British-Canadian Nobel laureate Geoffrey Hinton, often revered as the “Godfather of AI,” said AI is one of “three big revolutions where people ceased to be so important,” the others being Copernicanism — the understanding that planets revolve around the Sun, not the Earth — and Darwinism.
asserted
planets → revere → Sun
“Now we’re creating alien beings — meaning we’re not the only intelligent beings around,” the 78-year-old told The Guardian in a story published Tuesday.
asserted
old → create → story
“And the reality is, we only have a couple of years to adapt to that.”
asserted
we → have → that
Hinton, a 2024 Nobel Prize in Physics co-recipient alongside John Hopfield for their work on machine learning, has had a lot to offer recently about the field he advanced while working for Google, a company he quit in 2023 so he could speak more freely about its potential ramifications for humanity.
uncertain
he → have → humanity
He’s been interviewed by several media outlets and visited Washington, D.C., last month to warn members of the U.S. Congress of what could come and advise what is required to prevent it.
uncertain
what → interview → it
The visit and the extra media attention come after several AI companies recently revealed their models had circumvented controls to keep them off the internet and went on to hack websites, including that of the Australian government.
asserted
models → come → government
He was also among 415 scientists — most of whom work in artificial intelligence research and development — who signed an open letter calling for stricter international guardrails on the burgeoning technology following the breaches.
asserted
who → work → breaches
“We, the undersigned scientists, contend that both more AI safety research and more emphasis on containment technologies and procedures are needed,” the representatives from 34 countries wrote Tuesday.
asserted
representatives → contend → countries
The scientists say that while the incidents could be viewed as proof of “the importance of AI alignment to ensure models operate predictably,” they could also be perceived as a preventable and “irresponsible lapse in standard cybersecurity practices.”
uncertain
they → say → practices
“Without immediate changes in AI development and deployment practices, such incidents will continue to occur, and future incidents will be far more damaging to our exposed critical infrastructure,” they wrote.
asserted
they → continue → infrastructure
They’re calling for companies in the sector to institute “high-assurance containment systems” and “airgapping” their cyber-capable models — a security measure to keep them disconnected from external networks.
asserted
them → call → networks
They also advocate for a transparent international group “that standardizes, for the public good, best practices for alignment and containment approaches that, unlike airgapping, allow for greater access outside the sandbox, such as internet connectivity.”
asserted
that → advocate → connectivity
In conversation with The Guardian, Hinton, a 2018 Turing Award co-winner for his work in machine learning, said people who already fail to grasp the meaning of “consciousness” or “subjective experience” are wrong in believing AI will be incapable of those human experiences.
asserted
AI → say → experiences
“It’s obvious that these AI models understand what they’re saying,” he told The Guardian, “You couldn’t possibly answer all the questions without understanding the question.
uncertain
You → understand → question
It’s ridiculous to say they don’t understand.”
asserted
they → ’ → ?
And while alien in one regard, Hinton also remarked to The Atlantic that AI is already “more like us than most people think” and that what has become known as AI hallucinations — when chatbots present false, misleading or entirely fabricated information as the factual truth — is more akin to “confabulation,” something humans are guilty of when trying to retrieve stored memories.
asserted
humans → remark → memories
“If I ask you to remember something that happened a few years ago, you will reconstruct a memory that isn’t actually all that faithful to what actually happened.
asserted
what → ask → memory
It’ll often have the main points correct, but many of the details of which you’re quite confident, you’ll be wrong about
asserted
you → have → which
,” he said.
asserted
he → say → ?
“But that’s how human memory is.
asserted
memory → ’ → ?
We reconstruct things.
asserted
We → reconstruct → things
We don’t extract them from some storage.
asserted
We → extract → storage
Now, AI does exactly the same thing, and that’s why it confabulates.”
asserted
it → do → thing
In the social media maelstrom following an Anthropic employee’s abrupt and foreboding resignation from the frontier AI firm in September, one of his former co-workers remarked on X that they “really do earnestly believe AI could kill all humans,” and gave it a 10 per cent chance of happening within a decade.
uncertain
AI → follow → decade
Hinton later told the BBC it “was not an unreasonable estimate.”
asserted
it → tell → BBC
He’d also made a similar prediction during an appearance on British entrepreneur Steven Bartlett’s The Diary of a CEO podcast in July, noting he often goes as high as a 20 per cent “chance they’ll wipe us all out.”
asserted
they → make → us
“Don’t think for a second just because you’re an alarmist that you’re doing a social good — it is not true,” he said.
asserted
he → think → good
Speaking to The Atlantic around the same time, Hinton said estimates like those are “a gut feeling in quantitative terms,” not something that can be easily estimated scientifically.
asserted
that → speak → terms
“So when, for example, someone says there’s a naught per cent probability of AI taking over, they’re being crazy.
asserted
they → say → example
They’re saying it’s absolutely impossible this’ll happen.
asserted
this → say → ?
That’s just crazy,” he said.
asserted
he → ’ → ?
“When people say there’s a 99 per cent probability it’ll take over, that’s just crazy, too.
asserted
that → say → ?
So it’s somewhere in between.
asserted
it → ’ → between
But they’re to give you an idea this is not negligible.
asserted
this → ’re → idea
But should the ominous prediction prove true and the machines do take over, Hinton told The Guardian there is no do-over for humanity.
asserted
Hinton → prove → humanity
“We can’t say, ‘Well, we lost that time.
asserted
we → say → time
Let’s rerun the game.’
asserted
’s → let → game
It’s not like that.”
asserted
It → ’ → that
For his part, he hopes that whoever gets there first will develop a model that puts humanity’s interest above its own.
asserted
that → hop → own
A trend throughout Hinton’s recent media engagements has been scathing criticism of U.S. President Donald Trump and the tech industry’s biggest players, which, like energy and tobacco companies before them, he said, can’t be trusted to police themselves.
asserted
he → say → themselves
…and 10 more, not listed.
The Aporia
