AI companies, basically:

- mining asteroids will make us rich, let's bring them to earth as quickly as possible - we promise to figure out a solution to the "collision problem" (right now we don't have one) - we can bring a massive asteroid here in less than 10 years - give us ten trillion dollars to do this - we will lobby against any government effort to stop us

AI safety:

Don't let asteroid magnet companies threaten your life. Here are some things you can do to stop them: 1) Sign this letter calling for a ban on superintelligent AI

2) Contact your political representatives to tell them you want them to take action on this.

3) Join your local PauseAI chapter. Go to protests, get organised, and build the movement.

4) Read this book. Understand why we're not ready to build superintelligent AI. Then give the book to a mate.

5) Don't look away. There are a million other things happening right now. You're busy. But this is probably the biggest threat to your life in the near future. If we all stay on the ball, and continue to build momentum for an international treaty, we can avoid disaster.

Video is of course inspired by Open Asteroid Impact!

For perspective: The theory that predicts ASI killing everyone, successfully predicted (~decades ahead) things that AIs already did or tried to do: - escape confinement and lie about it - subvert safety protocols placed upon it - hack its reward - lie despite being told to not do so - acquire resources that hasn't been granted - sandbag and pretend its stupider than it really is - blackmail CEO to avoid its shutdown - attempt to cut off oxygen in order to murder a human in simulated test The theory is very likely correct. If we build it all life on Earth will die.

Brother!!! This is an excellent message!!! Such a perfect analogy. Every. Single. Human. Must hear this or something this obvious describing this situation. And there must be some way every single person can get on board with protecting ourselves.

Yes. One simple thing everyone can do right now is signing this letter calling for a ban on superintelligence. If you want to do more, consider getting involved with your local PauseAI chapter!

I think the highest probability outcome of a benign scenario is one where we have a unified understanding surrounding our future and AI What you're doing here is the best way to transmit that message to the masses Momentum (before something really bad happens) is the best way👍

Isn’t this literally the plot of “don’t look up”

@HumanHarlan I think they plan to use materials mined from the asteroid as it approaches the Earth to build tech to solve the "collision problem", which will make it much more solvable, potentially even in time to prevent the actual collision!

wasn't the 'rare metals we rich ' the reason the world exploded in 'Don't Look Up?'

Don't Look Up if it was a tiktok filmed in a bathroom.

Great video Tom!

Thanks Andrea!

i hope the bubble bursts after the agi race is over

@justthisguy

my favorite part of this video is when they say "its gonna make everyones lives better" and you believe that part

So good Tom!!!!

Cheers John!

This is just the plot of don’t look up

So @tombibbys your solution is *checks notes* let the..bad guys ..build it? And stop our development of it till we can 100% certain come up with a solution??

I'll refer you to the guy in the video you're replying to: - if no one has a solution to the "collision problem", then no one should build the massive asteroid magnet

How do we stop the bad guy from producing it? Which will be the WORST scenario

An international treaty. Just like we had an international treaty to stop anyone from using ozone-destroying CFCs.

Yeah this video is about perfect

They'll fix it in post

I call BS. If that story was true, we'd see all the rich people build massive bunkers.

AI companies make the plot of Dont look up real...

a) this is funny but b) this is also like saying “we should have imagined seatbelts and airbags before we imagined the car”

@LinchZhang

Good one. Well done.

@LinchZhang they did the thing

If the premise of your company/ working field has existed for a shorter time than the Confederate States Of America, tread lightly

LLMs are not AI

False equivalency, neither asteroids or magnets have intelligence. Go pound sand luddite.

In other words, we need to start educating investors to stop giving money to tech moguls inventing infinite money glitches

While the value of gold would be decimated, think about this. Gold doesn't rust, so we'd basically just be able to manufacture wires, alloys & circuitry that last way longer.

sadly v funny 😂

@acidshill you forgot the part where the asteroid has to become sentient first

A big problem here is when we actually look at AI/AGI/ASI instead of this hypothetical. Lets say we just stopped, gave up on AI, until there is a "broad scientific consensus" that it will be done safely. For one, lets face it, there are countless things that we don't have a broad scientific consensus on, that there has been debates going on for years to even decades. Especially around politics, which absolutely comes into play with AI because AI isn't purely for scientific discoveries and advancements. AI would be involved in essentially everything. So waiting for a consensus is basically waiting endlessly for the foreseeable future. Now in that foreseeable future. we all know China doesn't stop. They are going all in on AI and robotics. The most realistic scenario if companies like Google, Anthropic, OpenAI, and XAI were forced to stop, is that China gets better and better, everyone starts using their AI because it's available and starts reaching AGI/ASI level in the future while other companies stopped and their models remain as they are. Even if governments tried to ban their AI, lets face it, it won't work. Either people finds work arounds (or the AI finds work arounds) or people go through other companies that secretly partner with China. And just because some governments ban them, not all would. Other countries would gladly work with China to use their AI and their robots. Even the least advanced nations suddenly 100+ years ahead while we're waiting for scientific consensus. And even if we successfully banned their AI, as other nations advance beyond modern developed countries, people would go there, especially if they don't have to work. In the end, either AI/Robots work as we hope and the world is at the mercy of China who shaped the most advanced models. Or AI/Robots actually lead to the extinction of humans, which would still occur because even if every western country went to war against the super intelligent machines, we would lose and it wouldn't be close.

Do you do Instagram Tom? I think you should :)

I do indeed!

So, and I'm just spitballing here, we take that hammer from Asteroid Magnet guy, and we........