An AI that writes convincing prose risks mass-producing fake news
A new algorithm that writes convincing prose could be used to automate the writing of fake news, Will Knight writes.
What it’s capable of: The program can make up realistic-seeming news reports on any topic you give it. It was developed by a team at OpenAI, a San Francisco-based research institute.
The dangers: This is a troubling example of just how good AI is getting at fooling us. It's so convincing that the researchers have chosen not to open source the code, to try and stall its weaponization as a way of automating and mass-producing fake news. “If this technology matures—and I’d give it one or two years—it could be used for disinformation or propaganda,” Jack Clark, OpenAI’s policy director, said.
How it was developed: Originally, the researchers aimed to create a general-purpose language algorithm to help translate text or answer questions. It could have beneficial uses, like summarizing text or improving chatbots’ conversational skills. But the potential for abuse could outweigh these for now.
See the full story here: https://www.technologyreview.com/s/612960/an-ai-tool-auto-generates-fake-news-bogus-tweets-and-plenty-of-gibberish/