Realistic Course of Events in terms of AI Integration
We seem to be reaching an inflection point, and I want to remain optimistic that development will be paused/controlled, but I am not sure how plausible or realistic this actually is, but you can let me know what you think in the comments.
AI development is set to continue raising security concerns in terms of its capabilities to destroy crucial infrastructure systems. Over time, these systems will become more integrated and will be surveilled more closely by the government as a result. This will likely trigger many politicians to start really taking action against continued, unfettered development of AI capacities. Not only in the US, but in major nations around the world, the pursuit of AGI will be severely hindered politically due to the existential threat that is made very visible as its capacities continue to improve. I do believe we'll continue to see AI applications such as creating pitchdecks, financial models, and all around helping out with specific tasks. However, having one general AGI model I think will be out of the question.
I used to believe that this likely wouldn't happen, because there is already so much invested into its development, as well as the game theory issue with letting rival nations develop AI while our own goes more slowly. The sheer scale of the existential threat and the complete lack of control over intentions that will emerge as these systems get more advanced will effectively prevent all nations and groups from trying to build it. Could it happen secretly? I'm sure it can in a limited capacity, but it would be quite difficult because of the energy and financial needs a system like that would have. I would like to believe that we are capable as a society of recognizing the scale of the threat we face.