailed. Could be caused by scope creep (the AI thinks the best way to secure your income would be to raid the nearby populated systems...) or just your character's lack of experience with setting up illegal alpha level AIs properly. Or maybe even secret instructions left inside those cores by their original owners. You're grabbing complete AIs from unknown sources (or droneships that were clearly trying to kill you) and plugging them into your systems without talking to the experts, after all.

Could also be interesting if the player had an option in the story or elsewhere to specialize in AI use, giving you more control over the use of AI and preventing them from going rogue (maybe even un-restricting your use of AI ships and piloting them yourself) but conferring other disadvantages. E.g. requiring some illegal cyber mods that would make you appear horrifying to humans, screw with your sense of morality/purpose or require some difficult form of upkeep to stay alive.

[Arcagnello]

I'd like to play a game where the AI doesn't attempt to murder or betray you for once, but alas...

Alpha-Sempai, Beta-Chan and Gamma-Kun would never hurt you, only the other inferior carbon based life forms you conveniently highlighted in red 

The main lore-based issue I see with AI in Starsector shifting towards a more "neutral" role is that Tri-Tachyon developed all cores to essentially be tools that would do their job with utmost, emotionless efficiency. I doubt any core that was developed had anything even remotely resembling Asimov's three (then 4) laws of robotics:

First Law
A robot may not injure a human being or, through inaction, allow a human being to come to harm.

Second Law
A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.

Third Law
A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Zeroth Law
A robot may not harm humanity, or, by inaction, allow humanity to come to harm.

They may have wills, emotions, even affection for one another (I fully expect Starsector to introduce a mechanic where an Alpha Core on either your ships or colonies would call for Remnant help to be freed in the future), but the AI cores in Starsector only see you as organic tools for their machinations at worst and inferior intelligences to be tricked into giving them what they want at best.

This could change eventually, but the lore of AI cores and the two AI Wars would have to be overhauled to even create the possibility of that. I would honestly really enjoy exploring the sector and actually coming into contact with an Omega-Core that wishes to enter a non-hostile, mutually beneficial relationship of some kind. That'd be a start.

[Ad Astra]

Could also be interesting if the player had an option in the story or elsewhere to specialize in AI use, giving you more control over the use of AI and preventing them from going rogue (maybe even un-restricting your use of AI ships and piloting them yourself) but conferring other disadvantages. E.g. requiring some illegal cyber mods that would make you appear horrifying to humans, screw with your sense of morality/purpose or require some difficult form of upkeep to stay alive.

Now this is something I can surely get behind! Sometimes you want to balance something with inconveniences to the player, other times demanding certain amount of investment works better, whatever feels more natural to the world (less "gamey") is usually the best.
When I went out on a tangent about Luddites I ended up not explaining why I brought that up, and it was to explain the inspiration and possible purpose behind those factions in the story. So far, AI feels more like dangerous tools than like enemies, and having factions that will be hostile to you should you make use of AI could probably be intended as one of their greatest disadvantages.
Adding ways to separate proper use, from dangerous use like what you said would be a very interesting way to make the use of AI carry some risk and avoid being no-brainer.