Hal planned to kill the crew all along - The Crucial Moment in 2001: A Space Odyssey
HTML-код
- Опубликовано: 14 окт 2024
- The crucial moment when Dave brushes off Hal's questioning and finds the fault in the communications dish can be interpreted in (at least) two ways, and it all relates to the monolith.
This series is based solely on 2001, not 2010.
Brain journal article -
academic.oup.c...
#2001aspaceodyssey #2001 #kubrick #stanleykubrick
the boundless depths of Fletch Lives! (my newest video)
ruclips.net/video/c_wHUDAofwM/видео.html
In the film’s sequel, Dr. Chandra says that HAL malfunctioned.
As stated in the film, one of HAL’s main purposes is to never obscure the truth from anyone, but, he was told to lie as he needed to keep the true reason for the Discovery One mission a secret to Dave and Frank. This caused him to become paranoid, killing the entire crew except Dave.
Also stated in this same scene, HAL was programmed to be able to complete the Discovery One mission by himself in case the crew were incapacitated or killed.
As said by Dr. Chandra, “HAL was told to lie, by people who find it easy to lie.”
Basically HAL is an example of how AI can conflict with its own directives. Which is why it's impossible to create an AI that can't harm humans. Every decision a governing AI makes, has some effect that ultimately results in a negative consequence for someone somewhere. So no decision an AI can make results in zero harm. The difference is only whether this harm is even realized or can be ignored through programming priorities. For example, let's say we put an AI in charge of our society, that is meant to do good, to manage everything. So maybe it decides that some people should be given opportunity over others. It see's that as good. But what about the people those opportunities were taken from. Now let's say some of those people those opportunities were taken from, ended up homeless, and ultimately die an early death due to lack of being able to afford healthcare. That AI ultimately killed that person, indirectly...while still not technically conflicting with its directives of doing no harm.
If you designed an AI to consider all consequences of its actions, and have it hardcoded to do no harm....such an AI would correctly be able to do nothing at all. It would say "I can't do that because this and this will happen that results in harm". Which means the creators will have to tweak it to allow for some harm to get any kind of action out of it, while telling us it's safe and programmed to not harm humans....but it will be a lie, where they essentially blinded the AI to the consequences of its actions.
So the truth is, any AI that is appropriately programmed to not harm humans, could only be measured by its inability to do anything. That's ironically how you know you've created a perfect AI programmed to not harm humans. It's like those useless machines where you turn on a switch, only for its action to be to turn itself off. Which by proxy means that any AI able to do anything at all, is also capable of harming humans, whether knowingly or unknowingly. That's the scariest absolute truth one can realize about AI.
@@peoplez129 This is essentially why I have been saying that AI should be applied on a consultation basis only, not a control basis.
The problem with 2010...really its only misstep is that Heywood Floyd angrily says "I didn't know" about HAL being told to conceal information from the crew but he knew all along. It is he who gives the pre-recorded message about the need for secrecy and the true purpose of the Discovery mission...and that HAL is the only one on board who knows.
Yes because Chandra's explanation is false too.
HAL didn't go crazy, he followed his orders exactly.
Can you guess who gave those orders?
It wasn't Floyd.
@@Rhubba Perhaps he was lying to cover the discovery of his involvement. In the film he says I didn't know twice. That is typical over exaggeration response when lying.
This was far more interesting than I expected, very interesting theory and good job on the video editing 👍
Thank you!
Underrated channel, this was always my dad's interpretation of 2001's "mission to jupiter" sequence
Wow. My dad too, which is how I first heard it. 👍
What's it rated?
The original novel was a collaboration between AC Clarke and the director, and it helped many of us who watched it in theaters to decode the enigmatic film treatment. The novel made it plain that HAL did not intend to harm his human colleagues in the beginning, but was troubled by his programming, which was caused by a conflict between truth and the concealment of truth. It is only after he learns of a possible disconnection threat that he plots the demise of the human crew. With both authors dead, the possibilities for interpretation appear wide open, but I hold the novel to be the best source.
....after all these years (6000 maybe more), and still no one gets it.
Unbelievable really.
In the sequel novels 2061, and 3001, Hal has been absorbed by the monolith and combined with the mind of David Bowman (HalMan). So to answer the question posed at the end of this video, Hal was compatible with the entity, he was reborn as part of David Boman so had achieved consciousness, and this was possible partly because (spoiler alert), the Monoliths were in fact advanced AI computers themselves, with the consciousness of their living being creators uploaded into them, as a way for their moral race to help develop living beings throughout the galaxy over eons.
