I appreciate the humility and clarity of your writing. The capabilities versus uses framework is helpful, and I wonder what we can learn from other regulatory efforts to constrain the capabilities of useful technologies (nuclear, guns, automobiles). When does it make sense to transfer the burden of responsibility from the user (someone driving too fast) to the regulator/technologist (California’s proposal to limit the possible speed of cars)?
Thanks for the kind words! And great point - that's something I'm thinking about, and hoping to offer some tentative views about in the 5th post. It gets even more complicated when you have agents involved...
I appreciate the humility and clarity of your writing. The capabilities versus uses framework is helpful, and I wonder what we can learn from other regulatory efforts to constrain the capabilities of useful technologies (nuclear, guns, automobiles). When does it make sense to transfer the burden of responsibility from the user (someone driving too fast) to the regulator/technologist (California’s proposal to limit the possible speed of cars)?
Looking forward to the rest of the series.
Thanks for the kind words! And great point - that's something I'm thinking about, and hoping to offer some tentative views about in the 5th post. It gets even more complicated when you have agents involved...
For those like me that read better with their ears, I've run this through ElevenLabs for an AI reading.
https://askwhocastsai.substack.com/p/when-is-a-capability-truly-worrying