Some interesting things to read the first weekend in May

Dear Friends,

What exactly are our moral obligations to animals? Should we treat a snake with the same respect that we treat a bear? If we’re going to eat salmon, is it more ethical to eat one from a fish farm or from the wild? A smart friend of mine once argued that we should avoid eating cephalopods because they’re smarter than we are. I’ve struggled with these questions and have moved from veganism to vegetarianism to pescetarianism in my adult life. I’m not alone though, because, as Kelefa Sanneh argues in this wonderful new essay and book review, it is very hard to come up with a consistent philosophy of animal rights.

Peter Singer forcefully argued in Animal Liberation against “speciesism” and suggested that animals deserve rights based on how sentient they are. But if you accept that premise, must you also confer rights to humans based on their differing levels of sentience rather than on innate and universal human rights? Sanneh then takes us through the virtues and shortcomings of arguments by Martha Nussbaum and Matthew Scully. And while we’re looking closely at this issue, Sanneh explains, what are we to do with the fact that many of the strongest advocates for the rights of animals seem to be less attentive to the rights of the unborn—and vice versa? “It’s strange,” Sanneh writes, “that the people most concerned about the fate of human blastocysts take little interest in the fate of cattle or chimpanzees, and that the people who think carefully about the nervous systems of crabs take little interest in the nervous system of a human fetus.” 

The debate is made more complicated by new research showing that animal consciousness may extend further than we think. Did you know, for example, that bees seem to play, just for fun? The extreme response to animal welfare can lead one to dark places. I was friends in college with Joe Dibee, who joined a radical environmental group called the Earth Liberation Front. After a series of alleged arsons, he spent twelve years on the run (as detailed in the New York Times Magazine) before being captured in Havana in 2018. Sanneh concludes his piece with a typically perceptive line: “It is not easy to think carefully and consistently about what we do to animals. If the people who try often end up endorsing proposals that make us recoil, this may say as much about us as about them.”

The best book I read last month was Christine Yu’s Up to Speed which is about the science of female athletes. Yu, carefully and comprehensively, makes the point that sports research has been biased and directed toward men. Scientists study men because they believe that they are biologically simpler than women and then extrapolate those findings to all humans. The result is that women had to run forever without appropriate sports bras, that bike seats are designed terribly for women, and that certain things we know to be true—such as the benefits of drinking beet juice before marathons—may actually only be true for men. And while we’re on athletics, I also highly recommend this visually arresting essay from The Washington Post about the life of the Michael Jordan of bullfighting after he broke his neck. 

In technology, I loved this TED Talk by Vilas Dhar that begins on the forested paths of India and ends with an optimistic argument for making AI more just. It’s a wonderful talk, in part because he offers a vision of AI that one can be for instead of one to just be against. There is also this important new report from Stanford’s Human Centered AI-Lab that gives a necessary perspective on the larger trends in AI, in a year when the technology will become ever more embedded in our lives. I also was captivated by this weird story about how one of the most wanted men in the world revealed his location through social media detritus. (Speaking of which, here is a sobering essay from a young writer about what they missed because they never got to experience adolescence in a world without phones.) 

I also highly recommend this essay by Arthur Brooks on what we should do for our children; this way of thinking about old age by David Brooks; and this lovely profile of Albert Brooks by Adrienne LaFrance. Albert, among other things, is the author of perhaps the funniest quote-tweet I’ve ever received. 

This section is sponsored by Elastic, the Search AI Company. Elastic enables everyone to find the answers they need in real-time, using all their data, at scale. Its solutions for search, observability, and security are built on the Elastic Search AI platform—the development platform used by thousands of companies, including more than 50% of the Fortune 500. Visit here to learn more.

The public debut of ChatGPT on November 30, 2022 set off a race to build a new kind of search experience using natural language processing. I was at a conference in Arizona, and almost overnight, the act of typing queries into a search box felt passé. It was so much more engaging to ask questions of the new AI-powered chatbots, testing the limits of their knowledge and their seemingly magical ability to synthesize information. A host of companies started offering chatbots that felt like a trusted friend, a personal assistant, and an all-knowing genius, or, as time went on, a constrained all-knowing genius. 

Unlike traditional search, which returns links, AI algorithms try to summarize and explain what can be found in web pages, YouTube videos, Reddit comment boards, and everything else online. Classic search points you outwards toward the larger web—you can “surf” the internet. The chatbot invites you into a walled garden for one-on-one tutoring. It’s a fundamentally different experience from the open internet, and it’s hard to predict how chatbot search will alter the landscape of the web, which remains, at root, a mass of interconnected links. A further twist also presents itself: we are not far away from a time when much of the information on the internet will be written with the help of AI. What happens when AI chatbots summarize and synthesize AI-inflected content? Will it be smarter than what we’ve got now? Will the models collapse? Will it all, as Nilay Patel and Ezra Klein suggest in this fabulous podcast, turn into a mass of “enshittification”? Or we will just be in a world where there is more and more messy content, but better and better tools to make sense of it all?

In the meantime, I like to test the latest AI-powered search apps by asking them complicated questions about things that I don’t know a lot about. I recently had a wonderful long conversation about the causes of World War I, which helped me better join an argument between two of my children who are fascinated by this question. And I like to ask, too, about things that I do know about, just to find the limits of their knowledge. One recent favorite: “What are the specific physiological differences between the wear and tear of running a 50-mile race and running a marathon?” By exploring the responses to things we know, we can better understand the limits of answers to things we don’t know. I’ve also become a fan of the most specifically useful task I’ve found yet in my own work: uploading files of interview notes for a book I’m writing and asking the bot to find themes I haven’t noticed and quotes I haven’t appreciated. Here is also a podcast and interview where I talk both about The Atlantic’s future business plans and the ways that I’m using AI. And here’s some more about proper prompt engineering, which I think of as a skill that is incredibly useful now that is unlikely to matter much in the future.

As with any search engine, it’s helpful to have a general idea of what’s happening underneath the hood. And it’s also helpful to try to think through the larger implications of what happens as search evolves. Much of the current internet economy is built on advertising—you come to my web page, I show you an ad—but that model will be disrupted by AI-powered search that is scouring and summarizing the web for you. 

These companies must figure out how to compensate the institutions and individuals from which they draw information, and how to help humans judge the accuracy and authority of search results. It’s not an easy challenge, and great businesses will be built solving it. But, putting that aside, I’ve found that chatting with bots awakens one’s curiosity about the world, and pushes all of us closer to the famous intellectual standard set by the novelist Henry James: “Try to be one of those on whom nothing is lost.” 

Cheers * N