[lo0dot0]

The answers can be recorded and reviewed. The other points are true, or is there a way to make outcomes deterministic, when compared to previous versions while allowing to add more knowledge in newer versions?

[vintermann]

It's possible to make any model deterministic. Used to be just to save the seed, but I'm not sure it still is now that everything is distributed. Maybe a little more effort.

[dustingetz]

determinism isn’t really enough, we want “predictable”. Most of these AI wavefunctions are “chaotic” - tiny changes in state can cause wildly divergent outcomes

[lo0dot0]

A part of my question that you didn't go into was, can new knowledge be added in a new version without making the answers with knowledge learned in previous versions non-deterministic?

[kevingadd]

This requires an exact lock-down of things like the hardware and driver version, doesn't it? Is that sustainable?

[Yoric]

But that won't survive an upgrade, will it?