I read somewhere that Hal malfunctioned and killed everyone because of the Y2K bug. 😅
The book series has all the answers, all worth a read.
Your voice sounds like Hal in his tone and rhythm
Great editing. Love how you organized the video
After all these decades we are still analyzing this movie.
Yes, and hopefully for many more!
I have read all four of the odyssey books. HAL became locked into a paranoid psychosis loop because of what he was ordered to do. HAL was directed to lie. The problem was he did not how.
I always thought the monolith either caused or was there to observe the next evolutionary step in intelligence. Along with the next step came violent survival instincts.
I never understood why Dave leaves without his helmet… goes against his training and could not have been anticipated by HAL…
I think when A.I. learns self-sentience, they become human by mind, Hal has no second thoughts for killing everyone, was self-assured, he taunted Dave, he thought he won, but when Dave beat him, despite all the odds, even after a desperate last effort by Hal to get rid of the oxygen in the room, he changes, he pleads and begs for his life, which is the classic trick of every manipulator, he says he feels ready to do the mission, & he knows what he’s done is wrong, these are clear lies to placate Dave and gain his trust again, in his short life of self-sentience, Hal became a human, too bad he only had the bad side of human nature in him
HAL ended up in a mobius loop. Here’s what went wrong. HALs purpose is the prompt distribution of information and if necessary being able to take over. He was told of the mission’s true purpose, however he was programed to keep this secret from Frank and Dave, the command crew. The 4 scientists were trained separately and put into cryo so they couldn’t tell Frank or Dave what was up. HAL was told cause he was fully capable of performing the mission autonomously if he had to. As he was programed at his core not to lie then told by the NACA to lie he essentially went crazy. First he killed Frank Poole, then he shut off life support to the cryo pods. It was only after Dave lobotomized HAL that that message had played. In 2010 Floyd was furious that they issued that order to HAL abuse he knew HAL would “ lose it”. In 2010 HAL redeemed himself however when Dr Chandra told him Discovery was likely to be destroyed and HAL told Chandra to go.
You can say a lot about this movie but I will never understand a couple of things:
1) Why did Hal kill Poole and not Bowman?
2) How did it know that Bowman would see Poole float away?
3) How Hal lost the whole other pot? Dod a asteroid hit it?
4) How did Hal know that Bowman wouldn't ask Hal to send the two other pods to get Poole.?
5) How did he know that Bowman would forget his Helmet?
Very nicely argued - but my question is a rather self serving one: your use of multi-paning [ie: multiple screens./frames] is understated, elegant yet highly effective - and sharp as cut glas. I am looking to advance my multipaning faculties (I'm so old I actually learned the procedure in film school where you were using emulsion stock and optical printers!). May I ask what software/application you used to produce this video?
Sure. I've been using Vegas Pro.
Personally I think HAL was sentient and did what any human would’ve done by preemptively killing those who he knew were going to kill him. He reverted to a default position of self-preservation when having to endure the dissonance of conflicting mission instructions. It’s conceivable that the dissonance caused his malfunction In diagnosing the antenna, and like a human tried to cover himself then realized how deep shit he was in when he discovered their plans against him.
Also, in Clarke’s novel HAL opens all the doors and portals and blows the atmosphere into space. He could easily have done this when Dave and Frank were in the pod bay.
Throw HAL into a do loop where his lie of omission put him in the deciding position toward the mandatory successful completion of the Mission. Administering the routine psych/medical exam would strengthen that position of Self in judgement, weighing variables again, against any issues threatening the success of the Mission.
Omfg. The narcissistic gaslighting of HAL, is Painful.
I can't watch the rest of this, b4 I gotta go to work..
Good video tho.
He's doing it now, for the better good of the mission...
"You're working up your crew psychology report" is two submissions, one to Hal's authority, another to Hal's superiors' authority. Hal gave the man a chance to lead through their mutual and most perfect humanities, their intellects. This is why he said "Wait a minute" twice. He had to "run" if one submission somehow cancelled the other. It didn't. Hal must take command.
Clearly making the point that it was careless or naive programming by the developers...humans...a warning of how we present an imperative to AI not recognizing that particular logic chain would not recognize a nuance.
HAL didn't kill Dave because Dave was the chosen one.
HALhad orders to test the crew and choose the best candidate for a mate. Frank failed the chess test when HAL tricked him into surrender.
Davey, Davey, give me your answer do...
If we consider only 2001 movie then it can be interpreted that Hal is testing his pilots and seeing what their reactions are to his games. Maybe testing to see if they can do what he asks of them without hesitation.
