> The fix isn't to abandon AI. The fix is to use it differently. To use it in a way that sharpens us instead of softening us. To use it in a way that makes us think harder, not less.
maybe the fix is to _not_ use AI in your domain of expertise, or at all.
or, the fix could be to use narrowly trained AI - as Roman Jampolsky argues. Rather than train an AGI / ASI on the whole of human experience on the internet, perhaps train the models for narrow use - on narrow focused domains like microbiology / nano-materials etc
or the fix could be to make training AGI illegal - a regulated munition.
That might be a way to avoid creating a new intelligent species, which is competing with humans for compute and other resources... while making us dumber via outsourcing all our thinking in the short term.
polalavik•29m ago
jgord•27m ago
Ctrl-Alt-R in firefox
evnp•25m ago
jgord•18m ago