Fed the right amount and type of data, machines will be able to learn to do jobs so well that they will radically alter the workforce says one expert.
The prediction comes from Anthony Goldbloom, founder of the data scientist crowdsourcing platform Kaggle, in an address which was recently uploaded by TED Talks.
“Over the past two years, we have made dramatic breakthroughs. Machine learning is now capable of far, far more complex tasks,” Goldbloom said.
He noted that from simple tasks like sorting mail, two recent separate challenges set up by his company Kaggle resulted in machines performing as good as humans in scoring essays and detecting the eye disease diabetic retinopathy.
“A teacher might read 10,000 essays over a 40-year career. An ophthalmologist might see 50,000 eyes. A machine can read millions of essays or see millions of eyes within minutes,” he said. “We have no chance of competing against machines on frequent, high-volume tasks.”
“On frequent, high-volume tasks, machines are getting smarter and smarter. Today they grade essays, they diagnose certain diseases. Over coming years, they’re going to conduct our audits, they’re going to read boilerplate from legal contracts.”
Nonetheless, the future will not become a place where robots have completely taken over all jobs. Goldbloom said that some jobs will remain in the hands of capable humans.
“The fundamental limitation of machine learning is that it needs to learn from large volumes of past data. Humans don’t,” he said. “We have the ability to connect seemingly-disparate threads to solve problems we’ve never seen before.”
He added that the future of a job when it comes to surviving automation can be gauged by asking a single question.
“To what extent is that job reducible to frequent, high-volume tasks and to what extent does it involve tackling novel situations?” he asked.