my notes ( ? )
The challenge of creating humanlike intelligence in machines remains greatly underestimated. Today’s A.I. systems sorely lack the essence of human intelligence: understanding the situations we experience, being able to grasp their meaning... research studies have shown that deep-learning systems can be unreliable in decidedly unhumanlike ways... the best A.I. programs can be unreliable when faced with situations that differ, even to a small degree, from what they have been trained on.... a malevolent hacker can make specific changes to documents that while imperceptible or irrelevant to humans will cause a program to make potentially catastrophic errors....
these programs do not ... understand the inputs they process or the outputs they produce... renders these programs susceptible to unexpected errors and undetectable attacks... we need to look to the study of human cognition... broad, intuitive “common-sense knowledge”... our core abilities to generalize what we know, to form abstract concepts, and to make analogies ... to flexibly adapt our concepts to new situations....
Read the Full Post
The above notes were curated from the full post