Geoffrey Hinton, the pioneering mind behind AI industry-transforming neural networks, who’s often referred to as a “godfather of AI,” says we need to infuse AI with “maternal instincts” to save humanity from rogue AI.

Though his work on neural networks helped to usher in the large language models (LLMs) that dominate Silicon Valley today, these days, Hinton is known for being somewhat of an AI alarmist: he believes that there’s a significant chance that superintelligent AI will wipe out humankind, and talks about this risk frequently.

As CNN reports, Hinton, who was awarded a Nobel Prize last year, elaborated on his dystopian vision at an AI industry conference in Las Vegas on Tuesday, arguing that AI will be too smart to be “submissive” to attempts at domination by humans.

AI agents “will very quickly develop two subgoals, if they’re smart,” Hinton told the conference, as quoted by CNN. “One is to stay alive… [and] the other subgoal is to get more control.”

Instead, Hinton advised that humans should opt for a different approach: essentially turn AI agents into mother-like figures, so they’ll care for their idiot babies, or humans.

“The right model is the only model we have of a more intelligent thing being controlled by a less intelligent thing,” said Hinton, as quoted by CNN, “which is a mother being controlled by her baby.”

“That’s the only good outcome. If it’s not going to parent me, it’s going to replace me,” the scientist continued. “These super-intelligent caring AI mothers, most of them won’t want to get rid of the maternal instinct because they don’t want us to die.”

Hinton’s theory isn’t just strange; it doesn’t appear to be backed by science and is shrouded in a murky, complicated history.

The concept of the “maternal instinct” suggests that women are born with an innate, almost mystical instinct to mother that kicks in automatically when a baby is born. The experiences of pregnancy and parenthood both alter the brain, studies show, but women’s post-birth experiences drastically vary. Many women struggle with their post-partum mental health and don’t form an immediate connection with their infant, while research continues to find that mother-infant connections are often learned over time, as opposed to always being intrinsic and instantaneous. The idea that a so-called maternal instinct is a biological truth was popularized largely by men, experts have argued, and is rooted deeply in religious stereotypes, eugenics, and gendered biases.

“The notion that the selflessness and tenderness babies require is uniquely ingrained in the biology of women, ready to go at the flip of a switch, is a relatively modern — and pernicious — one,” the Pulitzer Prize winning journalist Chelsea Conaboy, who wrote a book interrogating the flimsy science behind the theory of the maternal instinct, wrote in a 2022 essay for The New York Times. “It was constructed over decades by men selling an image of what a mother should be, diverting our attention from what she actually is and calling it science.”

That’s not to say that mothers — and parents in general! — don’t love their children and want to protect them. But the idea that we could somehow infuse a mystical maternal instinct, let alone in a measurable way, into superintelligent AI systems assumes that it exists at all.

According to CNN, Hinton did mention that mothers also experience social pressure to care for their child, and don’t rely solely on instinct. These social pressures do exist and are a powerful force. That said, though, social pressure is embedded into nearly every aspect of human society, and is far from exclusive to mothering and parenthood.

It’s worth noting that superintelligence is still theoretical, and there are more immediate AI risks we can focus on — for example, the furthering of existing social biases already baked into the training data of future AI models.

Besides, we can’t imagine that these two choices — trying to exert control over AI through an abusive process of domination and submission, or becoming superintelligent mommy AI’s helpless babies — are the only two possible paths forward, should artificial superintelligence ever come to pass.

And in the meantime, before we release the one robot mommy to rule them all, maybe the AI industry can work on tamping down the gender biases embedded into AI models, or hire more women to actually build its products?

More on AI guys saying things: Sam Altman Calls His Own AI Model “Annoying” After Being Forced to Raise It From the Dead

The post The “Godfather of AI” Has a Bizarre Plan to Save Humanity From Evil AI appeared first on Futurism.

Origianl Creator:Maggie Harrison Dupré
Original Link: https://futurism.com/godfather-ai-bizarre-plan-save-humanity
Originally Posted: Thu, 14 Aug 2025 16:58:19 +0000