Once upon a time, I was tired of emailing Word files back and forth, and so I built Writely (Google Docs). Today I’m tired of drowning in wildly conflicting takes on the most important issue of our era, so I'm launching the Golden Gate Institute for AI. (These are both team efforts! Writely with my friends Sam and Claudia; Golden Gate with @roonotroon and @TarenSK. Three is a good size for a founding team.) TL;DR: we’re here to help you make sense of the AI firehose; see the reply tweet for ways to follow us. If you’re reading this, you probably feel the pain. Every day there are a dozen major updates – a new model, a new research paper, new analysis. Many of them are credible, all seem contradictory. It’s too much to even read, and impossible to make sense of. Just in the last few weeks, we’ve seen “AI 2027” and “The case for multi-decade AI timelines”. AI is ready to unleash a wave of bioterrorism; it doesn’t know anything you can’t find with a Google search. It’s coming for your job; it’ll usher in an era of prosperity; it’s full of snake oil. We’re here to help – with your help. We’re going to be doing two things to help make sense of the cacophony. First, convenings: getting people with diverse viewpoints and expertise talking to one another in conducive environments. Rachel demonstrated this last fall by running the wildly successful conference The Curve, a hotbed of productive conversations. The conference assembled several hundred people from AI labs, DC think tanks, Silicon Valley startups, AI safety organizations, academia, and elsewhere – groups that don’t often mix. New York Times writer Kevin Roose summarized his experience by saying “It felt like an event where history was happening.” At Golden Gate, we’ll turn The Curve into an annual conference, and add many smaller events. Second, accessible analysis: taking ideas that are circulating within the AI community, identifying key points of agreement and disagreement, and presenting a synthesis for a broad audience without dumbing-down. This is what I’ve been trying to do for the last two years in my newsletter; we’ll be doing it at larger scale. To reflect our role, the newsletter has a new name: Once upon a time, I was tired of emailing Word files back and forth, and so I built Writely (Google Docs). Today I’m tired of drowning in wildly conflicting takes on the most important issue of our era, so I'm launching the Golden Gate Institute for AI. (These are both team efforts! Writely with my friends Sam and Claudia; Golden Gate with Rachel Weinberg and Taren Stinebrickner-Kauffman. Three is a good size for a founding team.) TL;DR: we’re here to help you make sense of the AI firehose; see the reply tweet for ways to follow us. If you’re reading this, you probably feel the pain. Every day there are a dozen major updates – a new model, a new research paper, new analysis. Many of them are credible, all seem contradictory. It’s too much to even read, and impossible to make sense of. Just in the last few weeks, we’ve seen “AI 2027” and “The case for multi-decade AI timelines”. AI is ready to unleash a wave of bioterrorism; it doesn’t know anything you can’t find with a Google search. It’s coming for your job; it’ll usher in an era of prosperity; it’s full of snake oil. We’re here to help – with your help. We’re going to be doing two things to help make sense of the cacophony. First, convenings: getting people with diverse viewpoints and expertise talking to one another in conducive environments. Rachel demonstrated this last fall by running the wildly successful conference The Curve, a hotbed of productive conversations. The conference assembled several hundred people from AI labs, DC think tanks, Silicon Valley startups, AI safety organizations, academia, and elsewhere – groups that don’t often mix. New York Times writer Kevin Roose summarized his experience by saying “It felt like an event where history was happening.” At Golden Gate, we’ll turn The Curve into an annual conference, and add many smaller events. Second, accessible analysis: taking ideas that are circulating within the AI community, identifying key points of agreement and disagreement, and presenting a synthesis for a broad audience without dumbing-down. This is what I’ve been trying to do for the last two years in my newsletter; we’ll be doing it at larger scale. To reflect our role, the newsletter has a new name: Second Thoughts. We won’t be the first to rush out hot takes on new developments. Instead, we’ll synthesize the myriad takes that inevitably emerge, and present a coherent picture. When there are seven conflicting takes on a key question, we want to be the organization you can turn to to make sense of it. Because, Goddamn it, I want to make sense of it too. Our ask is that you follow us and spread the word, using one of the links in the next tweet. We want people to find our work and come to our events. (On that note, retweets are welcomed!). We won’t be the first to rush out hot takes on new developments. Instead, we’ll synthesize the myriad takes that inevitably emerge, and present a coherent picture. When there are seven conflicting takes on a key question, we want to be the organization you can turn to to make sense of it. Because, Goddamn it, I want to make sense of it too. Our ask is that you follow us and spread the word, using one of the links in the next tweet. We want people to find our work and come to our events. (On that note, retweets are welcomed!)