Eric Horvitz, Microsoft’s head of research, has spoken out about the current debate surrounding artificial intelligence (AI) and the concerns many people have about the way it might be used in the future.
With Google having recently announced that a form of “machine learning” is already being used in its search algorithms, the race to come up with a true AI is on.
History of flight
Horvitz, the director of the Microsoft Research lab, recently spoke on the subject to a full room at MIT’s annual EmTech technology conference held in Cambridge, MA.
“I like to think about advances in AI through lens of the history of flight,” he began.
“It started as a few guys on a windy beach looking at a kite,” Horvitz explained, “and 50 years later, we have a 707.”
The editor of the MIT Technology Review, Jason Pontin, had begun proceedings by asking the audience: “What do we want from smart machines?”
Stressing the need for parameters for new technology to be determined by the twin factors of how it can help us and how it should be applied, he paved the way for Horvitz to address the question in light of the history of augmented knowledge.
Advancement of artificial intelligence
Horvitz knows about the subject of AI because as well as being a leading light at Microsoft, he is also the former president of the Association for the Advancement of Artificial Intelligence (AAAI), the world’s largest scientific society of AI researchers.
“It’s important to take stock of recent advances,” he said, noting that at Microsoft alone, there are currently 1,200 people working in research, of which a quarter are “solely focused on AI.”
Although he acknowledged “fast-paced progress” in AI, Horvitz warned that “it’s unclear how fast things will move with machine intelligence.”
“The mastery of AI has been much harder than expected,” he explained, believing that there is a need to focus on “the more deep general intelligences,” rather than simply building “narrow, deep wedges of intelligence.”