3 ms·
"Get me a Coke..." was accomplished in 1972 by Terry Winograd's SHRDLU and many 3rd gen AI systems since. NNs will almost certainly succeed where expert system
by randcraw 11y ago
"Get me a Coke..." was accomplished in 1972 by Terry Winograd's SHRDLU and many 3rd gen AI systems since.
NNs will almost certainly succeed where expert systems failed, largely because we now understand that no single monolithic pattern matcher will suffice alone. Any cognitive engine must be composed of many components, each attuned to a different purpose and context. And now we better recognize the huge need for learning, both for initial skill acquisition and for lifelong thereafter.
Minsky's "Society of Mind" is probably a better illustration of how AI will evolve (if not manifest), as well as how it must integrate with our myriad collective needs and personal lives.
- drzaiusapelord 11y agoSorry, I meant as a product. Why can't I buy a robot today that can handle "get me a coke" or "fill the dishwasher" or "mow the lawn."
- oberstein 11y agoThat is a problem of robotic hardware, not AI. It's really easy to hook up a microcontroller to get mic input and query a voice recognition server (backed by some sort of AI software) and then do commands based on interpretation of the result (more AI software). The hard part is actually being able to do the commands. There's a reason the service industry hasn't been totally destroyed by robotics yet, and it's simply because it's hard to get robots to interact with the physical world with any sort of generality. Though lawns are simple and standard enough there have been lawn-Roombas for a long time.
- notahacker 11y agoIsn't the fundamental problem with robotics not replicating the relatively simple mechanics involved in tasks like getting a Coke, but writing specific software drivers to perform each class of mundane manual tasks because AI isn't intelligent enough to figure out how to use its robotic limbs and computer vision itself.