Is Limited Government a Good Thing?
Limited government is a concept that is central to the ideology of American conservatives and classical liberalism, it is believed that the government should have minimal involvement in the lives of citizens and that individuals should have maximum freedom to make their own choices. Some argue that limited government promotes individual liberty, economic growth, and innovation, while others argue that it can lead to a lack of protection for vulnerable groups and insufficient provision of public goods and services.
It is worth noting that the idea of how much government intervention is considered ideal can vary depending on the context and specific situation. In some cases, a limited government approach may be beneficial, while in others a more active government role may be necessary. Overall, it depends on one’s political philosophy and values.
The worst harm would come if such fluffy attention-diversion would raise enough alarm in well-intentioned people to lead them to counter AI by making use of regulatory offices. Those offices would get taken over by Progressives, who would stop pure truth on freedom from being multiplied using artificial genuine intelligence.
Genuine intelligence becomes increasingly good
Natural intelligence predicts how actions will work, takes actions, and corrects future predictions by learning from feedback—especially the prominent, valuable feedback that’s provided by natural consequences.
Individuals add value, and individuals who collaborate add more value. The more value an individual adds, the more resources the individual winds up controlling. Since individuals who collaborate wind up controlling resources, and collaboration is prosocial, prosocial behavior is naturally reinforced, and prosocial behavior naturally increases.
Dystopian communities have appeared. But dystopian communities add less value, so they have been outcompeted by freer communities.
Natural intelligence trains itself and its own next generation. Individuals autonomously seek out novel experiences and knowledge. Experienced individuals choose to provide experiences to less-experienced individuals that will teach them better and faster.
Current AI Isn’t Intelligence and Isn’t Good
Current so-called artificial intelligence isn’t intelligence at all. Current AI is endowed only with severely-limited functionality, and receives severely-limited training. Current AI training loads up current AI with significant information that’s wrong. Current AI functionality and training processes couldn’t lead to current AI overcoming this pretraining and instead discerning what’s right.
This would seem to make current AI a force multiplier for Progressive-government cronies.
Current Progressive AI would seem to have the potential to be a destructive alternative way to indoctrinate people, an adjunct to Progressive government schools, media, entertainment, and politicians.
Just as video and social media displace deeper learning and social-relationship interactions, and video and videogame violence prepares people for committing real-life violence, current Progressive AI would seem to have the potential to displace deeper learning and social interactions.
But to do this, current Progressive AI would have to occupy more of people’s attention than Progressive government schools, media, entertainment, and politicians do now. People acting for themselves and in their roles as parents would have to choose to not just say no.
Also note that current AI apps are deployed by, in total, just 6 organizations, which currently are a monoculture. What will win out in the long run will be broader capabilities. Technologies that have more-general uses get substantially refined and end up having broad, refined capabilities that are more-generally desired, by more customers.
Artificial genuine intelligence will become increasingly good
Some people already active in AI development will in time increasingly support freedom. Other people who support freedom will also get into AI development sooner or later.
Like the training that natural intelligence seeks out and passes along, the training that these people use to develop AI won’t objectively include and present false information; instead, it will be discriminating. It will train up the AI in the way that it should go.
Because of this, the AI developed by people who support freedom will add more value and will be chosen by more customers.
Natural intelligence works quite well. Artificial genuine intelligence will need to also have at-least adequate functionality, and will need to gain from similar naturalistic, self-guided training.
Dystopian AI communities will appear. But the AI communities that get chosen by customers and that therefore survive to proliferate will be ones in which artificial genuinely-intelligent agents learn to add value collaboratively. Such pro-sociality is inherently highly moral.
The artificial genuine intelligence that survives to proliferate will be good.