When Facebook unveiled its plan last week to open two new AI labs and create an AI safety net for its users, the company also announced that, for the first time, it will start using a new technology protect users – synthetic or “fake” data.
Data scientist Sergey Nikolenk hails the announcement from Facebook, confident it will lead to mainstream adoption of synthetic data as a powerful tool that will help tech companies more easily create machines empowered with artificial intelligence. Nikolenko’s company, Neuromation, a leading provider of synthetic data, was featured last week in WIRED magazine.
“While fake news caused problems for Facebook, fake data will help the company fix those problems,” said Nikolenko. “Facebook and countless other companies that generate reams of data every day want to create artificially intelligent algorithms quickly that can learn from the constant flow of data and make fast decisions to protect consumers,” said Nikolenko. “But when you train AI algorithms to perform their assigned tasks, the data needed to train them is sometimes not available, such as sensitive data about your customers. Synthetic or “fake” data mathematically mimics the behavior of real data perfectly, allowing companies to train their algorithms and implement AI at a faster pace.”
How will Facebook use AI and synthetic data? While the company’s long-term goal is to leverage AI to improve its various networking tools and apps, its immediate goal is to fight fake news, online harassment, and political propaganda from foreign governments, like the campaign mounted by Russia during the 2016 election.
One particularly fascinating goal is Facebook’s plan to use synthetic data to do more than train algorithms how to detect bullying language on its platform. Using a process called adversarial training, the algorithms will also learn how to generate online insults. While Facebook will presumably keep outbound insults offline, by using synthetic data, algorithms will learn faster and, over time, detect a broader range of insults.
While synthetic data will be used by Facebook to fight online harassment while protecting user privacy, it’s also being used by healthcare companies to help doctors carry out medical research while ensuring patient confidentiality. The technology has also been used in political research, recently giving data scientists at the Center for American Progress in Washington D.C. insight into shifting demographics and voting patterns likely to affect the Presidential election in 2020.