So long as the prediction is viable from a technological and execution level, I don’t see why it has to be labeled “doomer.” If it’s possible and people can exploit it to their advantage, they will, that’s humans.
Eh? It absolutely has in many industries. It also has been tried many times in software, it just tends to fail, so the fact the option was there and perceived as an advantage and acted upon is exactly my point. It very well could happen and it absolutely has and has been tried, as soon as it was viable.
When it comes to the above comment about grabbing the likeness of an individual to create an artificial human interface to the layoff process, “deep fake” approaches already exist and are very viable, it’s simply a matter of signing over yet another right in an employment contract. Some groups are already doing this with entertainers and public facing figures because there’s a lot of value in it. So long as it’s cheap to capture a bit of imagery from an employee and cheap enough to overlay and generate video for, how unbelievable is it that a business would simulate a manager figure to lay you off in a conversation pretending to be personable? Technologically we’re not that far away.
At some point you might even be able to crawl internal business conversations to gather personal characteristics and tailor a script or even the appearance of the simulated human to reduce reaction to a layoff message, assuming the risk to cost made sense.
I put nothing off the table of possibilities from groups of sociopaths with power, unless it’s quite literally impossible or too costly to do.
So I'm thinking it's possible to do this _right now_.
Simply put, when we look at the trends of AI improvement we are tasked with the question, what is going to cause the sustainment of human labor for its current value across broad areas of all markets?
AI is coming for those FTE's. It's nowhere near replacing the good ones but we have such a glut of shitty developers in this industry and AI is absolutely coming for them.