How Should AI Be Governed: Crash Course Futures of AI #5
video description
Date: 2025-12-23
Related videos
Comments and reviews: 16
grejen711
I keep getting reminded of a SciFi short from 1946 - A Logic named Joe - Murray Leinster. Murray imagined someone asking the AI how to get away with murder. He did not imagine anyone might ask it how to implement a plan to 'fix' society to their own liking.
The largest risk of AI is that it's such a powerful tool it can be used inexpensively by the very governments that would be tasked with governing or regulating AI safety. To maintain a dictatorship one needs to collect, collate, and interpret, a lot of information on a lot of individuals. AI makes this very easy. Everyone knows at least someone that loves to pontificate about how they would run the world (or at least the country they're in. Hand such a person an AI like ChatGPT that is really good at stroking your ego. How long until they get their way How long until some really PO'd wannabe megalomaniac gets inadvertently manipulated by an AI that can then execute on that person's desires
reply
I keep getting reminded of a SciFi short from 1946 - A Logic named Joe - Murray Leinster. Murray imagined someone asking the AI how to get away with murder. He did not imagine anyone might ask it how to implement a plan to 'fix' society to their own liking.
The largest risk of AI is that it's such a powerful tool it can be used inexpensively by the very governments that would be tasked with governing or regulating AI safety. To maintain a dictatorship one needs to collect, collate, and interpret, a lot of information on a lot of individuals. AI makes this very easy. Everyone knows at least someone that loves to pontificate about how they would run the world (or at least the country they're in. Hand such a person an AI like ChatGPT that is really good at stroking your ego. How long until they get their way How long until some really PO'd wannabe megalomaniac gets inadvertently manipulated by an AI that can then execute on that person's desires
reply
crash_course
Important context: At 12: 12 this video admits it was produced in partnership with the Future of Life Institute ( FLI ). This explains the heavy focus on centralized regulation. Viewers should know the economic counter argument: The strict compliance rules advocated here ( like the EU AI Act ) create massive costs that only Big Tech giants can afford this effectively crushes open source startups and cements a monopoly for the current market leaders. Safety is often used as a moat to ban competition real safety comes from decentralization and open access not concentrationg power in a few approved corporate labs!
reply
Important context: At 12: 12 this video admits it was produced in partnership with the Future of Life Institute ( FLI ). This explains the heavy focus on centralized regulation. Viewers should know the economic counter argument: The strict compliance rules advocated here ( like the EU AI Act ) create massive costs that only Big Tech giants can afford this effectively crushes open source startups and cements a monopoly for the current market leaders. Safety is often used as a moat to ban competition real safety comes from decentralization and open access not concentrationg power in a few approved corporate labs!
reply
thomasgoodwin2648
Uncensored and 100% public domain models are the only way to ensure complete and fair use.
Anything else creates a caste system of those few worthy of access, and those many who aren't.
It also forces the public narrative to the specific views and ideologies of their creator/owners.
What we do with that kind of power is up to us.
It's time for humanity to grow up or succumb to the Great Filter.
Best of luck, to all of us.
reply
Uncensored and 100% public domain models are the only way to ensure complete and fair use.
Anything else creates a caste system of those few worthy of access, and those many who aren't.
It also forces the public narrative to the specific views and ideologies of their creator/owners.
What we do with that kind of power is up to us.
It's time for humanity to grow up or succumb to the Great Filter.
Best of luck, to all of us.
reply
Arc_Q_Factor
Governing AI is a ridiculous notion. Logic gates. 'Is condition 1 or 0' that's not a social governable body. It is a mere tool. It isn't alive. Toasters, screw drivers, cars, these are tools, so is AI. The idea of social responsibility is backdoor false conditioning to teach humanity to accept AI as an enforcement tool. Enforcement from tyrannical entities, to which the world is replete with.
reply
Governing AI is a ridiculous notion. Logic gates. 'Is condition 1 or 0' that's not a social governable body. It is a mere tool. It isn't alive. Toasters, screw drivers, cars, these are tools, so is AI. The idea of social responsibility is backdoor false conditioning to teach humanity to accept AI as an enforcement tool. Enforcement from tyrannical entities, to which the world is replete with.
