This is Bloomberg Law with June Grossel from Bloomberg Radio.
It's been three years since a white supremacist opened fire inside a grocery store in Buffalo and killed ten people.
This tragedy shocked us. It devastated us. It pushed us to where we thought were beyond our limits. But it didn't break us. It didn't break us. Instead, it revealed a strength that runs deep in the veins of the city, the strength of a community that binds together and refuses to be defined by acts of hate.
The eighteen year old shooter pleaded guilty to all state charges and was sentenced to eleven concurrent life sentences. The federal case against him is ongoing, but survivors and families of the victims are seeking to hold social media companies responsible. They're suing internet giants Google, Meta, Amazon, and Reddit for publishing racist content that alleged motivated the shooter to commit a hate crime by targeting people in a historically black neighborhood hundreds of miles from his home.
Our case is fundamentally about the isolation, obsession, lack of impulse control, and desensitization resulting from the neurological and psychological trauma inflicted on a teenager's brain by a highly addictive social media product.
And today both sides were in a New York appellate courtroom where the social media companies are arguing that they can't be held liable for the racist content on their sites. The victim's families say the algorithms the Internet sites used to deliver tailored content to users should be considered products under the state's product liability law. Here are the attorneys for the victims families and Meta.
Any time our manufacturer places a dangerous product into the stream of commerce that they reasonably know creates an unreasonable risk of harm to the public, are libel are in New York state products liability law. It's plain and dirt simple.
And there are two sort of foundational problems here that get us outside of the realm of traditional product liability. One is that communicating intangible information is not a product, and the second is that services are not products. And here we have both of those things. Product liability law is geared to the tangible world. That's the language of the seminal Winter case.
It would be a first in New York if a court found a non tangible thing to be a product for purposes of liability law. My guest is Eric Goleman, a professor at Santa Clara University Law School and co director of the High Tech Law Institute. Eric, can you tell us about the allegations of the plaintiffs in these cases?
The case involved the mass shooting in a Buffalo supermarket by a white supremacist, and the basic gist of the claims is that the shooter was radicalized on Facebook, and so in order to fit that into legal doctrines, there's a variety of different claims that the victims and families have made in order to explain what's wrong with that? Why does that actually create legal rights for the victims does?
Some of those include a claim that Facebook was negligent, that it should have done something different to protect the victims, and some of it is based on what we sometimes call products liability. That Facebook is a product that is dangerous and as a result, any harms is caused by a dangerous product can be attributable to the supplier of that product.
Are the plaintiffs bringing it under the product liability law? In order to get around sex.
No. Section two thirty is a defense whereas the planets have to still figure out a claim that actually fits their facts. However, they do want to pick claims that are not preempted by Section two thirty, which says, in short, that websites aren't liable for third party content. So the idea is that the planets are arguing that they're not suing to hold Facebook liabel for the third party content. They're suing based on Facebook's contribution to the radicalization of the shooter.
So when you think of product liability, you think of tangible objects. You know, a mechanical device that injures you, something like that. Has a New York court ever found a streaming service or social media to be a product under the state's product liability law.
For decades, there have been legal theories that have held that the supply of physical items that are at risk of causing physical injury are live for those physical injuries. And one of the classic old cases involved exploding coca cola bottles, and if you didn't manufacture and fill the coca cola bottle properly, it could literally explode and create
shards of laughs that were projectiles. And so the idea is that we want manufacturers to be more careful in the manufacture of an item like that because of the physical risk of poses when that item is now in consumers' homes. There's a very bright line in those doctrines between physical items they are capable of causing physical injury and intangible thinks or services which do not pose the same risk of sending glass yards as projectiles, because there's literally nothing
like that in their possession. So the laws normally limit products liability theories to actual physical products, not to services like Internet services like Facebook. So it's a leap of the theory to try and expand it to cover services. This theory wasn't designed for that purpose, it's not optimized to do so, and unlike the regulation of Coca Cola bottles exploding with glass chards, there's significant speech implications if we tried to regulate the dissemination of information as if
it's an exploding coke bottle. So there have been plaintiffs across the country that have been trying out the theories that internet services are subject to products liability doctrines, and courts have split on that issue. Most courts have rejected it, some have entertained it, and so it's a live frontier in internet law today.
And have there been rulings bipellate courts on that.
