Enlarge / Screenshot from the documentary Who Is Bobby Kennedy?
In a lawsuit that seems determined to ignore that Section 230 exists, Robert F. Kennedy Jr. has sued Meta for allegedly shadowbanning his million-dollar documentary, Who Is Bobby Kennedy? and preventing his supporters from advocating for his presidential campaign.
According to Kennedy, Meta is colluding with the Biden administration to sway the 2024 presidential election by suppressing Kennedy’s documentary and making it harder to support Kennedy’s candidacy. This allegedly has caused “substantial donation losses,” while also violating the free speech rights of Kennedy, his supporters, and his film’s production company, AV24.
Meta had initially restricted the documentary on Facebook and Instagram but later fixed the issue after discovering that the film was mistakenly flagged by the platforms’ automated spam filters.
But Kennedy’s complaint claimed that Meta is still “brazenly censoring speech” by “continuing to throttle, de-boost, demote, and shadowban the film.” In an exhibit, Kennedy’s lawyers attached screenshots representing “hundreds” of Facebook and Instagram users whom Meta allegedly sent threats, intimidated, and sanctioned after they shared the documentary.
Some of these users remain suspended on Meta platforms, the complaint alleged. Others whose temporary suspensions have been lifted claimed that their posts are still being throttled, though, and Kennedy’s lawyers earnestly insisted that an exchange with Meta’s chatbot proves it.
Two days after the documentary’s release, Kennedy’s team apparently asked the Meta AI assistant, “When users post the link whoisbobbykennedy.com, can their followers see the post in their feeds?”
“I can tell you that the link is currently restricted by Meta,” the chatbot answered.
Chatbots, of course, are notoriously inaccurate sources of information, and Meta AI’s terms of service note this. In a section labeled “accuracy,” Meta warns that chatbot responses “may not reflect accurate, complete, or current information” and should always be verified.
Perhaps more significantly, there is little reason to think that Meta’s chatbot would have access to information about internal content moderation decisions.
Techdirt’s Mike Masnick mocked Kennedy’s reliance on the chatbot in the case. He noted that Kennedy seemed to have no evidence of the alleged shadow-banning, while there’s plenty of evidence that Meta’s spam filters accidentally remove non-violative content all the time.
Meta’s chatbot is “just a probabilistic stochastic parrot, repeating a probable sounding answer to users’ questions,” Masnick wrote. “And these idiots think it’s meaningful evidence. This is beyond embarrassing.”
Neither Meta nor Kennedy’s lawyer, Jed Rubenfeld, responded to Ars’ request to comment.
Instagram users have started complaining on X (formerly Twitter) after discovering that Meta has begun limiting recommended political content by default.
“Did [y’all] know Instagram was actively limiting the reach of political content like this?!” an X user named Olayemi Olurin wrote in an X post with more than 150,000 views as of this writing. “I had no idea ’til I saw this comment and I checked my settings and sho nuff political content was limited.”
“Instagram quietly introducing a ‘political’ content preference and turning on ‘limit’ by default is insane?” wrote another X user named Matt in a post with nearly 40,000 views.
Instagram apparently did not notify users directly on the platform when this change happened.
Instead, Instagram rolled out the change in February, announcing in a blog that the platform doesn’t “want to proactively recommend political content from accounts you don’t follow.” That post confirmed that Meta “won’t proactively recommend content about politics on recommendation surfaces across Instagram and Threads,” so that those platforms can remain “a great experience for everyone.”
“This change does not impact posts from accounts people choose to follow; it impacts what the system recommends, and people can control if they want more,” Meta’s spokesperson Dani Lever told Ars. “We have been working for years to show people less political content based on what they told us they want, and what posts they told us are political.”
To change the setting, users can navigate to Instagram’s menu for “settings and activity” in their profiles, where they can update their “content preferences.” On this menu, “political content” is the last item under a list of “suggested content” controls that allow users to set preferences for what content is recommended in their feeds.
There are currently two options for controlling what political content users see. Choosing “don’t limit” means “you might see more political or social topics in your suggested content,” the app says. By default, all users are set to “limit,” which means “you might see less political or social topics.”
“This affects suggestions in Explore, Reels, Feed, Recommendations, and Suggested Users,” Instagram’s settings menu explains. “It does not affect content from accounts you follow. This setting also applies to Threads.”
For general Instagram and Threads users, this change primarily limits what content posted can be recommended, but for influencers using professional accounts, the stakes can be higher. The Washington Post reported that news creators were angered by the update, insisting that Meta’s update diminished the value of the platform for reaching users not actively seeking political content.
“The whole value-add for social media, for political people, is that you can reach normal people who might not otherwise hear a message that they need to hear, like, abortion is on the ballot in Florida, or voting is happening today,” Keith Edwards, a Democratic political strategist and content creator, told The Post.
