Tech industry workers have signed an open letter calling for a six-month pause in Artificial Intelligence (AI) development.
But Cambridge University professor Jonnie Penn told Q+A this morning that their concerns around the dangers posed by AI are misplaced.
Penn is concerned — not about the possibility of machines taking over humans — but rather of the big tech players taking over the AI industry, pushing out the small to medium-sized businesses and institutions.
"Thankfully," he said, the community of AI researchers "has come around to that".
"There's still, in the far future, concern that we'll reach a point that machines will become, you know, people allude to it as sentient or something more cognitive," he said.
"But historically, I just don't think we're close to that at all."
He says the driver of AI research is only "marginally interested in the brain, they like to mimic it, they like to borrow from it, but they're not interested in copying it one-for-one."
The letter, calling for a pause in AI research, was signed by the likes of tech billionaire Elon Musk and Apple co-founder Steve Wozniak, with similar concerns being voiced by Geoffrey Hinton — dubbed the godfather of artificial intelligence.
Penn didn't sign that letter, "in part because I think we should take that idea of restraint, hold it and hit it with a hammer and break it into a thousand pieces and use that idea of holding back and being more deliberate with precision."
He supported an approach he calls "rest engineering", where researchers "deliberately avoid some uses and focus on the areas that we know it can be useful. And broadly, the uses that we ought to avoid are around predicting social phenomenon."
Examples of predictive social engineering with AI include predicting success in school or the likelihood someone will commit a crime — a technology predicted by 2002 film Minority Report.
Penn says these specific uses are over-deterministic and potentially very dangerous, but the notion of a six-month pause on AI research — as espoused in the high-profile letter is "arbitrary and infeasible".
"Who's gonna go for that? And I think even the people who signed it know that... where in particular, what uses are you so worried about?"
That doesn't mean that AI research ought to be progressed without limits, in his view. In a time of "overlapping crises", Penn believes we need to start "finding ways to give people time to step back and reflect on where the world's going."
"We've got this technical innovation, what is the social innovation that matches it? It could be the three-day weekend, it could be, for parents out there, paid childcare.
"If AI is truly here to save us time and to improve our productivity — prove it."
When it comes to regulating AI development, he said it's likely that there will eventually be some sort of global watchdog — but he's more excited about bottom-up, counter-cultural responses.
Examples Penn alluded to include drivers disavowing touchscreens and wanting to revert back to kinetic controls, young people increasingly turning away from dating apps and supermarkets instituting slow lanes in some stores to maintain traditional face-to-face connections for seniors.
"We need social innovation, who does social innovation? Care workers, nurses, teachers — regular people, who deal with other regular people and design a world that regular people want to live in."
Q+A is Public Interest Journalism funded through NZ on Air
SHARE ME