Well, it already is doing that to an extent, but I mean once the training wheels are off. Right now it is used as a tool to analyze trends in stock price and make recommendations to humans on which actions to take. I mean once it becomes an actual stock broker. What I am envisioning is an AI which looks at more than just stock ticker data and does more than just make recommendations. Here this guy made an AI bot to which he has delegated the authority to buy and sell stock on his behalf:
But it still is pretty limited, in that it only analyzes trends in stock price. What if the AI was monitoring social media, news outlets, press releases, reading quarterly reports, staying up-to-date to within seconds of current events, and making actual informed decisions(*) based on the dossier that it builds for a given company? Ten seconds ago an Apple employee on Reddit leaked something about an upcoming release of iPhone37 xs ++ Pro Mega, buy Apple stock. The president just tweeted about opening a new oil pipeline, buy stock in Kinder Morgan. Bud Light just announced a transgender spokesperson sure to offend rednecks, sell Budweiser stock.
(*) I already know it will be said that AI is incapable of making actual informed decisions, so if you're going to go there, please elaborate; why can't it do that? What are the barriers needing to be overcome before it can do that? How far away are we?
What are your thoughts and predictions?
I don't know enough about AI or the stock market to have a valid opinion on this from any angle, but I won't let that stop me. Here's my uninformed opinion presented before the AAC brain trust for target practice, so shoot holes in it; tell me why I'm wrong:
I think this will happen sooner or later and it will fundamentally change how our economy works. I predict that once the first firm implements an AI to do this and is wildly successful, there will be a race to the bottom as no human will be able to compete. The other firms will all find themselves "behind the curve" and having their actions be dictated as reactions to whatever the other firm's AI is doing. So they will implement their own AIs, some fully featured ("proactive") like I described, but mostly the "easy" kind like the one in the video ("reactionary"). The reactionary ones will still be "behind the curve" like humans, but only by a few seconds, and will exacerbate the issue. They will turn every minor up-tick or down-tick into a dramatic spike or plummet. Stocks will have already tanked or exploded before any human even understands why. It could be in response to a badly received Facebook post, maybe even one made by a hacker, and by the time it's sorted out it will be far past too late. People will go to bed destitute and wake up millionaires, and vise versa. There will be less development of needed industry and more development of wanted industry. They will be active in all markets in all countries in all time zones, the effect will be global.
Ok, there it is, fire away.
But it still is pretty limited, in that it only analyzes trends in stock price. What if the AI was monitoring social media, news outlets, press releases, reading quarterly reports, staying up-to-date to within seconds of current events, and making actual informed decisions(*) based on the dossier that it builds for a given company? Ten seconds ago an Apple employee on Reddit leaked something about an upcoming release of iPhone37 xs ++ Pro Mega, buy Apple stock. The president just tweeted about opening a new oil pipeline, buy stock in Kinder Morgan. Bud Light just announced a transgender spokesperson sure to offend rednecks, sell Budweiser stock.
(*) I already know it will be said that AI is incapable of making actual informed decisions, so if you're going to go there, please elaborate; why can't it do that? What are the barriers needing to be overcome before it can do that? How far away are we?
What are your thoughts and predictions?
I don't know enough about AI or the stock market to have a valid opinion on this from any angle, but I won't let that stop me. Here's my uninformed opinion presented before the AAC brain trust for target practice, so shoot holes in it; tell me why I'm wrong:
I think this will happen sooner or later and it will fundamentally change how our economy works. I predict that once the first firm implements an AI to do this and is wildly successful, there will be a race to the bottom as no human will be able to compete. The other firms will all find themselves "behind the curve" and having their actions be dictated as reactions to whatever the other firm's AI is doing. So they will implement their own AIs, some fully featured ("proactive") like I described, but mostly the "easy" kind like the one in the video ("reactionary"). The reactionary ones will still be "behind the curve" like humans, but only by a few seconds, and will exacerbate the issue. They will turn every minor up-tick or down-tick into a dramatic spike or plummet. Stocks will have already tanked or exploded before any human even understands why. It could be in response to a badly received Facebook post, maybe even one made by a hacker, and by the time it's sorted out it will be far past too late. People will go to bed destitute and wake up millionaires, and vise versa. There will be less development of needed industry and more development of wanted industry. They will be active in all markets in all countries in all time zones, the effect will be global.
Ok, there it is, fire away.

