How similar is ChatGPT-5 to the human mind? How should we measure AI progress? How close are we to transformative AI? Are "super-intelligent" AI sytems likely to kill us all?

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.conspicuouscognition.com/subscribe

This is a free preview of a paid episode. To hear more, visit www.conspicuouscognition.com

This week, I talked with the great Elle Griffin about “free speech in the age of social media.” Elle is a writer, essayist, and general polymath who writes the excellent newsletter The Elysian. I would highly recommend following her work and Substack.

We talked about a wide range of topics, including:

* What is misinformation?

* Is the misinformation problem worse today than in the past?

* Why is censorship bad? And why is free speech good?

* In what ways is misinformation symptomatic of deeper problems like polarisation and institutional distrust?

* Should we treat online spaces like public parks?

* How can we design social media platforms so that they encourage prosocial behaviour? (Elle had some very interesting ideas here).

* In what ways is the enforcement of “epistemic norms” (e.g., against lying) different to the enforcement of other norms (e.g., against littering)?

* Do we really want social media platforms to be free of conflict? Should all platforms be comfortable?

I got a lot from the conversation and hope people enjoy it.