(Note: I did this interview with Mathew Ingram in late February – before the recent flood of new anti-Section 230 activities in DC that I still need to blog. I always planned to share it here, but the problem was considered that Burned Back The pandemic has killed more than 120,000 Americans and closed our country. I share it now because so many people in DC give priority to destroying Section 230 over the many other devastating problems in our country.)
Ingram:… Eric, could you first give us a brief overview of your thoughts on section 230 and whether you think it needs to be changed or not – and if so, how or if not, why not?
Goldman: The main arguments for § 230 have not changed since 1996:
First, Congress passed Section 230 because it couldn't imagine what the Internet could be. It remains so today. Even though we now have internet-established companies and a well-developed internet architecture, we still cannot reliably predict what the internet could look like in 25 or even the next 10 years. I am confident that the internet will evolve in unexpected and social ways advantageous ways. As in 1996, Section 230 provides the legal infrastructure for the next generation of innovations.
Second, Section 230 has reduced barriers to entry into the UGC (User Generated Content) industry. In 1996, this secured a broad and diverse number of new market participants. Although we now have Internet operators who would be happy to see their competition thwarted, Section 230 still keeps the door open for newcomers. The next Google or Facebook killer will only appear with the support of Section 230.
Third, section 230 removed the “moderator's dilemma” – the concern that Internet companies will not attempt to moderate content if they are held responsible for errors in their moderation efforts. When section 230 is limited, the moderator's dilemma comes into play again, and internet companies make decisions that we don't want to have – such as simply eliminating certain categories of UGC, as it is impossible to manage those categories perfectly.
Since the main reasons for section 230 apply today as they did in 1996, I continue to support the law as it is. However, I think Congress could improve its effectiveness by coupling it with a federal anti-SLAPP law in cases where § 230 exclusions are brought for unsocial or censorship purposes.
Ingram: Thanks, Eric. In my conversation with James Grimmelmann of Cornell on Galley, he argued that section 230 was out of date and in need of change because, among other things, he said that it was “too strong if it was applied to really malicious actors who incredibly upset their user populations Publish harmful content that is technically from third parties so that the platform itself is immune. “How would you react to that?
Goldman: It is true that we saw some rogue actors relying on Section 230 to advance their anti-social agendas. Nevertheless, I disagree with Prof. Grimmelmann's conclusions for at least three reasons.
First, Section 230 does not apply to federal law enforcement measures. Rogue actors can and will be defeated with these existing crimes.
Second, other law enforcement agencies have been quite smart in extending their legal authority to rogue actors. For example, Section 230 has not prevented the FTC from closing down revenge porn operators.
Third, fraudulent websites rarely last very long. They quickly catch fire due to a combination of poor public relations, legal risk and monetization problems. Indeed, providers who defraud websites often intervene to terminate services – and these providers often rely on section 230 to make their termination decisions.
Ingram: Thanks, Eric. When I spoke to Olivier Sylvain of Fordham Law earlier this week, he repeated arguments he made in an essay for the Knight First Amendment Institute discussing how Section 230 implementation missed an important point – namely, how platforms Like Facebook Use data to engage users and shape their experience, and not just publish their content. He argues that the information we see could be "qualitatively different or, worse, disproportionately harmful," but under current law, courts can never judge whether this different treatment is illegal or otherwise harmful. How would you react to this kind of argument?
Goldman: All publishers try to address their users and to shape their experiences in the normal publishing process. Obviously, different media enable different ways to address users and create experiences. Still, I don't get much valuable insight when I differentiate between the remix type of online platforms and other types of publishing processes.
Nonetheless, I took note of the scenarios in which the Facebook algorithms can result in ads with a discriminatory impact, even though the selected advertising criteria did not signal advertisers potential discrimination. In this case, the Facebook algorithms can produce discriminatory effects that would otherwise not have occurred. I think it is a difficult legal question to find out how section 230 applies to this fact.
Ingram: Thanks, Eric. When I spoke to him earlier today, former NSA Minister of Homeland Security and General Counsel, Stewart Baker, said section 230 needs to be changed because legal liability protection is essentially a subsidy and these large companies do not need this subsidy or earn more. He also argued that the major platforms suppress conservative language and therefore do not deserve protection because they violate people's right to speak. Any thoughts on any of these issues?
