Facebook Whistleblower Haugen’s Historic Google Patent Filings Might Raise Eyebrows
All of the posts, advertisements, and suggestions presented to users on social media sites like Facebook and Instagram are not happy little accidents. The Facebook platform uses complex algorithms and AI that analyze usage and engagement patterns, which then infer the type of content that the user is most-likely to engage with. It is that curated content that is presented to users – interspersed with the social media platform’s preferred bias du jour, of course.
The specific content presented to users is at the crux of the Facebook whistleblower Frances Haugen’s disclosures last week. Haugen, a former data scientist and product manager at Facebook, contends that Facebook knowingly spreads content optimized strictly for engagement or reaction, "even when that content is knowingly harmful, based the social network's own research”. It seems obvious that Facebook would present users with content they are most likely to engage with, but it’s the second part of the allegation that is drawing ire. If Facebook knows particular content is harmful, and presents it to users anyway, that's a problem.
Haugen’s damning disclosures claim that Facebook is ultimately putting profits and growth over user health, which landed her in front of Senate lawmakers clamoring for some sort of meaningful regulation of Facebook and its various properties.
Much investigation and discussion is needed before any rash moves are made regarding Haugen’s allegations or Facebook’s responses, especially considering the technical ineptitude of many of our elected officials. Some recently discovered patent filings made by Google featuring inventions by Haugen (and partners), however, paint an intriguing picture of the mind and potentially motives of the Facebook whistleblower.
The breadth of patents show Haugen clearly has a firm grasp of the methods and mechanisms necessary to not only filter unwanted (or potentially unsafe) content, but to present specific content to users based on predetermined criteria and engagement goals.
All of the patents where Frances Haugen was featured are visible in this advanced search of the US Patent Office’s database. Considering they were all filed while Haugen was at Google, many of the patents are geared toward search results, but titles like, “Evaluating content in a computer networked environment”, “Providing suggestions to a user based on social activity patterns” and “Determining content of interest based on social network interactions and information” seem to be squarely in line with Haugen’s allegations excoriating Facebook. One patent in particular, “Generating a filtered view of a content stream” encompasses the allegations succinctly.
The summary for that patent plainly states, "In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include the actions of receiving a request to filter content items displayed in a content stream, with the request specifying one or more criteria for filtering of the content stream; traversing a social graph to identify items of data with attributes satisfying at least one of the one or more criteria and connections to the user; identifying content items associated with the identified items of data; generating a filtered view of the content stream, wherein the filtered view includes the content items identified and excludes the other content items displayed in the content stream." Further, a patent application from 2018 titled, “Using Metadata to Summarize Social Media Content” also seems apropos to the discussion.
How to interpret these insights will ultimately depend on your opinions of Frances Haugen and Facebook. Is Haugen a hypocrite that knew exactly how to fix Facebook’s problems and did little to sway the company during her tenure? Or is she a frustrated subject matter expert trying to effect change for the greater good of society? Let us know what you think in the comments.