Meta’s blog noted that “professional accounts on Instagram will be able to use Account Status to check their eligibility to be recommended based on whether they recently posted political content. From Account Status, they can edit or remove recent posts, request a review if they disagree with our decision, or stop posting this type of content for a period of time, in order to be eligible to be recommended again.”
Ahead of a major election year, Meta’s change could impact political outreach attempting to inform voters. The change also came amid speculation that Meta was “shadowbanning” users posting pro-Palestine content since the start of the Israel-Hamas war, The Markup reported.
“Our investigation found that Instagram heavily demoted nongraphic images of war, deleted captions and hid comments without notification, suppressed hashtags, and limited users’ ability to appeal moderation decisions,” The Markup reported.
Meta appears to be interested in shifting away from its reputation as a platform where users expect political content—and misinformation—to thrive. Last year, The Wall Street Journal reported that Meta wanted out of politics and planned to “scale back how much political content it showed users,” after criticism over how the platform handled content related to the January 6 Capitol riot.
The decision to limit recommended political content on Instagram and Threads, Meta’s blog said, extends Meta’s “existing approach to how we treat political content.”
“People have told us they want to see less political content, so we have spent the last few years refining our approach on Facebook to reduce the amount of political content—including from politicians’ accounts—you see in Feed, Reels, Watch, Groups You Should Join, and Pages You May Like,” Meta wrote in a February blog update.
“As part of this, we aim to avoid making recommendations that could be about politics or political issues, in line with our approach of not recommending certain types of content to those who don’t wish to see it,” Meta’s blog continued, while at the same time, “preserving your ability to find and interact with political content that’s meaningful to you if that’s what you’re interested in.”
While platforms typically update users directly on the platform when terms of services change, that wasn’t the case for this update, which simply added new controls for users. That’s why many users who prefer to be recommended political content—and apparently missed Meta’s announcement and subsequent media coverage—expressed shock to discover that Meta was limiting what they see.
On X, even Instagram users who don’t love seeing political content are currently rallying to raise awareness and share tips on how to update the setting.
“This is actually kinda wild that Instagram defaults everyone to this,” one user named Laura wrote. “Obviously political content is toxic but during an election season it’s a little weird to just hide it from everyone?”
On Monday, ChatGPT maker OpenAI detailed its plans to prevent the misuse of its AI technologies during the upcoming elections in 2024, promising transparency in AI-generated content and enhancing access to reliable voting information. The AI developer says it is working on an approach that involves policy enforcement, collaboration with partners, and the development of new tools aimed at classifying AI-generated media.
“As we prepare for elections in 2024 across the world’s largest democracies, our approach is to continue our platform safety work by elevating accurate voting information, enforcing measured policies, and improving transparency,” writes OpenAI in its blog post. “Protecting the integrity of elections requires collaboration from every corner of the democratic process, and we want to make sure our technology is not used in a way that could undermine this process.”
Initiatives proposed by OpenAI include preventing abuse by means such as deepfakes or bots imitating candidates, refining usage policies, and launching a reporting system for the public to flag potential abuses. For example, OpenAI’s image generation tool, DALL-E 3, includes built-in filters that reject requests to create images of real people, including politicians. “For years, we’ve been iterating on tools to improve factual accuracy, reduce bias, and decline certain requests,” the company stated.
OpenAI says it regularly updates its Usage Policies for ChatGPT and its API products to prevent misuse, especially in the context of elections. The organization has implemented restrictions on using its technologies for political campaigning and lobbying until it better understands the potential for personalized persuasion. Also, OpenAI prohibits creating chatbots that impersonate real individuals or institutions and disallows the development of applications that could deter people from “participation in democratic processes.” Users can report GPTs that may violate the rules.
OpenAI claims to be proactively engaged in detailed strategies to safeguard its technologies against misuse. According to their statements, this includes red-teaming new systems to anticipate challenges, engaging with users and partners for feedback, and implementing robust safety mitigations. OpenAI asserts that these efforts are integral to its mission of continually refining AI tools for improved accuracy, reduced biases, and responsible handling of sensitive requests
Regarding transparency, OpenAI says it is advancing its efforts in classifying image provenance. The company plans to embed digital credentials, using cryptographic techniques, into images produced by DALL-E 3 as part of its adoption of standards by the Coalition for Content Provenance and Authenticity. Additionally, OpenAI says it is testing a tool designed to identify DALL-E-generated images.
In an effort to connect users with authoritative information, particularly concerning voting procedures, OpenAI says it has partnered with the National Association of Secretaries of State (NASS) in the United States. ChatGPT will direct users to CanIVote.org for verified US voting information.
“We want to make sure that our AI systems are built, deployed, and used safely,” writes OpenAI. “Like any new technology, these tools come with benefits and challenges. They are also unprecedented, and we will keep evolving our approach as we learn more about how our tools are used.”
