4 ms·

Nuke Game Theory all over again, with AGI to add too!

ethbr0 4y ago
The fundamental difference with nuclear theory is that nukes don't self-assemble more nukes. The exponential replication of biological pandemics means there are very different optima. Whether that's in favor of GoF or anti-GoF, I can't say. It's a tradeoff between {increased risk of pandemic & more prepared for pandemic} vs {decreased risk of pandemic & less prepared for pandemic}. A sane response probably means we should build two neutral, BSL-4 (+) labs and require by UN treaty that all GoF research take place there, outlawed in the rest of the world, with verification inspection of labs. Of course, you'd be prying power out of the hands of the US/Russia/China, so there'd have to be a pretty big bone to get them to bite. Which would probably look like complete autonomy and confidentiality within the lab, in their respective areas. With strict perimeter and time-quarantine -based controls.
- telchar 4y ago
  That’s the fundamental difference from the anti-GOF side. The fundamental difference from the pro-GOF side is that understanding exactly how a nuke works gets you no closer to being able to protect a population from nukes because there is no such thing as a nuke vaccine. Your suggestion seems like a likely workable compromise if one of those labs is primarily in the US/Euro sphere and the other is primarily in the Russia/China sphere, with enforced transparency.
  - ethbr0 4y ago
    Yeah, I pontificated about physical location for a while and... wow, that'd be a difficult set of negotiations. Probably Switzerland, because, hey, Swiss and history. Other than that, it's almost have to be as neutral / third-party of a state as you could find. Maybe New Zealand or one of the small island nations that's willing to declare neutrality in exchange for hosting? Islands and caves are hard to argue against.
- danenania 4y ago
  Is there any concrete evidence that gain-of-function research actually increases preparedness for pandemics? It seems like the benefits are all still very hypothetical? Considering it would need to prevent a massive pandemic just to break even at this point, perhaps it's time to throw in the towel.
- jtc331 4y ago
  > The fundamental difference with nuclear theory is that nukes don't self-assemble more nukes. It's worse than that: nukes don't learn to become even more powerful nukes while being used. > BSL-4 (+) labs One of the issues is that effective safety in labs is virtually impossible in reality (lab leaks that are effectively non-material events are relatively high frequency even from the best labs).
dathinab 4y ago
IMHO it's not the same, very different nonce and factors of control leading to quite different dynamics

The fundamental difference with nuclear theory is that nukes don't self-assemble more nukes. The exponential replication of biological pandemics means there are very different optima. Whether that's in favor of GoF or anti-GoF, I can't say. It's a tradeoff between {increased risk of pandemic & more prepared for pandemic} vs {decreased risk of pandemic & less prepared for pandemic}. A sane response probably means we should build two neutral, BSL-4 (+) labs and require by UN treaty that all GoF research take place there, outlawed in the rest of the world, with verification inspection of labs. Of course, you'd be prying power out of the hands of the US/Russia/China, so there'd have to be a pretty big bone to get them to bite. Which would probably look like complete autonomy and confidentiality within the lab, in their respective areas. With strict perimeter and time-quarantine -based controls.

telchar 4y ago
That’s the fundamental difference from the anti-GOF side. The fundamental difference from the pro-GOF side is that understanding exactly how a nuke works gets you no closer to being able to protect a population from nukes because there is no such thing as a nuke vaccine. Your suggestion seems like a likely workable compromise if one of those labs is primarily in the US/Euro sphere and the other is primarily in the Russia/China sphere, with enforced transparency.
- ethbr0 4y ago
  Yeah, I pontificated about physical location for a while and... wow, that'd be a difficult set of negotiations. Probably Switzerland, because, hey, Swiss and history. Other than that, it's almost have to be as neutral / third-party of a state as you could find. Maybe New Zealand or one of the small island nations that's willing to declare neutrality in exchange for hosting? Islands and caves are hard to argue against.
danenania 4y ago
Is there any concrete evidence that gain-of-function research actually increases preparedness for pandemics? It seems like the benefits are all still very hypothetical? Considering it would need to prevent a massive pandemic just to break even at this point, perhaps it's time to throw in the towel.
jtc331 4y ago
> The fundamental difference with nuclear theory is that nukes don't self-assemble more nukes. It's worse than that: nukes don't learn to become even more powerful nukes while being used. > BSL-4 (+) labs One of the issues is that effective safety in labs is virtually impossible in reality (lab leaks that are effectively non-material events are relatively high frequency even from the best labs).

dathinab 4y ago

IMHO it's not the same, very different nonce and factors of control leading to quite different dynamics