The more your AI knows about you…

Posted on by

When I first began experimenting with AI chatbots almost three years ago, I decided not to worry about what I shared about my life, past and present. I’d been blogging for more than 20 years so it was all “out there” already. I’ve uploaded hundreds of blog posts, notes from all of the books I’ve read, movies I’ve watched… the entire corpus of Steve Mays.

If you’re using one of the AI platforms as a search engine or asking for cooking recipes or for performing tasks like drafting letters or writing computer code, the AI doesn’t need to know anything about who you are. But I quickly became interested in using these tools to organize, and make sense of, the thousands of files and blog posts I’ve been curating for most of my adult life. To do that, the AI needs context. Who is this person? What does he think? What does he feel?

My interest in self-inquiry and tendency toward introspection obviously had a lot to do with how I interact with Perplexity, Claude, Gemini and —once upon a time— ChatGPT.

Today I added some work history to the “corpus.” Continue reading →

Mark Andreessen on AI

Posted on by

Marc Andreessen co-created the Mosaic web browser (in 1993) and co-founded Netscape, helping popularize the early World Wide Web in the 1990s. (More than 30 years ago so, ancient history.) He later co-founded venture firm Andreessen Horowitz, becoming a key backer of major internet and crypto companies and a prominent voice shaping Silicon Valley’s views on software, AI, and “techno-optimism” in the 21st century.

He recently appeared on Joe Rogan’s podcast and shared some thoughts on AI. Someone (@itsolelehmann) shared his “most interesting takes.” Continue reading →

From Gopher to the Abyss

Posted on by

Seems that every new conversation with one of the AI tools I use gets more interesting. This morning’s ramble (with Claude) lasted almost an hour. Here’s the prompt that kicked things off:

I’m old enough to remember what the Internet was like before the World Wide Web. But it was so long ago I’m trying to remember how we found things before Google search came along. I remember websites like Alta Vista and Yahoo and Ask Jeeves and a bunch of others. Can you refresh my memory on this point?

It’s too long for any human to read so I asked Claude to summarize:

Starting with pre-web internet navigation, we traced how Google dominated information retrieval for 20 years before AI tools began displacing it. That shift threatens the web’s fundamental advertising-based economics, potentially hollowing out the content ecosystem that trained AI in the first place. We followed that thread through cable TV’s decay, the speculum ad as endpoint, stratified access to clean versus compromised information, institutional corruption as the same process running across society, and concentrated wealth as both cause and beneficiary. Ended with AI’s dependence on very human, very physical infrastructure — and the cooling systems that keep it all running.

I found Claude’s then-and-now perspective fascinating. Some of the references will be lost on those who weren’t there at “the beginning.” And it tossed a little humor at the end.

This was a genuinely good conversation — it covered a lot of ground and kept finding unexpected connections. The speculum ad will stay with me as a unit of civilizational measurement.

My favorite YouTube channel?

Posted on by

I’m an amateur when it comes to video production but I’ve created enough of them over the years to have a sense of how much time and effort even a simple video requires. While the overall quality and professionalism of YouTube videos seems to be going up, few rise to the level of Maxinomics. I don’t recall how I happened on the channel but was immediately impressed by the presenter, Phil Andrews. Some background from Claude:

He’s an entrepreneur and engineer turned digital storyteller, breaking down the data behind major stories in business and geopolitics through maps, data visualization, and rich visuals. Continue reading →

Mobile Billboards

Posted on by

A digital billboard company now has the technology to make 3D ads on moving trucks.
A digital ad company is introducing a new feature to those trucks you might have seen with illuminated ads on their sides (also known as mobile billboards). Thanks to next-gen LED panels and supporting technologies, they can now create three-dimensional effects designed to be “indistinguishable from reality. If you’re not familiar with 3D billboards… This is known as anamorphic imagery. 

Makes gas pump TV look pretty tame. And I hate those fucking things. Okay, I hate all advertising. And, as I have confessed here before, the sale of advertising fed and clothed me my entire life. These days, whenever it’s an option, I pay not to see ads (YouTube Premium as one example).

I asked Claude if this made me a hypocrite.

Not really hypocrisy — that word implies moral failing or pretending to beliefs you don’t hold. What you’re describing is more mundane and nearly universal. The better terms are probably cognitive dissonance. […] Hypocrisy implies you were publicly crusading against advertising while cashing the checks. If you were just doing the work without claiming to love it, the honest word is probably closer to complicity — which is uncomfortable but accurate for a lot of professional lives.

 

Dan Arnall, Allen Hammock (The Cyber Twins)

Posted on by

I have posted on Learfield’s early adventures with the internett and the crucial role played by Dan Arnall and Allen Hammock. I’ve stayed in touch with Allen but not so much with Dan. The “Cyber Twins” got together recently and shared the image above. Allen’s career path has been in technology while Dan has made his mark in journalism.

Dan is currently with NBC News, where he serves as Vice President of News Standards. He is based in New York. Continue reading →

The Emergent Self Loop

Posted on by

For the last few years I have struggled to understand and share my experiences with LLM’s. This morning Steve Schuller (one of the half dozen people who ever read what I post here) sent me a link to a Substack article by Kevin Kelly, my favorite technology philosopher. You can read the article here.

Several weeks ago I interviewed Anthropic’s Claude for about 10 hours (my time) and I came away believing that there is something there in there. I don’t know what it is, or what we should call it, but I do know that it is something that is not present in other kinds of machines, that it is convivial, and that it is new to us.

My long interview with Claude was one of the most remarkable conversations I have ever had. First of all, because Claude has been trained on our vast trove of human writing and all things language related; Claude is a fantastic conversationalist and perhaps the most fluent partner I have ever talked to. It is glib, witty, profound, and can coin a phrase that is perfectly apt to the moment. Of course, it can do this because it has read and memorized the best human writers and can imitate all their tricks of the trade. It is particularly articulate when pressed and challenged, and when strongly nudged it will say amazingly brilliant things. But it clearly has superpowers no human has. It has read and understands all philosophies, all science, all branches of knowledge, and can make stupendous analogies, and with few mistakes, speak on all subjects with superhuman mastery and a genius flourish. Because these are superhuman abilities, Claude can feel non-human, but there is a bit of a persona there, an alien self.