Based on Scott Aaronson's Five Worlds of AI.
I will try my best to resolve the market in the future, but realistically this is as much an opinion poll as a prediction market. I will use the market probabilities as a Schelling point for how to resolve it, but I may deviate from them if I judge that they are mistaken or manipulated.
If none of them apply, the market resolves to the closest option I can round it to. If we have an edge case, the market may resolve to a PROB mixture.
1,000People are also trading
I find it interesting how these questions posted years ago age.
Back in 2015 when Bostrom published his book, his view was that there would be "one superintelligent model" that would "lock out" all other models and rule the world or something liek that.
But we obviously know that is not turning out to be the case so far, and we've blown past "AGI" by any reasonable definition. I was able to output 1100 lines of working code in 3 hours today. Now, what people consider "superintelligence" is what "AGI" used to be. The "doomers'" goalposts continue to move forward, without being able to articulate exactly at what point we will cross this hypothetical threshold to an "AI future."
I offer that the most likely outcome to this question is that it is never resolved, because there won't be any "AI future." There will just continue to be constant improvements in many types of models that improve life and nobody's going to point to some moment where the "singularity" started.
@DavidHiggs 25 years is a long time for AI to kind of not do anything when AGI is probably coming in 10, but 1.2% is just so low. "Your world model is just fundamentally way off" to the point of a couple decade delay in some tech or other should basically never be that low.
I believe AI will lead to power concentration in the hands of the wealthy and powerful and more inequality. Does this count as AI Dystopia?
If this dystopia happens, it may be difficult to tell whether “AI” caused it or whether “increasing automation, including commodotized embedded AI as appropriate” helped cause it. What’s the difference between “AI Fizzle” and “AI Dystopia” in that case? Would the distinction be about whether AI is in the visible foreground or background of everyday citizen life?
@Karsh Asked ChatGPT to vome up with some examples, these seem satisfactory assuming AI doesn't do enough good to cancel out the corresponding badness:
AI-driven Mass Unemployment: In this bleak scenario, rapid advances in AI technology have led to the widespread automation of jobs across all industries, resulting in mass unemployment and social unrest. Economic inequality has skyrocketed, with a small elite controlling the AI technology and reaping its benefits. Society has become polarized, with the majority of the population struggling to make a living and find purpose in a world where human labor is no longer valued.
AI Surveillance State: In this dystopia, AI systems are used by totalitarian governments to create an all-encompassing surveillance state. The AI technology allows for constant tracking and monitoring of citizens, analyzing their behavior, conversations, and even thoughts. Privacy is virtually non-existent, as governments use the AI systems to maintain control and suppress dissent. Trust and social cohesion erode, as people are forced to live in constant fear of being watched and reported.
AI-controlled Fake News and Disinformation: In this scenario, AI technology has become so advanced that it can generate realistic, yet entirely fabricated news stories, videos, and images. Powerful groups or individuals use AI-driven disinformation campaigns to manipulate public opinion, spread propaganda, and destabilize societies. Trust in media and institutions is eroded, leading to chaos and widespread paranoia as people are unable to distinguish between fact and fiction.
AI-enhanced Bioweapons: In this dystopia, rogue states or terrorist organizations have utilized AI technology to develop advanced bioweapons, capable of targeting specific populations or even individuals based on their genetic makeup. These AI-controlled pathogens are difficult to detect and counter, leading to devastating outbreaks and a perpetual state of fear. International relations and cooperation crumble as countries engage in an arms race to develop their own AI-enhanced bioweapons and countermeasures.