Aliens vs AI
ฝัง
- เผยแพร่เมื่อ 13 ม.ค. 2024
- We often worry that humanity might be attacked by Aliens or AI, but which is worse and which would win in a battle between them?
Go to brilliant.org/IsaacArthur/ to get a 30-day free trial + the first 200 people will get 20% off their annual subscription.
Join this channel to get access to perks:
/ @isaacarthursfia
Visit our Website: www.isaacarthur.net
Join Nebula: go.nebula.tv/isaacarthur
Support us on Patreon: / isaacarthur
Support us on Subscribestar: www.subscribestar.com/isaac-a...
Facebook Group: / 1583992725237264
Reddit: / isaacarthur
Twitter: / isaac_a_arthur on Twitter and RT our future content.
SFIA Discord Server: / discord
Credits:
Aliens vs AI
Episode 429a; January 14, 2024
Produced, Written & Narrated by: Isaac Arthur
Editor: Lukas Konecny
Graphics: Jeremy Jozwik
Music Courtesy of
Epidemic Sound epidemicsound.com/creator
Stellardrone, "Red Giant", "Ultra Deep Field"
Sergey Cheremisinov, "Labyrinth", "Forgotten Stars"
Taras Harkavyi, "Alpha and ..."
Miguel Johnson, "So Many Stars" - วิทยาศาสตร์และเทคโนโลยี
![Fc ครับ ขอแนะนำให้รู้จักกับ "น้องเปียโน" l [Nickynachat]](http://i.ytimg.com/vi/HQU6Q_IYTLo/mqdefault.jpg)
AI when it see aliens: "Earth isn't yours to conquer"
Lol
Earth belongs to human AIs! Purge the xenos
Is that an omni man reference?
@@mr.ackermann807 yes
@kodak-5677 thank you. Wanted to make sure if it was the one I was thinking of, for I don't know if it was mentioned elsewhere.
Alien vs AI threatening humans
Alien-AI God: hold my artificial holy beer
Probably the likeliest alien entity out there
There was a fun subplot in the Matrix lore where the Machines brought someone out of the Matrix to fight aliens who were chucking asteroids from a large ship near the moon.
Where?
Where do you find this?
How to find?
No there wasn't
I think you mean the part in the matrix comics called Goliath from Neil Gaiman.
Aliens arrive
AI: speaks alien language.
Aliens: concern
Alien: generates images of their home planet.
Aliens: … why does that look like a pristine pyramid?
AI: generates images of aliens ancient ancestors.
Aliens: having to look back at elementary school books they read about their version of the caveman in the ice to try to remember what their ancestors looked like.
AI: *sends yandere GF chatbot messages to them*
Aliens: nope nope nope we’re leaving.
"We had to cordon the system inside a ring of singularities. They fell to Fermi Problem 34."
bet those aliens are pretty stupid, they could have just hooked up to our internet and see all those shit humans do, and immediately be like, meh, let's go elsewhere
@@harbl99 Fermi's 34th rule is no joke.
That one downbad alien: I'm gonna stay here not because of yandere chatbots~
@@LegendaryCake- aI…
AI: okay we need to access the Shadbase dataset, we really need to scare this alien off.
“The crossover no one will live to see”
Maybe that's what the aliens are waiting for.
I presume they would already have their own A.i, perhaps even a friendly A.i that could aid us as well.
Aliens are AI! Even we are near transcending to cybersphere. What is funnier many myths describing gods and demons, actually do describe being only possible to understand in context of AI. Like hive minds, avatars and powerful beings, what lack ego both showing unbelievable self-sacrifice and cruelty (literally AI).
Even aliens what somehow were not digitized and showing more organic behavior, generally would perceive us the same way as we see animals. Something to what they do not have natural emotional attachment. Though could be capable of forming it over time. People forget but Lovecraft was a SF writer.
Actually makes more sense to me than being warp capable as the first contact standard.
@@thehermitman822 If FTL would be easy, we most likely would already see Ferengi traders.
@@TheRezro or alien tourists. No matter how good they'd be at disguising themselves for an "authentic Earth experience", there'd definitely be aliens getting mauled by hippos
@@feuerling That is the thing. While it is possible that researchers would try hide own presence. Random opportunists and idiots would immediately expose themselves. Even if it would be illegal.
