“Assistants are all the rage right now. Everyone seems to be working on one,” Charles Jolley writes for VentureBeat. “People imagine them becoming the new interface to computers — why bother with apps and searching the web when you can just ask your assistant to do it for you?”
“Yet, a major challenge stands between the dream assistant and the current reality. It’s called the multi-agent problem, and most companies are reluctant to talk about it,” Jolley writes. “The solution will ultimately determine how much of an impact assistants will have.”
Jolley writes, “How does an assistant, with limited knowledge of the world and a limited set of isolated agents, many of which might claim to do the same thing – choose which one to activate for every command in a way that will make users happy?”
Much more in the full article – recommended – here.
MacDailyNews Take: We’re still in the infancy of personal assistants and artificial intelligence. Many breakthroughs will come.
Whatever the problem, all we have to do is trust Tim Cook. He knows.
Another stupid click bait article touting “serious problem” while the last paragraph surmises “This category is still very young and is still evolving.”
Utter nonsense and pure drivel what passes for journalism these days.
Automated services face an unsolvable dilemma: Humans like to ask questions without any specifics, expect the answer they seek, and often become irritable the moment they don’t get it. IE: A person sees a result displayed and asks “Siri, What is that?” It’s up to Siri to guess what aspect they are focussing on. Often as humans we can see where they are looking or pointing, but a computer can’t. Is it a house in a field in N.Dakota? A jpeg? A map? Which is the right answer? Even worse is the common use of innuendo or sarcasm, often merely implied by tone, and if it flies over most humans’ heads then even the best AI has little chance. What lazy humans really want is instant gratification from something that reads their minds. And that won’t happen anytime soon. The best they can do is gradually train *you* to ask the right questions using preprogrammed key phrases like “Siri, show me nearby restaurants” … and that’s kind of backward isn’t it?