Goldman: It is incredibly short-sighted to want to reform Section 230 because it benefits Google and Facebook. The Internet ecosystem has a vibrant and extraordinarily long range of services that benefit from the Section 230 “subsidy”. Changes to section 230 would disrupt or eliminate the services in this long tail, with the unintended consequences of further anchoring Internet giants like Google and Facebook. In this short essay, I examine the "subsidy point" and the competitive effects of section 230.
Implications that internet companies only target "conservative language" are really not credible. All content moderation decisions prefer some content over other content. Persons who have not achieved the desired moderation results will routinely complain that the decisions against them were "biased". Due to the massive volume of Internet services, they provide all sorts of anecdotal evidence of “bias” towards any class or view of speakers. However, this does not prove that internet services were somehow biased against these speakers. Instead, the evidence just proves that Internet services make a large number of decisions to moderate content.
(I also remind readers of the worrying overlap between “conservative” language and repulsive language, which should definitely be moderated).
After all, internet companies do not violate people's right to speak. They apply their editorial standards to decide what content to publish. Users have no legal “right” to speak about Internet services. Instead, the obligation for Internet services to transmit language that they do not want to transmit would violate the constitutionally protected rights of Internet services to publish (or not to publish) content at its own discretion.
Ingram: Thanks, Eric. Danielle Citron and Benjamin Wittes have argued that section 230 protection should be limited to the extent that it is only available to those who make "reasonable efforts" to moderate the content on their platforms (although what needs to be defined is what appropriate is not spelled). Do you see any merit in this proposal?
Goldman: No. The proposal would inherently reduce the material scope of Section 230. Some may cheer on that; Not me. More importantly, the proposal would functionally remove the key procedural benefits that section 230 offers. I explain this angle here. As a result, the mechanisms grossly misunderstand the mechanisms that make section 230 effective as immunity. I think the proposal "adequacy requirement" is equivalent to the repeal of section 230, which is only packed in less obvious terms.
Ingram: Thanks, Eric. One last question: why has Section 230 been attacked by so many different parties in the past year or two? There appears to be an ongoing attack by the government and those with party political agendas, but also by some members of the legal community. Is it just that Facebook appears so dominant and is involved in so many negative things and nobody can think about what to do about it?
Goldman: We have seen a wide-ranging techlash in the past half a decade, and Section 230 has been involved in this techlash downdraft. What triggered the techlash? There are many causes, but a few specific incidents are noteworthy: Russia's interference in the 2016 presidential election, including manipulation of voters through social media; Facebook's data loss at Cambridge Analytica; and Nancy Pelosi's "cheap" video (which slowed her speech down to give the impression that she was drunk). The Nancy Pelosi video was perhaps the most damaging because Congressmen were shocked and angry that Facebook would not protect someone who looked like them from such an obviously wrong attack.
The enormous size and performance of Google and Facebook also contribute to the Techlash. We generally mistrust large and profitable companies. and anti-elite populism has affected public perception of Google and Facebook in some circles.
Regardless, regulators view Google and Facebook as a threat to their own political power, and regulators routinely push back such threats. In particular, section 230 provides a practical negotiating chip for Congress members trying to advance their regulatory agenda. If Facebook and Google estimate Section 230, Congress members increase their bargaining chip by endangering it.
Similarly, corporate competitors and enemies of Google and Facebook have targeted Section 230 as part of their own gamble of power. For example, technology companies like IBM and Oracle, and entertainment companies like 21st Century Fox and Disney have spoken out against Section 230. The two similarities of these companies: (1) They are essentially not exposed to UGC at first hand. So they have no downside when section 230 disappears, and (2) they see Google or Facebook as a threat to their business for various reasons (e.g., the total war of Oracle against Google, waged by Oracle against Google, and the long time of the entertainment company – long-term strategy to undermine the legal protection of UGC in every possible way). When other technology companies speak out against Section 230, regulators misleadingly claim that the "tech community is divided over Section 230", masking the unanimous rejection of Section 230 reform among those actually affected.