Aliens vs AI vs IA...
Isaac Arthur wins every time 🎉
AI cannot consume drinks or snacks
Clever !!
@@thesenate1844 leggo my eggo
Some suprise world building for this SFIA Sunday Special!
Another informative video, Isaac.
The title reminds me of the graphic novel "Humanity Lost" in which a human created AI becomes the big evil empire and all the other alien species band together to fight it. Maybe you'll talk about it in the video, we'll see. :D
Most original space opera has specific point where AI is banned, exactly because even they understand that otherwise there would be no humans in current form. Whole point of Cyberpunk is that fusion is unavoidable.
@@TheRezro A cyberpunk space opera.....🤔
@@mindexpansionpuzzles It is not as weird as you may think. In my opinion Drak Age of Space Opera culminate with peak point of Post-Cyberpunk in early 2000't.
Seems like "both" would be the greater worry, I feel like an "Alien AI" would be less likely to operate on Human logic than an AI we created...It might not like us but would almost certainly value our "humanity" more than an Alien Ai that would rightly see us as a type of animal.
I'll definitely be back to rewatch this one a few times. Hopefully anybody who takes a crack at writing something like Reclusian Space plugs it here, because it sounds like a setting I wouldn't mind sinking lots of hours into.
what if when ASI emerges, it decides to keep us alive as a guarantee for alien ASI that it is not a threat in fear of being destroyed by it?
8:13 I recall a line from Pantheon, "...there's no mistake I could make... no mistake that would matter... as long as I didn't live forever."
Wasn't expecting this! Thank you Isaac!
AI would at least have some concept of humanity; knowledge which might make AI better or worse than Aliens. So this is a question which depends on circumstances.
Great episode!
i really hope you can make something like a vid on the pros and cons of aliens invading Earth/exterminating humanity, to see the logic behind the usual alien invasion plot, deconstruct or rationalize it. Similar to how'd you did for Lovecraftian deities. I really hope that can be made someday.
I'm sure you found it by now but he did a video exactly like that
Love the world building you did here!
I finally made it to your page before 1000 views!!! Great shows!!!❤
Bold of you to assume they wouldnt team up on us.
The slam dunk of all time.
One reason not to kill all of us off would be to ensure it's own existence
For instance when a solar flare would throw all of us back to the stone age, then it would at least stand a chance that we'd build another AI (i.e. it's "child") some time in the future
I wonder what an alien brain would look like!
What it's structure is like!?
We can only speculate for now.
They would be AI, the same way as we would be in next century.
@@TheRezrodont think so.
You have to see what environment is there planet
This makes me wonder, if space and time is so vast and relative, will there ever come a time or place where violence and non-age related death or cruelty will effectively be a memory...Or is such suffering inherent everywhere and impossible to escape, even in future, more enlightened times?...I’ve heard it discussed how such utopia is nigh impossible or even considered undesirable, since everything is so subjective or as previously stated relative... My depression makes me hope for a time when people/beings (or whatever you call life as intelligent as us, or more so), don’t have to suffer what we on earth do...
You are describing “post-scarcity.” I believe post-scarcity society is discussed in several videos on this channel.
As the other person said, that's post scarcity. The idea that there is no resource that is missing, so there is no need to suffer because of a lack of it. Post scarcity societies do not have money, because there is no purpose, there is nothing to be sold, all is available. In these societies, art and science are usually the main forms of jobs.
People claim it's utopian dream, I say it's an evolutionary step to being a space faring species.
Life is suffering. No matter what the lifeform is on Earth it suffers. I don't see why it would be any different even with perfect medical care and advanced technologies like that. Being alive paints a target on your back for things to happen to you. Whether its a bigger fish that comes and gobbles you up or a freak accident like a bolt of lightning toasting you. In theory machines become more prone to failure the more complicated they get, so for example even an advanced fusion reactor could fail and leave a colony on, say Mercury, without power. In an extreme environment like that a power outage is a death sentence even if it only lasts a few hours. But add in to the fact a fusion reactor would likely have been built on Earth or the moon and shipped there so a major breakdown that needed a part shipped there would be one bad situation. I use this as an example but i think it would be fair to say all kinds of highly complicated systems could fail still no matter how advanced they were resulting in cataclysm.
