Op-Ed: Seattle schools vs Big Tech, but how do you make a case?

Facebook owner Meta was fined for a data breach – Copyright AFP Justin TALLIS

The lawsuit filed by Seattle schools against a virtual Who’s Who of major social media companies is getting a lot of coverage. This is an extremely ambitious case. The problem is that it’s also an extremely difficult case for the schools. This is a true landmark lawsuit.

The schools are suing on the allegation of mental harm to students. A lot of people would agree with them. The brutal social media environment is leaving a very long trail of news stories, most of them pretty grim. There’s no doubt that rampant bullying and a truly vicious culture are a huge load on kids.

…But how do you prove it to the point of winning such a huge court case? The terms of the suit claim social media causes anxiety, depression, and mental harm. How do you prove that to the extent of getting meaningful legal results, and solving the problems?

The law is not in a good position here:

• There are some precedents regarding specific cases. There are no precedents regarding an entire class of social media.
• This case applies nationwide, not just Seattle.
• You need to prove the companies were negligent, and/or legally at fault, causing harm.
• Are social media companies legally responsible for user actions? To what extent? (If only this was a rhetorical question.)
• The court’s finding must either provide a solution, or dismiss the lawsuit.
• A solution would be difficult to put it mildly. Dismissal of the lawsuit would be a step backwards, and definitely not the end of such lawsuits.

Those are just the basics. The situation now gets a lot deeper and far more difficult:

• The schools are without a doubt acting very much in the public interest to the best of their ability. They’re doing what they can to manage a gigantic problem.
• Social media companies know all too well they have a problem. The fixes are likely to be laborious and expensive, and may not work.
• Any number of civil cases may be generated by a finding for the plaintiffs in this case. Existing civil cases may also be affected.
• There are extraordinary degrees of difficulty in even the terminology of the case. Try defining “mental harm”, for example. Even a psychologist would think twice about any glib definition.
• There are also serious cases of suicide and massive harassment which have civil law ramifications beyond the case.

The argument for the schools’ case suffers from being so general. The scope of the lawsuit is truly vast. Nobody would deny kids frequently do have problems on social media. A lot of people would dispute the efforts of social media to manage the many known issues, too.

That, however, doesn’t quite add up to a simple court finding. What’s the court supposed to do? Can it simply order a remedy to all these things? How? When? What sort of oversight should be in place to make sure the order is enforced?

Solutions – Algorithms aren’t people

There’s a further problem here, whatever the court finds regarding the lawsuit. The mindless faith in algorithms on social media monitoring is already bad enough. It’s inefficient. It often doesn’t get context or even syntax in statements.

For example – I referred to Russian riot police as pigs on Facebook. I got a 3 day penalty for “bullying”. Bullying whom, you may ask? To this day, I don’t know. Nobody even replied to the comment. Only the Russian riot police were targeted by the statement. Yet someone was “bullied”. Any human would understand the context, but the algorithm didn’t.

The point is that this is how algorithms work and don’t work, and they’re likely to be the default solution in this case. Auto-monitoring can’t, won’t, and doesn’t, work. It simply can’t do the job. That means you need human oversight, and that’s expensive.

Any finding, in this case, has to deliver a practical solution. That may well be tougher than the lawsuit.

_________________________________________________________

Disclaimer
The opinions expressed in this Op-Ed are those of the author. They do not purport to reflect the opinions or views of the Digital Journal or its members.