Summits Yellow

Facebook security chief fires back at critics over 'broken' algorithms

Tyrone Stewart

FacebookFacebook’s chief security officer, Alex Stamos, has hit back at criticism of the social network’s algorithms in an impassioned tweetstorm aimed at journalists for their lack of understanding surrounding the complexities of algorithms and their role in detecting fake news.

Stamos’ tweets came in response to tweets from Lawfare’s associate editor Quinta Jurecic, who questioned Facebook’s decision to have politically-targeted ads reviewed by humans rather than designing a better algorithm following the ongoing Russian ads scandal.

Stamos, who is a key figure in Facebook’s work to figure out how the Russian ads got through its system, did not take kindly to Jurecic’s words, however.

“I am seeing a ton of coverage of our recent issues driven by stereotypes of our employees and attacks against fantasy, strawman tech cos,” he said on Twitter. “Nobody of substance at the big companies thinks of algorithms as neutral. Nobody is not aware of the risks.

“In fact, an understanding of the risks of machine learning (ML) drives small-c conservatism in solving some issues,” he continued. “For example, lots of journalists have celebrated academics who have made wild claims of how easy it is to spot fake news and propaganda.

“Without considering the downside of training ML systems to classify something as fake based on upon ideologically biased training data.”

Though Stamos’ tweetstorm may raise some good points, it’s interesting to see a Facebook exec break the company’s usual measured, prepared, and deliberate, approach to dealing with criticism. In addition, Stamos makes a point of the lack of knowledge of people when it comes to the work of algorithms, but doesn’t address the fact that Facebook has the means to provide this knowledge to the people criticising it.