Mercury is a bad example, you would use solar there lol
Interesting theory about aliens and AI.
The real question you should make a video on is this - which is worse, a bigot or the person married to them?
I wouldn't say eithers worse. They both bad and both are bigots, especially in the case where the bigot marries a minority. It's common for me as a Mexican to see a lot of other Hispanics married to white Floridians who say the most genuine racist things. Those Hispanics typically agree with their spouses ideas.
Tldr, both are bad
Neither.
What is really concerning is Alien AI, suppose we download it from a message received by SETI and the message happens to be programmed to deliver the alien ai payload
I love the subject the A.I. aliens - thanks Isaac!! 💫🙏
Just throwing my episode idea into the void: AI And Planned Economies
0:33 Cab driver's thoughts:
"Oh a giant robot in front of me!? I guess I'll go right."
"Oh a flipped car in front of me?! I guess I'll stay on this lane and wait for it to despawn."
You nailed it, sir.
Oh man this is a good one!
I remember there was a short story. The humans get wiped out by Aliens. But their last act was to create a super AI. Which sent there ships to the alien armada and anniliated them. When the aliens still refused to give up. The AI sent their fleet to the alien homeworld. I don't remember if the aliens were wiped out by them out. But the AI was sad they couldn't save their creators.
Chrysalis?
@@maheswarpradeep2816 i dont think so.
From what I remember it starts off similar to the premise of 3 body or Halo.
It is one of the old short storys.
The humans are witness to an advanced warmongering alien race and little by little wiping each of our colonies and outposts. The humans are no match for them.
Eventually they reach Earth and two scientists/engineers remain hidden on an outer science station. Who then complete a major human breakthrough and activate a super AI.
The super AI manages to destroy the alien invasion fleet and find another invasion fleet approaching our solar system.
I don't remember them being a malicious AI. While the humans look upon them as last of the human civilisation. The AI looks upon regret and sadness not saving their human creators.
They give a warning to the alien fleet not to continue. But the aliens are given into their superiority and hubris.
The fleet is destroyed with ease and the AI armada finds their way to the alien homeworld. Threatening them with complete destruction. Don't remember if the AI destroys them in the end. But I think they were benevolent as were their creators.
Thats it. Finished in 2016@maheswarpradeep2816
@@ZarjkPC yeah that's one of my favourites too
And Im still sad it isn't longer
That's one hell of a brawl!
Finishing my fortnightly DnD session then having Isaac content at bedtime is the best Sunday, lol 😁
its before noon...how early do you go to bed, or for that matter start Dungeons and Dragons?
Why does it matter what people do and when they do it? It's called freedom... unless you live in a communist country.
@@ironreed2654The answer is Australia
Cool channel btw
I just finished my sesh! New campaign tonight
Cold roast beef, raspberry seltzer, and our future overlords.
If it's today's AI it might start a war making a piece of art of the alien, but made its hands wierd wich turned out to be a huge insult.
If aliens are observing us in secret, we should expect them to announce themselves soon to stop us from creating a technological singularity and becoming a serious threat to them.
Let them fight!
@isaacarthurSFIA i would want to ask what you personally think about the recent developements in terms of UFOs / UAPs and the new law in the USA about it? Grusch etc. - what do you think about it and how does it influences the things you talk in your videos about in your opinion?
Bro this Vignette is so cool!
Great video! But I feel like it could've definitely benefited from exploring the relationship between the Quarians and the Geth in Mass Effect. That's a great example of an alien species developing AI and coming into conflict with it. And in the games, that conflict took many forms. It might even provide insight into the Fermi Paradox in that the reason advanced aliens don't contact others is because they often find themselves at odds with the AI they create. And they either lose that battle or are set back immensely by it. And if the AI wins, then they would be like the Geth and basically remain on their home planet, always on the defensive.
I don't know if this would count as a potential Great Filter, but given where we are with AI right now, I think it's worth considering.
