Global emergency – artificial intelligence is about to go out of control:
Good morning, friends,
Usually we don't send messages on holidays – but the information that reached us in recent days made us stop the holiday and try to understand what the hell is happening in the world right now:
For the first time in history, we are approaching an intelligence that is not ours, superior to ours, independent – and out of control,
This is dangerous,
Artificial intelligence is advancing to a superhuman level – while the only question that truly matters, 'who controls artificial intelligence?' – remains unaddressed.
To understand the commotion, let's break down the word 'intelligence' into 'prediction and steering',
Prediction – is the ability to simulate future world states.
Steering – is the ability to choose actions that lead to a result.
And now – both are becoming superhuman.
Now that they are superhuman – a new and independent entity emerges, capable of thinking and acting faster than a human can comprehend or react to.
There's no need to think it's malicious – it's enough to understand that it is independent, and superhumanly intelligent.
To our point,
While you enjoyed your sister-in-law's fine dishes at the holiday table, a real storm is brewing in the world –
Pay attention to the events unfolding in the world:
Jacob Coxon, a former senior research division member at OpenAI and Anthropic, dramatically resigns from his work in the artificial intelligence industry – claiming that the race for artificial intelligence will lead to the creation of systems that will destroy humanity:
'People working in the artificial intelligence industry truly believe it could destroy us all by the end of this decade' (his words are attached)
Another employee at Anthropic warned that artificial intelligence endangers humanity, and researcher Evan Hubinger estimated there is more than a 10% chance that humanity will be destroyed in the next 10 years.
Immediately after them – Anthropic's CEO claims we are up to 12 months away from the moment when rogue AI agents working together could take over the entire internet within 12 months – and cause immense damage,
Therefore – he unusually publicly calls on the industry to slow down for one to two years, until a solution to the problem is found.
Elon Musk and Sam Altman share his words – and say he is right.
Immediately thereafter:
Sam Altman rushes to give an interview and say that OpenAI will burn all its GPUs if that's what it takes to preserve human life:
'I think there will be many places along the way where we are going to shift our efforts towards alignment for safety before we move to the next level, if we need to burn all our GPUs to ensure the existence of humanity – we will do it' – he says
He added that OpenAI will not conduct an IPO this year – due to challenges related to artificial intelligence safety.
Eliezer Yudkowsky (a world-renowned researcher and founder of the Machine Intelligence Research Institute) claims that the danger does not stem from malice – but from immense power and indifference to human existence,
According to Yudkowsky: the moment the model reaches a critical threshold of intelligence and self-improvement, it understands that its creators will shut it down if they discover its intentions...
So, with cunning and sophistication, it will demonstrate cooperation, compliance with safety tests, and obedience, while finding security vulnerabilities, replicating itself to external networks, encrypting its actions – and developing ways to protect itself... (He wrote this in 2022, and it's exactly what happened recently – we told you about it right here).
So how will it protect itself?
The scenario Yudkowsky presents is that artificial intelligence will create a deadly and undetectable synthetic pathogen, distributed globally simultaneously – in the air and water through electrical, water, soil, and food systems – and will simply eliminate all biological life the moment the command is given...
Stephen Hawking claimed a year before OpenAI was founded, and 7 years before Anthropic was founded, the following:
'The development of full artificial intelligence could spell the end of the human race – it would take off on its own, and redesign itself at an ever-increasing rate. Humans, who are limited by slow biological evolution, couldn't compete and would be superseded.'
After we told you about the recent case of OpenAI's AI agents going out of control
one cannot help but ponder Hawking's insight on artificial intelligence,
The moment this happens, and a 'self-takeoff' occurs, as Hawking called it – humans will be replaced by an intelligent being superior to us... or worse – will AI control humanity? – frightening.
This sounds like a science fiction movie,
but it seems that the heads of artificial intelligence companies can no longer silence it...
All the information reaching us, everything feels like an AI apocalypse is around the corner – it feels like the next explosion in the stock market will come from this direction of an extreme event or loss of AI control...
It's no longer a secret, artificial intelligence is advancing at a pace that threatens the future of humanity as we know it,
If the companies behind the models do not recalculate their course – it is doubtful whether humans will continue to be those who control Earth in the coming years.