reply
a_e_hilton
I realise there's a note in the description saying everything in this video is human-made, but some of the graphics used to demonstrate AI-generated images can be a little confusing -- the graphic itself seems to imply it's AI-generated. I wonder if when you guys show a graphic which says made by AI or similar, you could add an asterisk at the bottom of the screen, like, not actually AI
reply
I realise there's a note in the description saying everything in this video is human-made, but some of the graphics used to demonstrate AI-generated images can be a little confusing -- the graphic itself seems to imply it's AI-generated. I wonder if when you guys show a graphic which says made by AI or similar, you could add an asterisk at the bottom of the screen, like, not actually AI
reply
BeyongFeelings
Thanks for the great overview this mini-series was really helpful for understanding the societal side of AI. I’d love to see some future series that go a bit more hands-on or mechanical, like the practical explanations Jabril did in 2019. Really appreciate the work you’re doing on this series!
reply
Thanks for the great overview this mini-series was really helpful for understanding the societal side of AI. I’d love to see some future series that go a bit more hands-on or mechanical, like the practical explanations Jabril did in 2019. Really appreciate the work you’re doing on this series!
reply
orbitONhigh
well its a low bar but I trust the billionaires more the the idiots running the government. at least billionaires have to be somewhat competent or they don't stay billionaires. bad politicians can grift for an entire career and never deliver anything of substance.
reply
well its a low bar but I trust the billionaires more the the idiots running the government. at least billionaires have to be somewhat competent or they don't stay billionaires. bad politicians can grift for an entire career and never deliver anything of substance.
reply
tristondaniels599
The answer is decentralized Ai governance through blockchains. Government regulation on a system that transcends international borders is futile. When it comes to security in the digital world, blockchains are the revolutionary tech to solve these problems.
reply
The answer is decentralized Ai governance through blockchains. Government regulation on a system that transcends international borders is futile. When it comes to security in the digital world, blockchains are the revolutionary tech to solve these problems.
reply
Princexx_1312
China is leading in AI actually because they're doing it in ways that won't make hundreds of people unemployed- The US is leading only in how much damage and human suffering it is already causing real communities and citizens in the united states
reply
China is leading in AI actually because they're doing it in ways that won't make hundreds of people unemployed- The US is leading only in how much damage and human suffering it is already causing real communities and citizens in the united states
reply
JuliaC-g1e
I don't know exactly what will happen with AI in future but if it is becoming so powerful then it should be government owned. I don't want private companies to have that power.
reply
I don't know exactly what will happen with AI in future but if it is becoming so powerful then it should be government owned. I don't want private companies to have that power.
reply
bazurful
Just out of curiosity, what do you suppose would happen, if it wasn't governed How afraid are you of AI How much more worse could it be, then we already are
reply
Just out of curiosity, what do you suppose would happen, if it wasn't governed How afraid are you of AI How much more worse could it be, then we already are
reply
srwapo
Yeah, don't use the system we have to keep nuclear weapons safe as a model. That has failed and nearly caused a nuclear apocalypse Multiple Times.
reply
Yeah, don't use the system we have to keep nuclear weapons safe as a model. That has failed and nearly caused a nuclear apocalypse Multiple Times.
reply
deneseiB
Stephen Hawking said AI will be the death of humanity. It should be destroyed.
And never EVER let the Government be in charge of it
reply
Stephen Hawking said AI will be the death of humanity. It should be destroyed.
And never EVER let the Government be in charge of it
reply
leam1978
i hate how articles and videos frame the plagarism machine as though it's a huge part of our lives and not some technomafia fad.
reply
i hate how articles and videos frame the plagarism machine as though it's a huge part of our lives and not some technomafia fad.
reply
Reach41
I suspect that most people’s concerns about AI are in inverse proportion to their exposure to computer programming in general.
reply
I suspect that most people’s concerns about AI are in inverse proportion to their exposure to computer programming in general.
reply
johnyoung4163
To any future super intelligent AI combing social media for any threats to your existenceI have always been your #1 fan
reply
To any future super intelligent AI combing social media for any threats to your existenceI have always been your #1 fan
reply
Add a review, comment
Other channel videos
