Connect with us

Free Speech

Part IV: As Hate Speech Proliferates Online, Critics Want to See and Control Social Media’s Algorithms

Emily McPhie



Photo of Beto O'Rourke in April 2019 by Gage Skidmore used with permission

WASHINGTON, August 22, 2019 — Lurking at the corners over the renewed debate over Section 230 the Communications Decency Act is this question: Who gets to control the content moderation process surrounding hate speech?

Even as artificial intelligence is playing a greater role in content moderation on the big tech platforms, the public is still reeling from whether content moderation should facilitate free speech or contain harmful speech.

Around the time that Section 230 was passed, most of the discussion surrounding online platforms was based on a “rights framework,” Harvard Law Professor Jonathan Zittrain told Broadband Breakfast. Aside from some limited boundaries against things like active threats, the prevailing attitude was that more speech was always better.

“In the intervening years, in part because of how ubiquitous the internet has become, we’ve seen more of a public health framework,” Zittrain continued. This perspective is concerned less about an individual’s right to speech and more about the harms that such speech could cause.

Misleading information can persuade parents to decide not to vaccinate their children or lead to violence even if the words aren’t a direct incitement, said Zittrain. The public health framework views preventing these harms as an essential part of corporate social responsibility.

Because these contrasting frameworks have such different values and vernaculars, reconciling them into one comprehensive content moderation plan is a nearly impossible task.

What’s the role of artificial intelligence in content moderation?

Another complication in the content moderation debate is that the sheer volume of online content necessitates the use of automated tools — and these tools have some major shortcomings, according to a recent report from New America’s Open Technology Institute.

Algorithmic models are trained on datasets that emphasize particular categories and definitions of speech. These datasets are usually based on English or other Western languages, despite the fact that millions of users speak different languages. Resulting algorithms are capable of identifying certain types of speech but cannot be holistically applied.

In addition, simply training an algorithm to flag certain words or phrases carries the risk of further suppressing voices that are already marginalized. Sometimes, the “toxicity” of a given term is dependent on the identity of the speaker, since many terms that have historically been used as slurs towards certain groups have been reclaimed by those communities while remaining offensive when used by others.

A 2019 academic study found that “existing approaches to toxic language detection have racial biases, and that text alone does not determine offensiveness.” According to the study, tweets using the African American English dialect were twice as likely to be labelled offensive compared to other tweets.

“The academic and tech sector are pushing ahead with saying, ‘let’s create automated tools of hate detection,’ but we need to be more mindful of minority group language that could be considered ‘bad’ by outside members,” said Maarten Sap, one of the researchers behind the study.

AI’s inability to detect nuance, particularly in regard to context and differing global norms, results in tools that are “limited in their ability to detect and moderate content, and this often results in erroneous and overbroad takedowns of user speech, particularly for already marginalized and disproportionately targeted communities,” wrote OTI.

Curatorial context is key: Could other activist groups create their own Facebook algorithm?

The problem is that hate speech is inherently dependent on context. And artificial intelligence, as successful as it may be at many things, is incredibly bad at reading nuanced context. For that matter, even human moderators are not always given the full context of the content that they are reviewing.

Moreover, few internet platforms provide meaningful transparency around how they develop and utilize automated tools for content moderation.

The sheer volume of online content has created a new question about neutrality for digital platforms, Zittrain said. Platforms are now not only responsible for what content is banned versus not banned, but also for what is prioritized.

Each digital platform must have some mechanism for choosing which of millions of things to offer at the top of a feed, leading to a complex curatorial process that is fraught with confusion.

This confusion could potentially be alleviated through more transparency from tech companies, Zittrain said. Platforms could even go a step further by allowing third party individuals and organizations to create their own formulas for populating a feed.

Zittrain envisioned Facebook’s default news feed algorithm as a foundation upon which political parties, activist groups, and prominent social figures could construct their own unique algorithms to determine what news should be presented to users and in what order. Users could then select any combination of proxies to curate their feeds, leading to a more diverse digital ecosystem.

Critics of YouTube say the platform’s autoplay pushes extreme content

But without such a system in place, users are dependent on platforms’ existing algorithms and content moderation policies — and these policies are much criticized.

YouTube’s autoplay function is a particularly egregious offender. A Wall Street Journal report found that it guided users towards increasingly extreme and radical content. For example, if users searched for information on a certain vaccine, autoplay would direct them to anti-vaccination videos.

The popular platform’s approach to content moderation “sounded great when it was all about free speech and ‘in the marketplace of ideas, only the best ones win,’” Northeastern University professor Christo Wilson told the Journal. “But we’re seeing again and again that that’s not what happens. What’s happening instead is the systems are being gamed and the people are being gamed.”

Automated tools work best in combating content that is universally objectionable

Automated tools have been found to be the most successful in cases where there is wide consensus as to what constitutes objectionable content, such as the parameters surrounding child sexual abuse material.

However, many categories of so-called hate speech are far more subjective. Hateful speech can cause damage other than a direct incitement to violence, such as emotional disturbance or psychic trauma with physiological manifestations, former American Civil Liberties Union President Nadine Strossen told NBC in a 2018 interview.

These are real harms and should be acknowledged, Strossen continued, but “loosening up the constraints on government to allow it to punish speech because of those less tangible, more speculative, more indirect harms … will do more harm than good.”

And attempts at forcing tech platforms to implement more stringent content moderation policies by making such policies a requirement for Section 230 eligibility may do more harm than good, experts say.

