 The knowledge economy Much of human interaction must be understood as a status game. People often refuse charity when they need it, because the humiliation of accepting donations or gestures would be more painful than the hardship of doing without; they spend money on things they don’t need to demonstrate that they have it to spare. This is cross-cultural, and even overrides basic needs: “There were Polynesian chiefs,” writes the philosopher Dan Williams, “who grew so accustomed to having servants carry out their tasks that they would sooner starve than be seen feeding themselves.” But this framework does not only apply to material goods, says Williams. People like to show off their intelligence and wisdom, and form beliefs “not to acquire knowledge but to signal their impressive qualities and loyalties.” Furthermore, he argues, competition for status might be behind the “the widespread populist rejection of claims advanced in institutions like science… and mainstream media.” By rejecting the knowledge disseminated by elites, Williams argues, populists and conspiracy theorists reject their claims to intellectual superiority: “It is the conspiracy theorist, not the elites, who knows things that others — the gullible sheeple — do not.” Counting down The pace of progress in artificial intelligence is remarkable — and scary. The CEO of Anthropic said this week that he expects AI to utterly transform society within two to ten years; other thinkers have similar timelines. But the programmer Dwarkesh Patel, who has interviewed many AI experts on his podcast, says he predicts somewhat longer timelines than many of his guests. He thinks that AIs’ difficulty in learning continuously will be a significant bottleneck: Where humans can learn incrementally, when you close an AI’s chat window, most of its knowledge is lost. “How do you teach a kid to play a saxophone?” Patel asks. “You have her try to blow into one, listen to how it sounds, and adjust.” But with large language models, he says, it’s more like one student tries to play the sax; when she fails, you send her away and write up why she went wrong. You then give those instructions to the next student, who tries to learn from them. As a result, he anticipates things taking somewhat longer than the most bullish estimates. But even Patel, a somewhat skeptical voice, thinks we will enjoy “a relatively normal world up till the 2030s or even the 2040s,” but that after that, “we have to expect some truly crazy outcomes.” His race is run Bob Emmerson was a fixture of the Northampton Parkrun in the UK. Parkrun is a wonderful institution: A simple five-kilometer (3.1 mile) run, not a race, free and open to all. It began in west London but now operates across parks and cities in 23 countries every Sunday morning; among its many millions of participants are Olympic-level runners aiming for sub-15-minute times, and first-timers just trying to get round at a walking pace. Emmerson, who passed away last month aged 92, was among the few who notched up 500 runs, after taking up Parkrun in his early 80s. Running blogger Craig Lewis felt it worth commemorating Emmerson, who described himself as an “old jogger” but was much more than that. In his mid-40s he had taken up cross-country running and had discovered a love for it, completing marathons and eventually ultramarathons; he broke age category records at 30-mile, 40-mile, and greater distances. He ran 24-hour races and was the British 100km champion. Even in his 80s he was running 30-minute 5Ks, although that had slowed to 45 minutes by his last years. He ran more than 100,000 miles in his lifetime, “all neatly recorded in a pocket log book.” |
