In an article in Wired magazine earlier this year, Alessandra Ram, a writer, wife and mother based in the San Francisco Bay Area, noted: “There’s a strange and under-discussed side-effect of the AI boom.” That is: “what it’s doing to family dynamics”. By which she means: how it’s potentially destroying those dynamics.
She writes from experience. She’s married to an AI-pilled chap who has a newish job in an AI company. He phones her at 11pm in Berkeley, California, where she is home alone with their 10-month-old daughter. It’s 2am in Massachusetts, where he is, and he wants her to look at something on his computer.
“Are you looking?” he shouts. She isn’t. Instead, she’s looking at their real baby and thinking: “There are two babies in this household now: the small human one and the large language model. Both demand constant attention. Both keep us up at 2am.”
Being an award-winning journalist, Ram doesn’t leave it there. She goes out interviewing her contemporaries in the Bay Area because that’s where this particular psychological crisis seems most acute. She finds that it goes like this: “He works in AI, and she does everything and anything else. Other times, it’s bleaker: He desperately wants to work in AI – or feels he must work in AI – and she wants him to do literally anything else.”
Now we need to remember that northern California is not a normal place at the moment. It’s the centre of a frenzy reminiscent of the gold rush of 1848-55, with Jensen Huang of Nvidia selling the contemporary equivalents of picks and shovels. And the demographics of the tech industry are chronically skewed: it’s basically a male-dominated playground. This applies not only to the tech companies, but also to those who invest in them or who want to work in AI labs.
Ram thinks that implies that there are “hundreds of thousands of spouses, partners, and girlfriends, holding the fort while someone mansplains the singularity to them”. The evidence in her piece is mostly anecdotal, but I think she’s on to something.
At any rate, it chimes with my own experience of watching this space. Five years ago, I annoyed some of the industry’s male protagonists by pointing out that its most perceptive critics tended to be women. Broadly speaking, this is because women are more sensible than men in these matters. I write as a long-term gadget freak who is constantly surprised that my female friends are invariably unimpressed by the latest gizmo I have acquired.
So you can see why I sympathised with former Miami Herald columnist Dave Barry, who once took his wife for a ride in a brand new Humvee. As they rumbled along, he proudly mansplained to her that, among the many features of the 4×4 military vehicle, there was a switch that enabled him to inflate or deflate its tyres as they drove along. His wife very sensibly asked: “Why?” – to which he had, of course, no answer. She was likewise unimpressed by the Humvee’s mighty winch, which, she was told, could pull down a house.
As far as AI adoption is concerned, there’s an increasing amount of empirical evidence that women are more judicious than men in approaching the technology. A key Oxford study, for example, suggests that the emerging “gender gap” in AI use “is associated not only with differences in skills, access or socioeconomic position, but also with perceptions of AI’s societal risks”. Concerns related to mental health, environmental impact, data privacy and employment are closely linked to women’s approach to the technology.
The researchers found consistent evidence that these risk perceptions are more strongly associated with adoption among women than among men. Women who express concerns about AI’s broader social consequences are less likely to use generative AI tools frequently, and – interestingly – these differences are especially pronounced among younger and digitally literate groups.
This looks to me like encouraging news. It suggests that what the Oxford researchers found among young, digitally literate women isn’t technophobia but a grown-up, cost-benefit calculation that the AI boom’s protagonists seem constitutionally unable to make.
Which brings us back to Ram’s “two babies”. One of them will, in due course, grow up and become judicious. I wouldn’t bet on the other one, though.
What I’m reading
Irish eye
Gaslighter | Denier is Sinéad O’Sullivan’s devastating critique of Ireland’s political elite.
Mutual friend
US novelist Neal Stephenson expounds on what writers can learn from Charles Dickens in Say It, Don’t Show It.
Triple lock
The Pundit’s Trilemma finds Henry Farrell tackling a problem that many writers face posting on Substack.
Photograph by Justin Sullivan/Getty Images