[HN Gopher] The Princeton researchers calling out 'AI snake oil'
       ___________________________________________________________________
        
       The Princeton researchers calling out 'AI snake oil'
        
       Author : irtefa
       Score  : 22 points
       Date   : 2023-09-19 20:44 UTC (2 hours ago)
        
 (HTM) web link (www.semafor.com)
 (TXT) w3m dump (www.semafor.com)
        
       | pedalpete wrote:
       | https://archive.is/tqUg2
       | 
       | I'm surprised they are only providing 2 buckets for AI,
       | predictive and generative, and that they say "most of the snake
       | oil is in predictive".
       | 
       | Most of the snake oil being in predictive isn't the same as
       | saying most predictive AI is snake oil.
       | 
       | With predictive most predictive AI, we can provide a result of
       | how often the AI gets the prediction correct.
       | 
       | We're in the sleep space, and do sleep stage classification using
       | neural nets. My co-founder decided for fun to see how the model
       | would work in predicting sleep stages, and it was surprisingly
       | good (above 80% coherence with manually scored data if I remember
       | correctly).
       | 
       | I think there is a different challenge in things like psychology,
       | where it is very difficult to get a true reading on what is
       | happening with the subject. However, we're also building EEG, so
       | maybe we'll get some insights there as well (but psychology is
       | not our focus).
        
       | dang wrote:
       | Related. Others?
       | 
       |  _Princeton 'AI Snake Oil' authors say GenAI hype has 'spiraled
       | out of control'_ - https://news.ycombinator.com/item?id=37243354
       | - Aug 2023 (101 comments)
       | 
       |  _ML is useful for many things, but not for predicting scientific
       | replicability_ - https://news.ycombinator.com/item?id=37089573 -
       | Aug 2023 (37 comments)
       | 
       |  _The greatest risk of AI is from the people who control it, not
       | the tech itself_ - https://news.ycombinator.com/item?id=36139852
       | - May 2023 (504 comments)
       | 
       |  _GPT-4 and professional benchmarks: the wrong answer to the
       | wrong question_ - https://news.ycombinator.com/item?id=35245626 -
       | March 2023 (185 comments)
       | 
       |  _ChatGPT is a bullshit generator but it can still be amazingly
       | useful_ - https://news.ycombinator.com/item?id=34618376 - Feb
       | 2023 (154 comments)
        
       | treprinum wrote:
       | [flagged]
        
         | dang wrote:
         | " _Don 't be snarky._"
         | 
         | " _Please don 't post shallow dismissals, especially of other
         | people's work. A good critical comment teaches us something._"
         | 
         | https://news.ycombinator.com/newsguidelines.html
        
         | archgoon wrote:
         | [dead]
        
       | dxbydt wrote:
       | The right hand doesn't know what the left hand is doing.
       | 
       | right
       | 
       | > even if we have a large number of people working on AI x-risk,
       | it does not really mean that this interest arose organically. It
       | has been very strategically funded
       | 
       | left
       | 
       | https://nitter.net/prfsanjeevarora/status/170388297344208091...
        
       | AISnakeOil wrote:
       | Finally my time to shine.
        
       ___________________________________________________________________
       (page generated 2023-09-19 23:02 UTC)