I’m talking about using AI to perform, direct, and improve upon nearly everything biologists do.
If he is going to insist that the AI is useful for more than analysis, I dont think he makes a very good case as for why an AI would be better at the trivial lab tasks, compared to a human.
Yes, the AI would be better at tasks which require analysis. I dont see how the AI is going to be an improvement over a human intelligence when it comes to starting and stopping centrifuges.
Examples of advances that work well in a technical sense, but whose impact has been substantially reduced by regulations or misplaced fears, include nuclear power, supersonic flight, and even elevators.
Elevators used to be dangerous, and those regulations were written for a reason. That reason is perhaps outdated and these regulations should be examined.
Aversion to Nuclear Power, on the other hand, was primarily the result of fearmongering. There wasnt a good, actual case to be afraid of nuclear power in the US; a total of 13 people have ever died at a nuclear power facility in the US. Only 4 of those deaths had anything to do with nuclear energy, and the most recent of the incidents was in 1964.
Supersonic planes? They literally generate a sonic boom over people's homes. If that can be mitigated and the issue solved, then by all means solve it. But until then, supersonic flight simply isnt viable.
you might think that the world would be instantly transformed on the scale of seconds or days (“the Singularity”), as superior intelligence builds on itself and solves every possible scientific, engineering, and operational task almost immediately.
This is a very grounded argument to make, but it also fails to see the forest from the trees. In reality, experiments do take time to run and an AI Superintelligence would not be able to make mold grow on bread any faster than Fleming could.
But a Superintelligence would certainly accelerate things. Discovery and research will always take time, but the superintelligence will better interpret those results. Subsequent research will be better directed, and more productive.
It wont solve every issue overnight, that notion is certainly hyperbole. But the advancement of society would accelerate the moment it is switched on.