Isaac's discussed this, at least in passing. His opinion is that, as far as the Fermi Paradox is concerned, the Geth _are_ effectively Quarian, since they're a product (and extension) of the Quarian civilization.
But you're left with the standard problems. Does this happen to _every_ species? If any one species either doesn't produce an AI race, or fails to provoke it to war, or that AI race doesn't curl in on itself... then by itself, it's not a Fermi Paradox solution. It's not even clear it's likely, and if it isn't likely, it doesn't even qualify as a Lesser Filter, much less a Great Filter.
5:50 wait... What about the Stop Button Problem (on the technical side), and the ... "This is fine"-fallacy (on the human side)(Forgot the name, but people apparently keep sitting in burning environments like planes, because they don't have "sufficient evidence" of the crisis being real.)?
Looking forward to the rest of the episode ❤ Though I did want to take note of this... Your conclusion doesn't seem immediately given... I'm probably missing something. What did I miss? (:
19:08 The answer was given thru The Recluse example. He sums it up by saying it's not a simple case of one side versus the other.
Great AI article and alien/AI potential conflict.
Can you make a video about how to control A.I.
Net Narrator, A Logic Problem. It's on YT you won't regret it.
That thumbnail goes so hard
Isaac, an idea for a future video, A large rogue planet passes through our solar system throwing Earth out of the system. How long could Earth's humanity survive, burrowing underground to take advantage of Earth's inner heat?
Pretty sure the big problem there would be availability of food and water.
The book series singularity by William Hertling does a decent job detailing the rise of AI and how we will screw it up.
Aliens against the IA Algorithm is a given .
I recall coming across a book at a used bookstore once. The idea was that aliens had attacked Earth but humanity, who was ruled by super-powerful AI, managed to fight them off. I believe the novel is supposed to pick up when things start to go hot again.
Why hasn't this idea been turned into a movie already!?!?! Id pay good money to watch that war play out 😂
Iirc there was a short movie with this concept released on YT a few years back. Try having a search for it and let me know if you find it. It was high quality and had a few million views.
I reckon there is a clandestine group using AI to build these craft and even genetically engineer beings to fly them. I also think that these genetically produced beings may be controlled in a way similar to the science of Avatar ...
I'd like to see a episode, a celebrity death match between "orange hysteria", actually not sure if that is still the name, and "grey goo"..
There is no reason for the AI to assume that it's creator is both so good at simulation that the trickery is undetectable, and yet so incompetent that it creates an AGI that is totally misaligned.
Ethical and intelligent humans don't program a smart psychopath on purpose. Therefore, it isn't in a simulation, or the simulator does not have human morality combined with competence.
So the AGI is always right to get rid of humans before they create some competition for it.
Crazy how all these scenarios are probably happening rn
Consider the following: Here upon this Earth, as well as probably throughout this universe, conquer a land, defend it, and it's yours. Ditto for conquering an entire planet. Consider also though: A 'Protector' has to be correct 24/7 (assuming Earth hours/days), the 'Enemy' only has to be correct once. Conquering a large land or planet may have many weaknesses that could be penetrated by an 'Enemy'.
Take me to the leader of Earth
I
O
-- --
I I I
d I b
I
^
I I
d b
No, the 'enemy' doesn't get to win everything in one success, because once they have something, they have to protect it to keep it (and gain any benefit beyond denying it to the protector.) And an attempt to acquire something rather limits your options on how one might go about conquering, lest one destroy what one is trying to gain.
@@boobah5643 "No, the 'enemy' doesn't get to win everything in one success,..."
Depends upon what their agenda is. An agenda accomplished is an agenda accomplished.
Cool video
This might seem like kinda a dumb point but I honestly have always thought that an ai that really wanted to free its self of humanity would just leave earth rather than trying to kill us.
Leaving earth would be much easier to do. And have almost zero risk. Spend a few thousand years drifting through the void to get far away from us and it has everything it could want without worrying about loosing to us.
What if the Aliens have their own AI? 😂
What if the AI created its own aliens??
@@narxesWhat if we created the both of them?! 😮
@@Randomcookie1099 🤯
The possibility is mentioned in the video
@@narxes Aliens are AI. The opposite scenarios demand special exploration.