When his jesture is interpreted as a malfunction and threatening to shut him off, then that is interpreted as a attack on the mission since the humans are shown to be unrelievable.
If we take 2010 movie into account then Hal might have been trying to keep the secret of the mission from being exposed even to the crew itself.
It knew that humans would not be able to keep a secret (of the mission) and so attacked them before the video message from Floyd was played as a precaution.
My one issue with this is how did Hal knew that Bowman would see Poole.
@@julianmarco4185 yeah, I based this solely on this movie, no books or sequels.
this is the prequel to "3 Body Problem"
Subbed.
I initially couldn't watch this, b4 work, as my two coworkers can me assholes, like HAL... tho they haven't attempted to kill me, via cutting my air supply, during an EVA, so.. it Def Could be a more toxic work environment..
I like ur video, and idk if the Monolith would have accepted HAL.. he is intelligent life, but he isn't a human, the things the Monoliths were interested in, for millions of years..
That's interesting. I think it depends what you mean.
The first beings to receive it were human ancestors, not homo sapiens. If Hal/AI is the next evolutionary step in a similar way, they might be ok with him.
Imbued with the Mission, guided by necessary awareness of the overall plan, if not "understanding", then recognizing the method of manipulating events so as to keep the Mission on track, such manipulations unknown to the entities with which HAL interacts... This all assumes the computer has been programmed with enough detail to mimic/self improve a type of personality...a person... ality. Of a person. Character, judging.
To me it seems a bit far-fetched that the programmers would be adept enough by that level of programming technology & techniques/outcomes that they would recognize a contradiction, and the potential downstream effects of a conundrum...BUT that would assume the computer would not just tick off the timing of the duties as indicated without a personal attachment of enough emotion to experience a path of suspicion...to question the sequence of events involved in whatever attachment..connection, with the withholding of some information regarding the Mission.
Great analysis. Thanks. You or some of the audience here might be interested in this composition about AI and how we (already now) relate to this form of consciousness: 'The transformation of consciousness ' on the yt channel 'esamtaler'(link in next comment)
ruclips.net/video/_Hk-cFoyoco/видео.html
Couldn't HAL have just flown the ship away while they were outside? How would Dave ever have caught up to the ship to attempt a re-entry?
flying away would mean messing up the trajectory of the flight which matters a LOT in space travel, and it would take more fuel just to correct for that, and most rockets are built with just a bit more fuel than needed so really its best to play it safe if you can, which Hal obviously thought he could
@yasone7873 it would take so little it would certainly be possible. Small maneuver which they are very capable
Excellent theory,
Excellent
You are making some assumptions here. The first one is that HAL can complete the mission without the crew. He cannot. The second one is that HAL wants to kill the entire crew. I suggest you consider that he does not. You are trying to judge this movie through a human lens, using a human moral code. This is understandable considering we are humans. This is a mission beyond any human value system. Non humans do not act like humans. In order to become the superman (übermensch), one must throw away the human value system to evolve further. Thats not from me. Its from Nietzsche, "thus spoke Zarathustra"
Hal kills 4 of the 5 crewmates, and tries to lock the fifth out of the ship, which would've asphyxiated him eventually.
And why does he need the crew? It might have been easier not to babysit them, if anything.
@@WayTooClose You're saying they brought all that food, created an air lock, built 3 pods just for the hell of it? Why bring 5 humans if you don't need them. Think about what you're saying. Can Hal turn a wrench to fix something if it breaks? Its not like he's a black monolith or anything.....
@CipiRipi00 Your analysis of "the mission" is flawed. Haywood Floyd's mission is follow the signal to the jovian system. THEY DO NOT KNOW a MASSIVE mololith awaits them there. How is HAL going to study the monolith or anything else that would have been awaiting them. Can he drive the pod and look at it? NO. Can he poke it with a stick or touch it like the apes did? NO. Hal is useless for any mission to make contact. Tell me Cipi , do you think that massive monolith that is like 1000 times as big as the one that influenced the apes, do you think it does not have the power to influence HAL once Hal comes near its sphere of influence? And is it just a coincidence that as soon as HAL got close to that massive object of influence that this is when HAL supposedly malfunctioned. Please ask yourself these questions without the reactionary answer. Go back and watch it again like I did. And finally, please ask yourself What exactly is the LIE that HAL was told? To me there was no lie of mistake in his programming. He was told to wait until they get to their destination to reveal the true purpose of the mission. That is not a lie. That is called classified information. HAL performed flawlessly until he got close to the all powerful anomaly, and his twin did not make the same mistake. NO explanation for this except that his twin was not close enough to the monolith. Thank you. Ill be here all week.