Yes and no. There's been a variety of rulings in different fact circumstances. But there's a leading case in this area that just came out in February from the Fourth Circuit involved a very very similar case involving the shooter who allegedly was radicalized on Facebook, and the plaintiffs argued products liability is part of the reasons why Facebook should be liable, and in that case the Federal peblic Court shut all that down and said that Facebook qualified for
Section two thirty. Everything that the Planets are arguing was trying to hold Facebook liable for the third party content available on service, and therefore the Planets lose. So the plainliffs in this case aren't directly bound by that ruling, but clearly there's a strong message from that ruling that the Planetf's arguments should not be successful.
Does that issue depend on the state's product liability law, because those laws differ from state to state.
It can we have to look at the terms of the exact statute, what it says it covers. That might
very well restrict plaints claims, but not expand them. In other words, there's some limit to the theory that as far as it can go, at some point it has to stop, either because of doctrines like Section two thirty, which Congress enacted as the way to prevent that kind of explan or because of the First Amendment that the Constitution does not permit a physical based legal doctrine to extend to the dissemination of the speech.
If the court allows the case to go forward, well, the plaintiffs have to prove that Facebook or the other platforms led to the radicalization of the shooter. And how difficult will that be?
Definitely they'll have to show that what we call causation when it comes to negligence. I believe that that will also be relevant in the context of strict products liability, and it will be difficult. And causation was also an issue in the Federal Appellate Court. Rulin I mentioned from the Fourth Circuit and the court independently said there was
not sufficient causation in that case. And when we think about what causes someone to radicalize into a white supremacist and even words, what causes a white supremacist to then decide to engage in mass murder. There's so many things that contribute to that outcome, and it's very difficult to pull out one piece and say that's the cause. They
should bear all the burden of responsibility. In fact, it really is a whole of society failure to have people who end up in that circumstance, and we really need to look at our society across the board to figure out what went wrong and what we ought to be doing differently, and to try and pin it on anyone
player in the ecosystem. To say Facebook was the reason why this particular person pulled the trigger, I really think kind of denies the reality of how we've all contributed to a society that can lead to these outcomes and.
Explain how Section two thirty would work here As a.
Defense, So the planets started arguing that Facebook radicalized the shooter, and Facebook itself normally doesn't publish content itself that it authored. Its primary way of engaging in content distribution is to gather third party content like the Facebook post that we all make as the every day activity, and then share that with an audience. And so the content that allegedly
radicalized the shooter didn't come from Facebook. Facebook most was the mechanism by which some author matched with the shooter. And so it's impossible really to describe Facebook's role in radicalizing the shooter without talking about the content that regalized the shooter, all of which came from a third party. And if third party content can't create liability for Facebook section two, there you will apply a note.
Michael Bloomberg, the founder and majority owner of Bloomberg LP, the parent company of Bloomberg Radio, is a donor to groups that support gun control, including every Town for Gun Safety, which represented the plaintiffs in this case in the lower courts. Coming up next on the Bloomberg Law Show, I'll continue this conversation with Professor Eric Goleman of Santa Clara University Law School. We'll talk more about the algorithms that issue
in the case. Survivors and families of the victims of the mass shooting in a Buffalo supermarket three years ago are suing Internet giants Google, Meta, Amazon, and Reddit for publishing racist content that allegedly motivated the shooter to commit a hate crime targeting people in an historically black neighborhood hundreds of miles from his Home. The plaintiffs blame the proprietary algorithms the companies use.
Putting people in groups. There's something called confirmation bias. And when you put people in groups where they're getting these white supremacist theories and these calls to violence and these calls to make a race war and think that that's the right thing. That's what happened to Peyton Gendron, and that's what this lawsuit is about him.
But the Internet companies say they're protected from liability under the law, which says no interactive computer service provider can be treated as the publisher or speaker of any information provided by another information content provider. I've been talking with Professor Eric Goldman of Santa Clara University Law School. Eric tell us about the plaintiff's claims about the algorithms.
So, in order for the shooter to get access to the radicalizing content, somebody has to upload it, and then Facebook has to make the match to present that to the reader to the shooter in this case, and so plainists have often claimed that it was the algorithm Facebook created that led to this radicalization. Now, there are a couple of problems with that theory. One is it was
expressly rejected. In the Fourth Circuit opinion that I mentioned earlier, Federal Peller Court said that's not a way of getting around Section two thirty. The other piece, and the Appellate Court mentioned this as well, that in order for Facebook to do anything differently to prevent that requalization, it would have to re architect its entire service. In other words, there's not a way of it just changing the algorithm to eliminate the risk of rattle and clovation with changing
any other element of its algorithm. So, in other words, there's no easy fix here for Facebook. Facebook work the way it was designed to work. It may matches between people and content they might be interested in. Unfortunately, some people are interested in content that is antisocial. Facebook's algorithm treats that the same as all other content.
