I don't think this is in response to anything specific. It's more an acknowledgement of an ongoing "boiling frog" problem. I recently heard Eliezer Yudkowsky m

I don't think this is in response to anything specific. It's more an acknowledgement of an ongoing "boiling frog" problem. I recently heard Eliezer Yudkowsky make this comparison in the (somewhat) related domain of AI safety: let's say you gained reliable intel that aliens were invading in exactly 30 years. Do you begin acting immediately, or do you wait? Anything you could do now, you could probably do next year. But the problem is, there's no obvious point at which to act; on any given day, you keep kicking the ball down the road. The same logic applies even if the length of time is unknown (X years). There is a high probability of future crises regarding both AI safety, and authoritarian states leveraging surveillance and social algorithms. In both cases, if we wait for a clear instigating event, it may be too late.