The California Assembly Bill 587 (AB 587) – an amendment to the Business and Professions Code relating to social media – has suffered a significant setback: the result of the legal challenge from X (the X Corp. v. Bonta case) is that the law’s core provisions have now been dropped.
We obtained a copy of the ruling for you here.
Treated as an unconstitutional censorship law by opponents, AB 587 was proposed by ten Democrat and one Republican member of the state Assembly and adopted in September 2022, pushed by Governor Gavin Newsom and Attorney General Rob Bonta as legislation ensuring “social media transparency.”
The US Court of Appeals for the Ninth Circuit sided with X and other opponents, arguing that the key demands imposed by the state were, in fact, unconstitutional, i.e., a violation of the First Amendment, and issued an injunction; and now the case has been settled.
One of those provisions that California’s Department of Justice agreed to remove as part of the deal required large social media companies to submit detailed reports about their platforms’ moderation practices to the state’s attorney-general, twice a year.
The intent was to force the companies to state if their terms of service include definitions of “hate speech or racism,” “extremism or radicalization,” “disinformation or misinformation,” “harassment,” and “foreign political interference,” and also provide those definitions.
Another contentious provision was the requirement to describe in detail how speech characterized in this way was “moderated,” as well as let the state authorities know about flagged content in these categories and what is being done about that – down to the number of times posts were removed for “disinformation,” etc.
The power of the attorney-general or the governor to have a hand in what kinds of “moderation” are implemented is where the law hit the First Amendment wall, that protects from government censorship.
What’s left of AB 587 is a requirement to submit the terms of service twice a year, describe in detail any changes, and how those terms are implemented; but the companies will not have to inform state officials or provide them with data related to “moderation” of “hate speech or racism,” “extremism or radicalization,” “disinformation or misinformation,” etc.
