#Academia appears clearly divided on use of #AI. One camp sees the huge potential for #research, ideation, exposure to different perspectives, id of research gaps and critical thinking. The other seems to think it should be ignored and traditional research and writing prevail. Which camp are you? @academicchatter
The poll ended with
the votes above. Some interesting comments to say the least. On LinkedIn, the poll had 88% in Camp 1 0% (if that's a figure) in Camp 2 and 12% in Other. Twitter was evenly split 50/50. @academicchatter
In that case, camp 2, but not for the regressive reasons offered. I can't think of any context in which I'm interested in mapping my inputs to statistical predictions of parameters to get things that look statistically likely to be the outputs for those parameters. And I don't want the added environmental costs, or to enhance the obscene wealth of a few tech billionaires, or the risk of AI pollution seeping into our most trusted data sources.
I do sometimes need to map my inputs to statistical predictions of parameters to get things that look statistically likely to be those outputs. (Models that can generally be fit on a desktop computer rather than a power-hungry GPU driven cloud)
But that is where my desire to use ML ends. Gen AI is a neat parlour trick, but I've never found a place it is actually useful, and it has too many ethical issues around it.