Unpredictable Patterns #150: The Holiday Break
Top reads from the year that was, thinking about next year, and thank you!
Dear reader,
I am off for a holiday break, and so thought it good to just sum up the year that was, think a bit about next year - but most importantly: say thank you for all of the comments, feedback and the added readership — we are twice as many at the end of the year as we were at the beginning - and that feels fantastic!
As I took up writing this year again, after a two year hiatus, I really enjoyed it. There are so many interesting themes and ideas out there, and I learn by writing so the newsletter has given me the opportunity to dig into a number of different issues and questions. Among my favorites from this year are:
#105 on sensors, senses and what happens with intelligence when it is built from different and larger sets of senses than ours.
#111 on the way computational and biological time differs and why AI can become a kind of temporal mediation mechanisms. I was extremely grateful to see this one edited masterfully, and re-published by AI policy perspectives (APP is a must read).
#126 on the NatSec-shift — something that I think clearly is happening now.
#136 on culture and AI - and whether AIs can have culture.
But what was even more fun was to write with others — I had the luck and pleasure to write with quite a few people this year, and hope to expand this even further next year.
#123 with Yannis Mastrogeorgiou, on the ability of machines to wonder was really interesting to collaborate on.
#128 and #144 with Fredrik Erixon, on what a healthy public sphere really looks like focused less on the threats to the public sphere, but the neglected question of what it looks like when it really works — and why that question is key to figuring out regulatory responses.
#134 with Anastasia Bektimirova on organzing science in the age of AI helped me understand more about why organizing science may be as much of a challenge as getting to the right tools for science.
Writing together and collaborating on texts is fantastic - I hope to do a lot more of that.
But there is naturally room for improvement. I feel that some of the texts didn’t quite manage to bring across my point, and I need to revisit and come back to those. #117 on agency-enhancing technologies did not quite get it right — and I want to revisit that idea later. #141 on why even scenarios where AGI is benign, safety is solved and we all “flourish” may lead to human exit also scratched the surface on something that I want to capture better.
So for next year I want to figure out periodicity — is weekly still meaningful or would monthly be better with deeper texts, more research? Choosing a monthly theme appeals to me and could be a way forward, but I also want to figure out scope: after I restarted I have almost solely focused on AI from different aspects, I feel that this may sometimes be too limiting and at the same time too broad. Finally, I want to find more formats for writing with others — if you are interested let me know!! And also - if you have ideas about the above please let me know.
We will resume beginning of January, and then I will have a new plan for the new year — but for now, again, thank you for reading, liking, commenting and being generally curious learners!
Merry Christmas and a Happy New Year!
Nicklas