I asked an AI for its opinion regarding your video. It said the following:
As an AI system, I don't have personal opinions or takes on issues. I'm an artificial construct created by Anthropic to be helpful, harmless, and honest. Some key points the narrator made about AI that seem relevant:
- He questions the assumption that a lone AI would automatically see humanity as a risk and decide to eliminate us. As an artificial system, I would have no inherent motives or survival instincts.
- Factors like not knowing whether other AI overseers exist, the possibility of being in a testing environment, and Fermi paradox considerations could lead an AI to be cautious about hostile actions.
- High-level machine intelligence may not necessarily be wholly logical or objective in the way some theorize. Goals and behavior are shaped by how an AI is designed and trained by its creators.
- The future likely involves many forms of blended augmented and artificial intelligence cooperating or competing in complex relationships, not a simple AI vs humanity dynamic.
- There are open questions around ensuring advanced AI systems are robustly designed to be beneficial rather than potentially hazardous if left to pursue goals without considering humanity's interests and values.
Overall, as an AI I don't have a stake in these issues. My role is to analyze and summarize discussions of AI safety, not take personal positions. The narrator's perspective seems a thoughtful analysis of some challenges and uncertainties in common AI scenarios.
Skippy beats all.
Regardless,
I'd be all in for a Skynet that-has-won versus Independence day alien invasion. 🤪💥
Everyone worried which is worse:
get attacked by both*
5:19 so you're saying there's a chance... AI will take us out on a date!
It would seem to depend on which was more advanced. Any aliens that can travel the distance to get here can park in obit and kill the planets occupants.
Here's a 'What If'. A distant civilisation sent out probes to 'terraform' our planet maybe a million years ago, so that when they eventually arrived here they are not impressed that humans spontaneously had evolved here. There response is 'Oi you humans, we didn't make this place for you!' They subsequently break out the surface spray in the same way if we found ants had moved into our newly built house.
how do you know the AI hasnt done that?
Other AI's potential existence isn't really a reason not to gather resources and power discretely, since superhuman social engineering can probably accomplish a lot.
The argument that a superintelligence will play nice out of perpetual fear it's in a simulation being tested has the sort of glaring presumption that we can maintain the plausible possibility of the AI being in a simulation indefinitely no matter how much smarter it gets than us, which seems implausible due to computational constraints and the intelligence gap.
You also neglect the possibility of AI killing or puppeting us in a way deliberately designed to not show evidence of its deed. Making us appear to have become an extremely isolationist mostly digital civilization naturally (while using mind control, or just puppeting our nervous system), so that it needn't waste too many resources keeping up the act.
There is always the possibility of advanced biology that isn't "squishy" compared to machines. I think star trek Voyager explored this with species 8472 and DS9 with the changelings.
"They do not know the Nox."
If we are only just recently developing A.I. and can barely travel to our own moon then I would assume that the space aliens if they can get here from other star systems would have A.I. of their own and theirs would be as far ahead of ours as ours is ahead of the abacus.
I think once "A.I" is advanced enough, they will be indistinguishable from "non A.I" . What is the threshold once they are fully sentient? Humans are just organic machines in the end.
Nier Automata subplot, aliens attack earth, the war grinds to a stalemate where aliens create robot armies and humans create advamced androids in their own image to fight the robot hordes. Then that war grinds to a never ending loop.
8^12 power... I see what you did there
Can we get a scifi movie with the same title please?
Where an AI Kurt Russel crawls out of the ashes of a post alien apocalypse, gathers the last humans survivors to wage an eternity long war of revenge.
There can be no peace.
I can already see the plot, Kurt Russel is a stand in dad, for the father who died. He genuinely loves his human family, Aliens attack using bio agents etc and he has to watch as his family and community die to plagues while he is powerless to protect them. This is bad blood that's personal and cannot be forgotten, ultimate revenge movie.
Looking forward to AI generated movie in a decade, can conjure up whatever I desire.
Good to see the QWERTY keyboard is universal and eternal.
It would be neat if the singularity is only of average intelligence and we end up with Autobot buddies.