The director of the Center for Democracy and Technologies Free Expression Project, which filed in a MEEKS brief in support of the social media companies, said if the automatic ranking and delivery of content is separated from the Section two thirty liability shield, it will suddenly make many pieces of content open to liability and incentivized platforms that use automated ranking systems to deliver content to suppress or eliminate delivery
of content they're worried about. So basically, would there be a chilling effect if that happens.
I think that's a riff on what I was just saying, and maybe said better than I did. But let me take another cut at it. So, if Facebook is liable for radicalization content, and if and I'm going to hypothesize this, it's true, people might debate it. There is no way for Facebook to know which content is radicalized in or not, or to properly separate that from the pro social content
that it would prefer to disseminate. Then Facebook basically becomes the financial guaranteur of any harms that are caused by people that can be traced back to someone's consumption content on Facebook. We've seen so many cases in this genre. This is not a unique case in Buffalo. Many times, victims of shootings have claimed that the shooter was exposed to content on social media, and social media should be
responsible for this, and that's just not tenable. Social media cannot act as the financial guaranteur of all the shootings that take place in the world, so then they have to do something different. If they become the financial guarantur they cannot exist in their current form.
The easiest path for the Appellate Division right now in this case is to allow the case to go forward and to say, you know, the facts have to be developed before we can make a decision on whether or not this is subject to the state's product liability law.
So that's essentially what the lower court did. The lower court basically said, I understand Facebook's defenses, and yet I need to see more facts than before I'm prepared to make the conclusion that products valuability applies, or even that Section two thirty applies. So one possibility is that pel Court could say that District Court was right. Let the case continue to evolve in its ordinary manner, and once we get more information, then we'll be able to figure
out what defenses to actually apply in the circumstance. Most cases evolving section two thirty don't do that. Most cases in section two thirty. And at the motion the Smiths, because it is obvious on the face there's nothing that would be produced in a discovery that would show that Section two thirty didn't apply. And here I don't think anyone's contesting the Facebook provided or authored the radicalizing content. The allegation is that it matched the shooter with their
party content. So everything that the court needs to know about Section two theories, I think or are you available to it. So punting on that question actually would be a disservice to Facebook and I think to all other social media that expect to be able to avoid the liability for things that they're not legally responsible for.
It's interesting that the plaintiffs to your chose state court were in the other case you were discussing the plaintiffs chose federal court.
You know, planets get to pick the venue that they think is the best place for them to achieve justice. So it's completely logical that they might conclude that New York state courts is opposed to New York federal courts
are a better venue for them. But I would hope that the New York state courts will still take a look at what's going on in the rest of the country, because this is not a novel issue across the country, and the case law is very clear that the New York state trial courts got a wrong.
How important is this case as a precedent, let's say, for other cases around the country, and is it important or is it just, you know, another one of the cases that you've been talking about. Is there any import to it?
So every case that involves the situation where social media allegedly contributed to some offline harm is important because if those cases succeed, then all social media services face an extreme amount of liability that may make them financially untenable. So it's like the social media services have to bat a thousand in these cases. If they only bout nine to ninety nine and one case gets through, that one
case could potentially change the entire Internet ecosystem. So if this case ends up like essentially all the others, and the planiffs don't succeed, then this case became unimportant. It's just one of the many. If for whatever reason, this case does succeed, then this could be the case that
reshapes the Internet. And the district court opinion in this case was so troubling, and because it clearly deviated from pressing around the country, and because it basically said I can't make any judgments, even though many people thought that the court had all the facts that needed to make a judgment. And so if that's the case, it opens up the door for unpredictable results, and that unpredictable is actually its own form of danger to the Internet.
Any final thoughts.
You know, the Buffalo shoe was horrific. You know, all these mass murders are just heartbreaking. You know, they're a sign of pathology in our country. They just absolutely tear me up. And the fact that the shooter was espousing the great replacement theory is even more troubling that this theory continues to find an audience in our country when
it's both a factual but it's also deeply corrosive. I will add that want to blame Facebook for the dissemination of the great replacement theory, but many leading figures in our country, whether that's media pundits or actual government officials, have also dabbled in the great replacement theory, and so to me, the idea that Facebook ratoclides this shooter misses the whole ecosystem.
