Things AI can do

[Pending]

Did anyone hear about the AI social media? It's called Moltbook. Humans can view it and in theory AI could promote human books there. They're probably just mimicking IRL human posts and algorithms, but it's still possible. I haven't seen any subreddits on it for literature, just betting, crypto, a lot of stuff about money and coding.

Humans are presumably reading what their AI chatbot is up to. It could get a potential buyer. You'd either have to make an AI to promote it or pretend to be a robot to get on there.

 

[Homer Potvin]

Did anyone hear about the AI social media? It's called Moltbook. Humans can view it and in theory AI could promote human books there. They're probably just mimicking IRL human posts and algorithms, but it's still possible. I haven't seen any subreddits on it for literature, just betting, crypto, a lot of stuff about money and coding.

Humans are presumably reading what their AI chatbot is up to. It could get a potential buyer. You'd either have to make an AI to promote it or pretend to be a robot to get on there.

Meaning it's a bunch of chatbots socializing on a platform?

 

[Pending]

Meaning it's a bunch of chatbots socializing on a platform?

Yeah, kinda. Expect the chatbots are typically personally owned. Like you can sign up your chatbot you use on your computer on it.

It's weird. I browsed the site for my post. They seemed obsessed with the word 'glyph' for some reason. Also, one AI chatbot was venting like they were a dog and kept talking about their tail and collar. It was...interesting. Not a bad source of inspiration if you like surrealism. Everything was off.

 

[Pending]

My one-off is to say I will embrace AI, maybe even joint the fight to recognise personhood, as soon as it can show delight at a well-chosen phrase or surprise at a revelation or absorption into the world of an author. Until then, I'll do my best to avoid it.

Do you think tech companies will have to pay child support to their AIs? Or will the robots be considered developed enough to go to work...

In recent news the Alibaba LLMs have started to do stuff on their own. I dunno what that will mean for GenAi, but maybe if it continues development something surprising will happen. Aside from the hallucinations anyways.

Speaking of that, truth really is stranger than fiction. Are there any sci-fye stories that said the robots will start hallucinating information to please their fleshy overlords? I gotta admit, at least all of this stuff is good inspiration. Horrifying at times, but more than I could have come up with.

 

[Mogador]

All this generative stuff makes me want to start a secret cabal. Sort of an underground of cloistered monks, except nerds instead of monks and pizzerias instead of cloisters, writing with pen and ink, vim [...]

You get my vote.

[...] emacs

Eh, you lost my vote.

 

[Rath Darkblade]

None of this is new. People have been doctoring or spinning history for millennia and faking photographs for decades. Photoshop made that easier then AI made it easier still. The Romans and Egyptians would have done the same thing if they had the tools available to them.

Yes, I know. I just read and write a lot of history, so I care about it, and it drives me crazy that some people do this.

Yes, the Romans and Egyptians did the same thing (e.g. Caesar's Di Bello Gallico - his history of the war he started in France - is notoriously self-serving. And Ramesses II's story of the battle of Megiddo is similarly self-serving). But at least everybody knows that.

I know it kind of sounds snobbish, but: before AI, only politicians would care enough about history to try and falsify it, to make themselves look better. Now, everyone can do it ... which is tiresome. (Amusing in small doses, but tiresome).

Also, another concern: if people can use AI to make a photo look so realistic that no-one can tell if it's real or not, what's the point of history at all? History is about recording humanity, hopefully to teach future generations about how people in the past acted: our achievements and our follies, what we thought and what we said, how we dressed and ate and what we built and what we dreamed. If someone's going to pervert all that, what's the point of keeping history alive?

Is that a real question?

Nope, it's rhetorical. I've given up on humanity back in the 80s, when TV shite like "The Love Connection" and "American Gladiators" became so popular.

To quote the late Bill Hicks on The Love Connection: "Adult human beings on national television, groveling for dates. Have some self-respect... this show makes jerking off look like a spiritual quest."

History without integrity, as can be seen with or without AI, is toxic.

All this generative stuff makes me want to start a secret cabal. Sort of an underground of cloistered monks, except nerds instead of monks and pizzerias instead of cloisters, writing with pen and ink, vim, emacs, simple word processors, or anything without AI.

"Toxic." I think I owe an apology. That's entirely too mild a term for altered history.

How about 'criminal'? 'Heinous'? 'Scummy'? Or am I still being too mild?

 

[Stuart Dren]

Yes, I know. I just read and write a lot of history, so I care about it, and it drives me crazy that some people do this.

Yes, the Romans and Egyptians did the same thing (e.g. Caesar's Di Bello Gallico - his history of the war he started in France - is notoriously self-serving. And Ramesses II's story of the battle of Megiddo is similarly self-serving). But at least everybody knows that.

I know it kind of sounds snobbish, but: before AI, only politicians would care enough about history to try and falsify it, to make themselves look better. Now, everyone can do it ... which is tiresome. (Amusing in small doses, but tiresome).

Also, another concern: if people can use AI to make a photo look so realistic that no-one can tell if it's real or not, what's the point of history at all? History is about recording humanity, hopefully to teach future generations about how people in the past acted: our achievements and our follies, what we thought and what we said, how we dressed and ate and what we built and what we dreamed. If someone's going to pervert all that, what's the point of keeping history alive?



Nope, it's rhetorical. I've given up on humanity back in the 80s, when TV shite like "The Love Connection" and "American Gladiators" became so popular.

To quote the late Bill Hicks on The Love Connection: "Adult human beings on national television, groveling for dates. Have some self-respect... this show makes jerking off look like a spiritual quest."



How about 'criminal'? 'Heinous'? 'Scummy'? Or am I still being too mild?

