tbh everything about the current implementation of "AI" is starting to look like hot porridge when it comes to real world products.
Is the prompting workflow so convenient that it’s worth having to spend twice or thrice as much time double checking the accuracy of the inference and fixing bugs?
How long until we collectively decide that to reduce the probability of errors we’re better off going back to writing our own functions, methods, classes etc. because it gives us granular control?
Last but not least, we’re devolving to mainframe and terminals…
I’m replying from the cold east coast (from the edge of a wood chair in a lovely iykyk type of restaurant) to a human posting from el cap on hn; We have achieved peak technology. Oh yeah, I’m working on urban logistics, powered by AI.
My immediate observation when I first learned of the Quipu and its use: nodes and edges.
Potentially a graph to be completed by the owner via verbal communication/interpretation as a supplement to the material instrument; a single source of information that could be interpreted differently depending on the societal role and vocation of the owner.
Something is not right. The sequence of the story starting with the accidental invite and the total ignorance that the "guest" was there. The pieces of the puzzle are not puzzling. The ideal next step would be a forensic device audit.
Also, taking into consideration something as simple as UNC5792, what if the "guest" was bridged-in? iykwim
Is the prompting workflow so convenient that it’s worth having to spend twice or thrice as much time double checking the accuracy of the inference and fixing bugs?
How long until we collectively decide that to reduce the probability of errors we’re better off going back to writing our own functions, methods, classes etc. because it gives us granular control?
Last but not least, we’re devolving to mainframe and terminals…