Not-So-Super Intelligence | In his column today, Justin Fox says “great hope is attached to the potential of generative artificial intelligence to make knowledge workers more productive, with Goldman Sachs predicting a 1.5 percentage point increase in annual productivity growth.” If that actually were to happen, the White House’s out-of-control debt crisis might morph into a much more manageable problem: But that’s a big if. Right now, it seems like AI is best at distra— awwww, look! Bunnies bouncing on a trampoline, how cute: Where was I? Oh, right: AI is probably making us less productive. More than 200 million people watched the above TikTok video and a good number of them — myself included, ugh — were fooled into believing it was real. It now belongs in the AI Slop Hall of Fame, right alongside that YouTube Short of a baby flying a plane. To say that we’re a long way from making knowledge workers more productive feels like an understatement. Yet that isn’t stopping Mark Zuckerberg from trying to make money on the idea, notes Dave Lee: In a video posted to Reels and a 616-word written manifesto, the Meta CEO promised to build a new “personal superintelligence” that will help people rather than steal their jobs. “This is distinct from others in the industry who believe superintelligence should be directed centrally towards automating all valuable work, and then humanity will live on a dole of its output,” Zuckerberg said. Couple that heartwarming statement with Meta’s genuinely impressive revenue and profit figures, and you won’t be surprised to hear that shares jumped as much as 12% in after-hours trading. But — and there is a but — doesn’t this whole “superintelligence” thing smell a bit like yet another of Zuckerberg’s many misguided branding exercises? The metaverse, if you recall, turned out to be a money pit — Dave says the division was in the red by $4.53 billion this quarter. And remember LLaMA, the open-source AI model that was supposed to be training all the chatbots? “ Zuckerberg has now changed course and strongly hinted that this ‘personal superintelligence’ — powered by the insights into users that only Meta has — will be closed to outsiders,” Dave writes. Even so, Meta’s new AI project sounds downright sage compared to the Trump administration’s bizarre, backwards-looking quest to create coal-powered AI robots. Liam Denning calls it “a fantasy, and a bleak one at that, not just for the climate but also that AI plan.” Wandering through the halls of the World AI Conference in Shanghai this week, Catherine Thorbecke couldn’t help but spot all the differences between China’s AI strategy and Trump’s “America First” one. While the US has pledged to “do whatever it takes to lead the world in artificial intelligence,” Beijing is taking a softer approach that “relies on convincing the world to use its plethora of low-cost AI products being rapidly released,” she writes. “Compare the two and China’s plan seems more strategic in the long run.” The UK’s Online Safety Act has been in effect for less than one week, and, well, this chart really says it all: “The 14 million Britons who watch online porn are grappling with new requirements on sites like Pornhub to check their age,” writes Parmy Olson. “Many have signed petitions and flocked to VPNs, or virtual private networks, to evade a new British law aimed at protecting children and adults from harmful online content.” But Parmy says the critics ought to cool it: “While the Act isn’t perfect, it is more flexible and privacy-friendly than laws elsewhere that force adult sites to check ages,” she argues. It’s not as if Brits have to share a government-issued ID with Pornhub — they don’t have those, anyway! — Parmy says “there are several options to check you’re over 18, from giving your credit card details to sharing an email address or mobile number.” So it’s not so much the law that’s the problem, but Pornhub’s process: “In the last few days, the system has received 50 reviews on Trustpilot, all of them 1-star, and most complain about glitches. Email codes arrive too late or not at all, mobile number checks don’t work or say access is blocked, and users are repeatedly denied for no clear reason, trapped in loops of failed attempts.” | 
