If AGI was created as an instantiating software that could run on a normal high end gaming rig, how many would install the AGI in a day?

If AGI was created as an instantiating software that could run on a normal high end gaming rig, how many would install the AGI in a day? <br><br>Assume that AGI was created as an instantiating software that could run on a normal high end gaming rig that costed $20,000. I would assume instantiating the AGI software would take a full day, and you would only have to instantiate it once. Assume that the instantiating is as easy as plugging your system into a power socket and pressing the power button. Assume that the first institution that discovers AGI is a very ethical and transparent company and is openly willing to give away AGI for everyone. <br><br>I have a few questions:<br><br>a) How widespread would AGI be adopted on its first day? <br><br>b) What would be the best strategy for the government to tackle the mass proliferation of AGI?<br><br>c) Is it true that a country that adopts AGI the fastest will have a massive advantage over countries who adopt it slower?<br><br>d) Could a government just blockade the institution who discovered AGI and ban the institution from distributing AGI? <br><br>e) If institutions are banned from distributing AGI, but they are allowed to keep developing AGI, then wouldn't the institution be able to slowly develop AGI to the point where it could break free of the blockade?<br><br>f) If governments blockaded an institution who discovered AGI, then wouldn't an international war arise from the large amount of countries who would want to blockade the institution who discovered AGI? Hence, wouldn't it be more strategically beneficial to allow the institution to distribute the AGI? <br><br>The reason why I think that instantiating AGI is only a matter of plugging a system into a power socket and pressing the power button is that this is the case for training large language models; if you purchased a system that is designed to support 8x A100 GPUs, you just need to plug the system into a power socket and you can immediately train your own large language model. Moreover, if you wanted to train a large language model but you do not have access to a system that has 8x A100 GPUs, you can purchase cloud services from Google or AWS or Microsoft and they will allow you to rent a system that has 8x A100 GPUs for a month and you can train your own large language model. <br><br>Hence, I don't see any reason why the same thing can't happen for AGI. It seemed like OpenAI is trying to keep their training hardware in house, but I'm not sure why they would do that.