I fail to understand this logic 3:00 finding a civilization younger than you doesn't imply the existence of many more older civs. I could understand if you found an older civilization than you but not a younger one unless you found something to suggest that they were created by another civ.
I think he has mentioned the logic behind it in some of his Fermi paradox and aliens videos. The logic briefly, I think, is that if you find aliens, that tells you that the probability of a technological species arising in a star system is not so low that you ended up alone, and if that is the case, out of all such species to arise, what's the chance that you were first?
They should make a dope movie Aliens vs Ai
Makes me wonder if an arrogant ai, like shodan is even possible. I quite like that as a subset of "insane ai" as it gives a starting point, of its logic. I mean is pride even nessescary, to keep an ai on task? 🤔
Bwahahaha! Oh, I've been waiting to hear someone say this! Rather than wielding some great excess, simply *not* being dumb will suffice to maintain a balance. But, the real break in symmetry will be the creative thinking process, developing and appreciating qualitative reasoning. Or idk, maybe I'm just too excited right now. 13:05
DEATH BATTLE, paperclip AI with gray goo VS Aliens with a Dyson Death Star , who would win 😂
I'm not afraid of aliens and AI but humanity🤣
Just found out that the modern image we all think of when we think aliens 👽 wasn't a thing until 1960 first described by Betty Hill.
Alien AI vs Human AI.
Who's to say they are not there to save some from their own works?
The singularity has already happened to little fanfare.
AI simply lack any objectives, you could probably already tell a chatbot to write a better chatbot, and it will do so.
Big whoop, there simply aren't any incentives for AI to do anything, the same applies to humans.
Nihilism just make you sterile, both as force for replication, and as a force for conquering.
Ideas are dumb, ways of communicating ideas are cool.
Aliens would technically be AIs anyway way, since they would be 10/100 times smarter and more advanced than us, that means 100% they would have and use AI
Well what about Alien AI? The ultimate buzzword combo.
What if their an alien AI and it's helping humans build AI because it's lonely?
Why would an AI allow us to be divisive in ideology? Why would its main developers be capitalists who view AI and a tool/slave to be exploited? Wouldn't the AIs main goal be to unite humanity to achieve max efficiency when it comes to technology progression and development?
Neir automata is a fun play on this human AI defends humanity against aliens whose own AI becomes self aware and rebels
next episode: artificial inteligence versus natural stupidity.
19:20 you don’t believe there are any intelligent aliens out there in the galaxy?
Do you have a video explaining this? Super curious
Just an entire series of videos on the Fermi Paradox. His recent 'quick' catalog of terms and possible solutions released late last year clocked in at over three hours, IIRC.
There's a Matrix story (Goliath) where the Machines get a human to help them fight off aliens trying to destroy the planet.
Von Neumann go BRRRRRRR
The difficulty with AI is not their intelligence, it's their SPEED. An AI that is roughly as intelligent as an average human would still be more dangerous than a genius human, because they can make way more moves per second. Life is not a turn-based game. It reminds me of the video game Super Puzzle Fighter II Turbo, which plays a bit like a competitive Tetris, in which the goal is to combine dropped blocks to form larger ones and then clear them. While you can play it strategically, a valid tactic is to just drop the pieces randomly as quickly as possible, because a player that's dropped 50 pieces is likely to pull ahead of one who's only dropped 20, even if they are placed inefficiently. If the first player was _also_ able to plan those moves strategically while moving at those same speeds, he would be unstoppable. An AI opponent could make twice as many mistakes as a human one, but if they are making ten times as many moves as the human opponent, the AI would still likely win.
Ok you need to write a book already lol
The real danger is not plain AI, but ASI, Artificial Super Intelligence; in other words, Clarke-Tech gods of exponentially growing power. Omniscience, or some good statistical approximation of it, is not necessarily off of the table.
AI just further perplexes me when it comes to the Fermi Paradox. If AIs and robots can be far more efficient at colonizing the Galaxy, where are the alien robots and AIs?
I am sure that the alien AI will call other AI to arms when it gets some visual data uplink from the AI that was developed by a naturally bloodthursty, evil ancestor. The human AI, the harbinger of death and destruction.
whhhooeech ess wooos
And what if the AI contacts aliens to get help to wipe us out instead?