Self-improvement isn’t limited to the human experience. 

AI is capable of something called “self-recursive” improvement, in which it reviews its own code, finds flaws, fixes them, writes new code, and tests the update. 

This isn’t a bad thing. I mean, who doesn’t love a good transformation story? It might as well make its own glow-up reels. 

But in my own personal journey from paranoid technophobe to occasional AI user, the idea of a self-improving AI unnerves me somehow.

If you’ve ever used AI for a creative endeavor, then you know it has limits. Don’t get me wrong, it’s really good. But there’s something missing about it. Something off. 

A lack of soul, perhaps.

But what about future versions – the AI frontier?

I was reading a book the other day and a section really stood out for me.

“Assume this is the worst AI you will ever use,” writes Co-Intelligence author Ethan Mollick. “We are playing Pac-Man in a world that will soon have PlayStation 6s.”

Maybe AI will never have a soul. That’s a question for philosophers and theologians. But barring an unforeseen technological catastrophe, it will continue to improve. 

And you know what?

I plan to do the same.

Here are some other things I’m thinking about this week…

---

Something New:

• AI-driven ransomware attack. Look Ma, no humans! JadePuffer, a “self-narrating” agentic AI, just conducted possibly the first known fully autonomous AI ransomware cyberattack. Cybersecurity experts say this is the first of many in a dangerous new era of AI-generated hacks. On the bright side, a human did indeed initially set up the AI and point it in the right direction. Small mercies.

Something Old:

• Terminator 2: Judgment Day (1991). My favorite Terminator film just turned thirty-five. In celebration, James Cameron announced that he’s re-releasing T2 in theaters later this summer, to remind us that “the good guys win against the AI superintelligence. And maybe that’s a message of hope we all could use this summer.” Nevermind the necessity of brute physical force.

Something Inspiring:

• Paralyzed woman uses Neuralink to create art again. Twenty years ago, artist Audrey Crews was in a bad car accident that left her paralyzed from the neck down. Last summer, she became the first female Neuralink patient. Now she uses the neural implant to control a cursor with her thoughts, creating digital art that she displays and sells online. Human imagination meets cyborg tech. Pretty cool.

---

Okay, that’s all for now. Join me here next week for more.