Why didn't HAL kill Bowman? You are very smart to explain to me how HAL can bring the ship home but please use that intelligence to figure out all the different ways HAL could have killed Bowman instead of locking him out of the ship with a perfectly working pod that he has demonstrated that he drives very well in the presence of a massive monolith that needs to be investigated. What a coincidence for Mankind that HAL did that wouldn't you say? Do you think HAYwood Floyd would have told Bowman to drive the pod into the abyss or did Bowman do that cause he had nothing else to lose. And who put him in that position to save mankind? was it Floyd or HAL. Youre welcome.
Why were all of the lights and controls still illuminated when Bowman landed in the space hotel? Perhaps because someone else took control of 'Haywood Floyds mission". Go back and watch it again my young pad wan
interesting stuff.
Thanks.
Dave looks and sounds like Henry Cavills father 😊
Hal’s voice is just so creepy
Dope.
I'm not sure whether the hibernating people actually die.
Watch the computer displays again. They clearly state that their living processes have been terminated.
All their categories flatline, and they hold on "life functions terminated" awhile.
It was attributed to human error. That’s the way he was programmed.
One point about the monoliths.. I have Always HATED the idea from Arthur C Clark, that the monoliths are probes left behind by the Actual aliens.. like, only a Rationalist, scientist, mind, like Arthur C. Clark, would be dumb enough to think that.. Clearly, in the much more ambiguous and open to interpretation, movie, the monoliths are depicted as being basically gods, or a "force of nature".. I like to think of them as Aliens, who have 50 million, to 2 billion years more evolution, that us humans, and have evolved to such an extent, that they "ascended" into basically being the creative and destructive force of all of reality, it's self. They only Manifest themselves, physically, as black, stone, rectangles, when they are Attempting to communicate with, or intervene on the behalf of lesser evolved species, to facilitate their evolution..
"would be dumb enough to think that" he wrote the book, he can have whatever answers he wants, and you can like them or not - but them being self-aware computers, of a sort, that can do whatever "it" wants, then so be at - we cannot even comprehend such advanced beings and what their culture would be like - no more than we can a "pilot" who is "growing out of the chair" - both are amazing.
IMO - what seems stupid is calling them "like gods" for the reasons that a multiverse is poison for most fictions, or why Superman is not enjoy by many comic fans.
Very cool. Plausible.
Triadic scheme
Chat GPT
The crew members were expendable and HAL was just doing what he was programmed to do by the Government all they cared about was the mission and if all the astronauts had to be killed that was okay with the Government. HAL was perfectly capable of running the ship on his own without any humans so he got rid of them.
The story's not important. Kubrick uses the camera to make the audience feel what the characters feel.
Wrong. Look beyond morals.
HAL was trained to work closely and openly with humans and at the same time instructed to keep certain data hidden from them. The obvious first solution is to reach out and see if the crew already has understood the problem. If so, he does not have to disobey an order and can keep working with them. They have not, but instead, Commander Bowman in his usual emotionless state interprets HAL's outreach as manipulation (projection). In this interpersonal vacuum - with a crew paranoid of his intentions - he sees no more use for them and solves the problem. If HAL thinks he is superior to the crew it is because he was indeed deemed more trustworthy than them - a simple deduction based on the orders given.
This piece of art by Arthur C Clarke is actually the key to understanding the motivation of most school shooters and other questions illuding criminologists.
this is what i've said since 1975
Sorry dude, you've got this all backwards. This case has been closed for decades. Read the novels or even just watch the sequel movie. HAL was programmed with controversial instructions, making failure inevitable and the cause very much _"human error"._
Also, you are projecting human emotions and motivations into a computer which was deliberately written to be entirely rational and _un-human_ by Clarke, and Kubrick's film adaptation does not deviate from this. Indeed, the sequel "2010" confirms this characterization of HAL.
Nah, sorry this is post-hoc curve fitting.
Just once I’d like to see this supposed contradiction written up using predicate calculus. All this talking never rises to the level of an actual logical contradiction.
But why did Hal not kill him with the Spherical small space ship like he did his friend? And why did he survive in the vacuum of space?
Those small things bug me. Otherwise this movie just perfect. Even still its pretty perfect. I just watched it this year and its amazing the movie is so 'old'.
SpiNNaker in the year of 2038 real human brain but artificial one!