Facebook, Twitter Crack Down on AI Babysitter-Rating Service

Facebook and Twitter are taking action against an automated babysitter-rating system reported on last week by The Washington Post, saying the service broke the social networks' rules on user surveillance and data privacy.

Predictim, a California-based start-up, analyses babysitters' online histories, including on Facebook and Twitter, and offers ratings of whether they are at risk of drug abuse, bullying or having a "bad attitude." Facebook said it dramatically limited Predictim's access to users' information on Instagram and Facebook a few weeks ago for violating a ban on developers' use of personal data to evaluate a person for decisions on hiring or eligibility.

Facebook spokeswoman Katy Dormer said the company also launched an investigation earlier this week into Predictim's extraction, or "scraping," of personal data. That investigation is ongoing and could include further penalties.

Twitter spokesman Nick Pacilio said the site conducted its own investigation earlier this week and revoked Predictim's access to important site tools - known as application programming interfaces, or APIs - that would allow the start-up to review and analyse babysitters' tweets on a massive scale.

"We strictly prohibit the use of Twitter data and APIs for surveillance purposes, including performing background checks," Pacilio said.

The crackdown, which was first reported by BBC News, could restrict Predictim's ability to analyse babysitters with what it called "advanced artificial intelligence" and sell parents the results. The service was heavily criticised for offering unproven and potentially inaccurate results that could affect a babysitter's life.

But Predictim's executives said Tuesday they were undeterred and intend to continue pulling publicly available data from Facebook, Twitter and Instagram. Predictim chief executive and co-founder Sal Parsa said he also believed the social media giants' reaction was at least partially rooted in a competitive rivalry over personal data use.

"Twitter and Facebook are already mining our data. It's right there, user-generated data. Now there's another start-up that's trying to take advantage of that data to help parents pick a better babysistter, and make a little money in the process," he said. "I don't know why they ganged up on us. It could be because there's no benefit for them."

Dormer, the Facebook spokeswoman, said that Predictim's continued use of Facebook data would be a direct violation of the network's rules. "This is not about anything other than making sure the apps on our platform are protecting people's data," she said.

Parsa and Joel Simonoff, Predictim's chief technology officer, said they spoke with Facebook policy officials in recent weeks and received a letter from Twitter on Monday. The changes would not hurt their algorithms' accuracy, Parsa said, because the company had "decided to source data from other means."

The company, which advertises that it has trained its algorithms on "more than 6 billion data points" to assess a babysitter's "important personality traits," is working to expand its analyses to include data from babysitters' blog posts and Reddit activity, Parsa said.

"If you're not hiding anything, if you're not abusive, if you're not a bully, I don't see why you'd be scared of letting a parent see who you are," Parsa said.

It's unclear how the change could affect Predictim's expansion. Sandra Dainora, the head of product at Sittercity, a babysitter marketplace that had said it planned to use Predictim's analyses as part of a pilot screening program early next year, said Tuesday that "the safety and privacy of our users is paramount" and that it "will certainly take this news into consideration as we vet the service."

Brad Shear, a Maryland attorney who specialises in social media and privacy law, says Predictim's problems may run much deeper than that. The site, he said, appears to violate a ban on employers demanding job applicants verify or give access to their personal social media accounts. Such requests might run afoul of the law in 26 states, according to data from the National Conference of State Legislatures. Parsa said the service is "perfectly legal."

"What they're doing is purely against public policy: There are First Amendment issues, Fourth Amendment issues. If you talk to any lawmaker out there, they'll say it's absolutely disgusting what Predictim is trying to sell," Shear said.

"The fact they would think this is okay obviously demonstrates they have some ethical issues," he added. "They're selling snake oil they say can predict people's personalities and misleading parents along the way."

© The Washington Post 2018