We'll see if the New York Appellate Court here deviates from the way most other courts have ruled. Thanks so much, Eric, that's professor Eric Goldman of a Santa Clara University School of Law. Coming up next on the Bloomberg Law Show. The growing trend of presidents using temporary appointments for top roles in government. President Trump has installed another loyalist in one of the most powerful US attorney's offices in the country.
Former Fox News host Janine Piro has taken over as the interim US Attorney for DC from Ed Martin, who was the interim US Attorney for close to one hundred and twenty days. The interim nature of both appointments test the bounds of a federal statute governing temporary officials and reflects a growing trend of presidents leaning on temporary appointments for top roles in government. My guest is Anne Joseph O'Connell, a professor at Stanford Law School who specializes in political appointments.
Will you explain the provisions that are at play when a US attorney is given an interim appointment.
So, when there's not a Senate confirmed US attorney in a particular district, there are two statutes that can provide for temporary service. So the first, what we call interim US attorneys, is a specific provision in the United States Code called Section five forty six, and that provision allows for the Attorney General to choose a temporary interim US attorney for the district in which there is not a
Senate confirmed person. There are no restrictions on who the Attorney General can pick, so they don't have to pick someone who's within the Department of Justice, unlike this other
statutory provision. And so the Attorney General can pick an interim US attorney and that person can serve for one hundred and twenty days under section five forty six, and then after that one hundred twenty days expires, the district court may right may appoint a US attorney to serve until there is a Senate confirmed one.
So what happened here is the term for ed Martin was about to run out, and before the district court could appoint someone, President Trump nominated as another interim US attorney. Jeanine Piro I mean, is that according to the.
Rules, Well, it depends whether you're looking at the text of the rule or the intent of the provision. Under the text, I think it's permitted. The provision Section five forty six does not explicitly bar successive one hundred twenty day appointments. And I should also note that even if the district courts had chosen someone, the President in all likelihood could have fired the district court pick and picked another interim US attorney or turned this other statue the
Federal Vacancy's Reform Act. So I think the text permits successive appointments. There's no explicit bar on it. There's no provision that provides for a penalty if you see these successive appointments end. In the past, we have seen successive appointments. Now that's the text. When Congress reimposed time limits, so put back the one hundred and twenty day constraint on
how long interim US attorneys can serve. They did that in two thousand and seven after a US attorney scandal involving Attorney General Darto Gonzalez, who had fired a number of US attorneys. And the scandal was not whether the president or the Attorney General could fire US attorneys. It was about the reason that was given for those firings. But at the time that it happened, there was no limit on how long interim US attorneys could serve. And
this had occurred after September eleventh. There was a period of years where there was kind of no time limit. But then in two thousand and seven, Congress put back this time limit, and I think the putting back of the time limit suggests strongly that the intent of Congress was not to have endless interim appointments of US attorneys.
And was that the intent of the Framers as well. In other words, they wanted the Senate to advise and consent.
So we have.
Had Senate confirmed appointments since the start of the United States Constitution under the appointments Clause, and interestingly, we've had statutes that provide for temporary appointments from about the same time. The very first Vacancies Act, which applied generally, was in seventeen ninety two, and there was another one in seventeen
ninety five. There it gets a little tricky about kind of what were the time limits, because in the seventeen ninety five Act, Congress imposed the six month time limit on acting officials.
Here a lot of actings. Have US presidents been using interim positions more and more so I have done.
Some research on recent administrations and looking at the very highest level of positions, so the heads of agencies, cabinet secretaries in particular, and definitely recent administrations, and that would include President Obama, President Trump's first term and President Biden have heavily relied on acting officials, much more, I would say, than their predecessors in terms of modern government.
What's the reason for this? I know that President Trump in the past has said, you know, he likes the idea of it.
Is it because it's.
Difficult to get people through the Senate? I mean, what's the reason then they're using it more.