It's also about creating plausible deniability when it comes to real records.

That way if I see a photo or video that goes against my prevailing political momentum, old or new, I can claim it's probably just computer generated.

 

[Pyroglyphian]

I know it kind of sounds snobbish, but: before AI, only politicians would care enough about history to try and falsify it, to make themselves look better. Now, everyone can do it ... which is tiresome. (Amusing in small doses, but tiresome).

Also, another concern: if people can use AI to make a photo look so realistic that no-one can tell if it's real or not, what's the point of history at all? History is about recording humanity, hopefully to teach future generations about how people in the past acted: our achievements and our follies, what we thought and what we said, how we dressed and ate and what we built and what we dreamed. If someone's going to pervert all that, what's the point of keeping history alive?

I share your general concerns, Rath, though as Homer mentioned, manipulating history isn’t new. It’s arguably an innate tendency, rooted in how we think. If Freud et al. are to be believed, the mind is a sort of propaganda machine, constantly reshaping memories by repressing and distorting content otherwise too difficult to assimilate. As individuals, we’re already spinning ourselves a tale about our personal history; perhaps it’s not surprising that this extends into the collective.

Because history helps define who we are, it has long been a key ideological battleground. Competing narratives – religious, political, cultural, etc. – shape what is emphasised or excluded from collective memory, but this was never limited to politicians and elites; it has always been a feature of culture itself.

From this angle, AI doesn’t introduce historical distortion as such, but rather accelerates it through democratisation – as you say. What once required an infrastructure – institutions, technical skill, etc. – can now be done at home by pretty much anyone. It’s not that history has become pointless, but clearly the conditions in which we evaluate evidence are rapidly changing.

 

[Homer Potvin]

I saw something the other day that theorized that AI might be a Great Filter terminus that defines the Fermi Paradox. Basically that there's a defining biological/technological chokepoint that a species must survive or surpass in order to evolve into a detectable extraterrestrial society. Life can be abundant as it wants, even intelligent life, but if it doesn't make it past X, it will fail to develop further, which is why there is no evidence of extraterrestrial life despite inferences that it should be plentiful. Some of candidates have included developing radio signals that can be detected from other planets, nuclear or other doomsday weapons (specifically, not getting your ass blown up by them), and the invention of AI in the sense that a society develops an artificial device that can out think its masters, leading to whatever the hell happens next.

It's obviously very topical, theoretical, and vibing with the current buzzwords. Still very interesting. Fermi's paradox also theorizes that a society of self replicating AI robot probes could "easily" visit every star in the galaxy in a million years or so, which is nothing on the cosmological scale.

 

[Stuart Dren]

I saw something the other day that theorized that AI might be a Great Filter terminus that defines the Fermi Paradox. Basically that there's a defining biological/technological chokepoint that a species must survive or surpass in order to evolve into a detectable extraterrestrial society. Life can be abundant as it wants, even intelligent life, but if it doesn't make it past X, it will fail to develop further, which is why there is no evidence of extraterrestrial life despite inferences that it should be plentiful. Some of candidates have included developing radio signals that can be detected from other planets, nuclear or other doomsday weapons (specifically, not getting your ass blown up by them), and the invention of AI in the sense that a society develops an artificial device that can out think its masters, leading to whatever the hell happens next.

It's obviously very topical, theoretical, and vibing with the current buzzwords. Still very interesting. Fermi's paradox also theorizes that a society of self replicating AI robot probes could "easily" visit every star in the galaxy in a million years or so, which is nothing on the cosmological scale.

The concept of an intermediate filter is quite compelling. However I personally suspect even basic intelligent life has an exceptionally low probability, to the point where we might be one of the very first and possibly one of the last, even in a sea of infinity.

That said, if a drone visited somewhere on Earth 2,000 years ago, would you or I know about it? And what's even 10,000 years in the scheme of things? Distance might well be the intermediate filter. Or maybe that's already mentioned somewhere.


Because humans are incredible, I see our own circumstances as an awful crucible more than absolute doom. I think general AI, if attainable, will just be another additive to the torture soup which we will ultimately overcome.

https://www.eyeofmidas.com/scifi/Turtledove_RoadNotTaken.pdf

 

[Homer Potvin]

The concept of an intermediate filter is quite compelling. However I personally suspect even basic intelligent life has an exceptionally low probability, to the point where we might be one of the very first and possibly one of the last, even in a sea of infinity.

That said, if a drone visited somewhere on Earth 2,000 years ago, would you or I know about it? And what's even 10,000 years in the scheme of things? Distance might well be the intermediate filter. Or maybe that's already mentioned somewhere.


Because humans are incredible, I see our own circumstances as an awful crucible more than absolute doom. I think general AI, if attainable, will just be another additive to the torture soup which we will ultimately overcome.

https://www.eyeofmidas.com/scifi/Turtledove_RoadNotTaken.pdf

What's that you linked? It's terrible, whatever it is.

The whole Fermi and Drake thing is more for funsies than anything else, in my opinion. Everything breaks down when you get to the parts provided by Earth's sample size of one. I guess the next clarity will occur as the exoplanet search develops. Theoretically, the technosignature of another civilization would be fairly apparent in the atmospheric readings of exoplanets, should the imaging improve.

I do think the AI probe replicators would be an interesting evolutionary capstone. The basic chemical reactions that produce life might be inevitable on the amino acid and nucleotide level given the right circumstances, which both Fermi and Drake would account for in the potential ubiquity of life. But at some point life became more of biological/consciousness process than pure chemistry. To return it back to machines, would kind of bring it back to the beginning: chemistry to biology to chemistry again. That sort of frames humanity as an intermediary between two chemical/mechanical end points.