The medium is the message for social and AI

Marshall McLuhan's adage "the medium is the message' continues to be prescient when thinking about the tech we use today.

A recent After Babel post uses it to highlight why content moderation is a red herring:

Let me be clear: there is no way to make social media safe for children by just making the content less toxic. It’s the phone-based childhood that is harming them, regardless of what they watch. Kids need to be freed from the grip of smartphones and social media, especially through early puberty. This is why two of the four norms I propose for solving our collective action problems are about delaying children’s complete immersion in the virtual world.

It's also a helpful adage to think about problems with genAI chatbots. When we search for things on Google we expect to get a factual, finite answer. But with LLMs we're getting a facsimile of an answer - which may or may not be factual. The story of Gemini generating diverse nazis is exactly this problem - the technology doesn't work how everybody expects it to work. It's not just what the LLM outputs that is the problem, it's the fact we're using it in place of a search query altogether. I don't think we've fully reconciled what that means for how AI chatbots will impact our lives.

That's why it's useful to understand how the tech we use actually works. I really like this Pragmatic Engineer article on five real-world engineering challenges by someone on the applied engineering team at ChatGPT:

How ChatGPT works isn’t magic, and is worth understanding. Like most people, my first reaction to trying out ChatGPT was that it felt magical. I typed in questions and got answers that felt like they could have come from a human! ChatGPT works impressively well with human language, and has access to more information than any one human could handle. It’s also good at programming-related questions, and there was a point when I questioned if ChatGPT could be more capable than humans, even in areas like programming, where humans have done better, until now?

Lastly, the article Artificial investment from The Verge argues AI is in for a reckoning in 2024. AI is expensive, and the value in return is questionable at times. One reason I've had skepticism about the current genAI movement is because it's not clear how it can be profitable. I expect even more scrutiny in a ZIRP period. This was an interesting tidbit in the article about a potential new biz model for genAI:

Right now, most AI companies are generating money through premium pricing for better services, says PwC’s Greenstein. But that could potentially evolve into outcome-based pricing, he says. “The idea of paying on the outcome — you know, ‘I saved your money, I’ll get some, you get some, everybody wins’ — it’s a really interesting model in AI, because there are a lot of things [where] you can assure an outcome,” he says. “And I think those will be very compelling commercialization models.”

💩 Cool shit

The Silk Roads – I love this project. Learn about the Silk Roads through a choose your own adventure.

Password Basket – A fun spin on a password generator. Note: I don't recommend actually using it for your passwords. Get a password manager.

Trudy Tube – This is a fun journal of someone tracking all the songs they listen to on Spotify.

Animate Anyone – A scarily impressive paper about animating still images into video.

Social spaces directory – An Airtable listing digital online spaces.

Legal Lullabies – Listen to reading of a tech company's terms of service. You know, if you need help getting to sleep.

Ad Nauseam – A side scrolling game about advertising.

DrumHaus – A digital drum and rhythm machine.

Tool Finder – I wish I found this sooner. Search and read reviews about a range of productivity tools.

Share this with a friend because LAST ALERT: ALL YOUR PHOTOS WILL BE DELETED!