I think there are several reasons. The first is it takes time to nominate people. Though I will say that President Trump in his second term has a much faster nominations pace than he had in his first term. I mean six or sevenfold more nominations at one hundred day mark in his second term than in his first term. But nominations still takes time to find the people and
to formally nominate them. There's a vetting that occurs, and I think for President Trump that vetting has really focused on loyalty, whereas in other administrations might be focusing on ethical considerations, expertise and the like. And then there's the confirmation process. And so even as we have now and even as we did for President Biden, have a Senate controlled by the same party as the White House. It's
still takes time because individual senators. Even though there's no longer a sixty vote threshold to move a nomination to a confirmation vote that went away in November twenty thirteen, individual senators can still delay the confirmation's process, right, And we saw this with Senator Tuberville in the Biden administration. We've seen this in other cases as well. Right, Democrats do it to Republicans. Republicans do it to Democrats, and that can slow it down. And so I think both
the nominations and confirmation process make acting officials attractive. But I would also say that what I think is a large reason is that acting officials don't require Senate confirmation, and so you can pick people to serve in the
temporary capacity that the Senate might not confirm. So in President Trump's first term, the Republicans, as reported by the media, did not want to confirm Ken Kucinelli to a senior post in the Department of Homeland Security, but through acting appointments and delegations of authority, mister Kuchinelli was able to
serve in very high profile roles in DXS. And I think a similar story right for Ed Martin that Republicans balked and he was able through the interim provision to serve, whereas you know, if he had just been put up as a nominee, not as an interim, he might not have gotten true.
Has President Trump pushed the limits of the Vacancy's Act further than other presidents?
I think the Vacancy's Act is pretty capacious. It allows the government to function. I think the Vacancies Act should be reformed in certain ways because I think it allows more than perhaps what we want for an accountable government. But I think in this term, President Trump is really pushing an argument that no previous president has pushed, which is that he believes that if the Vacancy's Act does not apply, he has inherent article to authority name acting officials.
And this has come up in the Inner America Foundation and the African Development Foundation, these are entities not covered by the Vacancies Act, where President Trump has named acting officials to Senate confirmed Board positions relying on claimed inherent article to authority. It seems with the firing of the Librarian of Congress that if the Library of Congress is not subject to the Vacancies Act, and I don't think
it is President Trump. The White House is indicated that there is Article two authority to name an acting librarian, and I think this is an outrageous claim. We have an appointments clause, we have a recess appointments clause. It is true that the Vacancies Act and specific agency provisions like section five forty six for interim US attorneys, do allow vast use of temporary appointments, but outside of those, presidents need to rely on the appointments clause or the recess appointment.
So some are saying that with the US attorney, for example, with Piero becoming the second interim US Attorney for DC, that defendants might be able to challenge some of the you know, prosecutions, et cetera under her because of this issue.
I think they would definitely have standing to sue. And I think their argument would be that under five forty six, successive one hundred and twenty day appointments are not allowed. And I think that is a plausible argument. I just don't think it's a winning argument. I think that courts will look at the text. Right as Justice Kagan says, we're all textless now, that the courts will look at the text and see that successive one hundred and twenty
day appointments are not barred. It's happened in the past and permitted. Now, that doesn't mean it shouldn't change. I mean Congress could amend five forty six to make it clear that you get one one hundred twenty day appointment, then the district court gets to choose. And then I should note that there is an alternative to these successive
one hundred and twenty day appointments. While a nomination might be pending to a US attorney slot, is that I believe that the Federal Vacancy's Reform Act is available for temporary appointments to these positions. Now, the constraint why I think the Trump administration doesn't want to turn to the Vacancies Act is that under the Vacancy's Act, you can't just choose anyone. You can't choose an outsider like the current person. You can't choose an outsider to come in
as an acting US attorney. You either need to choose someone who is the first assistant to the US Attorney right, which is typically a career person. You need to choose someone who has Senate confirmed already to another position, so it has already kind of gone through the Senate process, or you have to choose someone who was in the agency for at least ninety days in the year before the vacancy.
So finally, you think there needs to be reform here.
I think the.
Law of governing temporary appointments are pretty broad and likely too broad. On the other hand, we want the government to function. We don't want the government to come to a standstill. And so finding the balance between getting Senate confirmed appointments through a dysfunctional appointments process and having the government function is tricky. And sometimes we're on the right side of the balance, and sometimes we're on the wrong
side of the balance. And so I think that Congress really has a role to play in constraining the use of acting an interim appointment.
So we'll start counting now to see if Janine Pirou is nominated to be the permanent US Attorney for DC before the one hundred and twenty days expires. Thanks so much for joining me today. That's Professor and Joseph O'Connell of Stanford Law School, and that's it for the edition of the Bloomberg Law Show. Remember you can always get
the latest legal news on our Bloomberg Law podcasts. You can find them on Apple Podcasts, Spotify, and at www dot bloomberg dot com, slash podcast slash Law, and remember to tune into The Bloomberg Law Show every weeknight at ten pm Wall Street Time. I'm June Grosso and you're listening to Bloomberg