The Chatbot Perfect Storm
Framing a magical time for technologists + tinkerers in today's tectonic terrain.
We’re back with another edition of Machine Learnings, brought to you by the folks at Heyday.
Heyday is an AI-powered memory assistant that resurfaces content you forgot about while you browse the web.
We have our base product. It’s powerful, helpful, and occasionally mind-blowing. But to take a true leap for everyone we hope to help, we’re crafting new experiments every day.
Sometimes they work, and we’re in awe. Sometimes they fail. Sometimes they’re confusing. Not all advances look like advances in the moment.
But at its root, AI is the closest thing to magic we’ve seen.
If you want to feel how work in the (near) future will make you a magician, join us today.
-@samdebrule
What we're reading.
1/ Let’s be very straightforward – it feels as if we’re crossing the chasm with chatbots. What makes that the case now? Learn more at Every’s Divinations >
2/ This week’s biggest mistake cost more than $100m. Hm. Learn more at BBC >
3/ If DALLE-2 felt like a distortion of the imagery we know, ChatGPT feels like the blurry version of paragraphs. Much to take away from that frame. Learn more at The New Yorker >
4/ It’s hard to stay away from Every’s work each week, especially when they’re producing some of the highest quality AI content on the web. Here’s a step-by-step guide for engineers and layfolk to build your own chatbot with GPT-3. Learn more at Every’s Chain of Thought >
5/ Ben Thompson’s talking about this week in Microsoft-Google, and as always, it’s a synthesis of the wider economic plane. This one also includes a recorded conversation with super important folks on both sides. Teaser below. Learn more at Stratechery >
6/ The new learn to code is likely learn to talk to AI. Incredible shifts, and one we all need to watch closely at work. Learn more at The Atlantic >
7/ …which leads us to a solution. One of our favorite internal prompt guides lays out a simple model to get the most out of talking to AI. Learn more at Github >
Research for this edition of Machine Learnings was enhanced by Heyday, the AI-powered memory assistant.
Author’s Note – We’re off next week, but we’ll be back on February 26.