Intelligence Rising – A strategic simulation of AI futures
asst prof at wichita state | director of TFI | opinions on data science & ai strategy
When conducting interviews with dozens of the world’s AI experts over the past few years I frequently asked a question about what they thought would be good indicators of true progress in AI . Oftentimes they would reply that the latest milestones which had received substantial media coverage were not signs of true progress in AI, and offered different suggestions for what would constitute true progress in AI.
The previous post explored the recent progress demonstrated by DeepMind’s Agent57 on Atari, which could be seen as an indicator that gaming milestones no longer carry the weight they used to. Perhaps they were a poor signal of progress all along, or perhaps the real signal was just foreshadowing an era of true progress. AI has struggled to overcome the productivity paradox , and it remains to be seen whether the economic value of AI recently heralded by leading firms like McKinsey and PwC will in fact come
Intelligence Rising – A strategic simulation of AI futures