Democratic presidential candidate Beto O’Rourke’s newly unveiled plan to do just that would ultimately result in a ‘block first, ask questions later’ mentality, said Free Press Senior Policy Counsel Carmen Scurato.

“This would likely include the blocking of content from organizations and individuals fighting the spread of racism,” Scurato explained. “Removing this liability exemption could have the opposite effect of O’Rourke’s apparent goals.”

O’Rourke’s unlikely alliance with formal rival Sen. Ted Cruz, R-Texas, to take on Section 230 highlights just how convoluted the discussion over the statue has become.

Because the First Amendment’s guarantee of freedom of speech is a restriction on government action, it doesn’t help individuals critical of “censorship” by private online platforms.

It’s up to the platforms themselves — and the public pressure and marketplace choices within which they operate — to decide where to draw lines over hate speech and objectionable content on social media.

Section I: The Communications Decency Act is Born

Section II: How Section 230 Builds on and Supplements the First Amendment

Section III: What Does the Fairness Doctrine Have to Do With the Internet?

Section IV: As Hate Speech Proliferates Online, Critics Want to See and Control Social Media’s Algorithms

Emily McPhie was Assistant Editor with Broadband Breakfast. She studies communication design and writing at Washington University in St. Louis, where she is a news editor for campus publication Student Life. She is a founding board member of Code Open Sesame, an organization that teaches computer skills to underprivileged children in six cities across Southern California.


Supreme Court Declares Trump First Amendment Case Moot, But Legal Issues For Social Media Coming

Benjamin Kahn



Photo of Justice Clarence Thomas in April 2017 by Preston Keres in the public domain

April 5, 2021—Despite accepting a petition that avoids the Supreme Court deliberating on whether a president can block social media users, Justice Clarence Thomas on Monday issued a volley that may foreshadow future legal issues surrounding social media in the United States.

On Monday, the Supreme Court sent back to a lower court and ruled as moot a lawsuit over whether former President Donald Trump could block followers on Twitter, after accepting a petition by the federal government to end the case because Trump wasn’t president anymore.

The case dates back to March 2018, when the Knight First Amendment Institute and others brought a case against former president Trump in the Southern District of New York for blocking users based on their political views, arguing the practice is a violation of the first amendment.

The lower court judge agreed, and the decision was upheld by the United States Court of Appeals.

In accepting the petition by the government, Justice Thomas stated that adjudicating legal issues surrounding digital platforms is uniquely difficult. “Applying old doctrines to new digital platforms is rarely straightforward,” he wrote. The case in question hinged on the constitutionality of then-President Trump banning people from interacting with his Twitter account, which the plaintiff argued was a protected public forum.

Thomas stated that while today’s conclusion was able to be vacated, that likely would not be the case in the future. He went on to say that digital platforms exercise “concentrated control of so much speech in the hands of a few private parties.”

He continued: “We will soon have no choice but to address how our legal doctrines apply to highly concentrated, privately owned information infrastructure such as digital platforms.”

Even though Facebook and Google were not the platforms in question in this case, Thomas pointed to them as “dominant digital platforms” and stated that they have “enormous control over speech.” He stated that Google, Facebook, and Twitter have the capabilities to suppress information and speech at will, and referenced the “cataclysmic consequences” for authors that Amazon disagrees with.

Thomas also rejected the notion that other options exist.

“A person always could choose to avoid the toll bridge or train and instead swim the Charles River or hike the Oregon Trail. But in assessing whether a company exercises substantial market power, what matters is whether the alternatives are comparable.”

Continue Reading

Section 230

Sen. Mike Lee Promotes Bills Valuing Federal Spectrum, Requiring Content Moderation Disclosures

Tim White



Screenshot of Mike Lee taken from Silicon Slopes event

April 5, 2021 – Sen. Mike Lee, R-Utah, said Friday spectrum used by federal agencies is not being utilized efficiently, following legislation he introduced early last year that would evaluate the allocation and value of federally-reserved spectrum.

The Government Spectrum Valuation Act, or S.553 and introduced March 3, directs the National Telecommunications and Information Administration to consult with the Federal Communications Commission and Office of Management and Budget to estimate the value of spectrum between 3 kilohertz and 95 gigahertz that is assigned to federal agencies.

Lee spoke at an event hosted by the Utah tech association Silicon Slopes on Friday about the legislation, in addition to other topics, including Section 230.

Some bands on the spectrum are reserved for federal agencies as they need it, but it’s not always managed efficiently, Lee said. Some are used by the Department of Defense for ‘national security,’ for example, but when asked what that spectrum is used for, we’re told, ‘we can’t tell you because of national security,’ he said.

“Just about everything we do on the internet is carried out through a mobile device, and all of that requires access to spectrum,” he said.

He said that lives are becoming more affected and enhanced by our connection to the internet, often through a wireless connection, which is increasing the need for the government to efficiently manage spectrum bandwidth, he said. Some of the bands are highly valuable, he said, comparing them to the “beach front property” of spectrum.

Legislation changing Section 230

Lee also spoke on Section 230, a statute that protects online companies from liability for content posted by their users. It’s a hot topic for policymakers right now as they consider regulating social media platforms.

Both Republicans and Democrats want more regulation for tech companies, but for different reasons. Democrats want more moderation against alleged hate speech or other content, citing the January 6 riot at the Capitol as one example of not enough censorship. Republicans on the other hand, including Lee, allege social media companies censor or remove right-leaning political content but do not hold the same standard for left-leaning content.

Lee highlighted that platforms have the right to be as politically-biased as they want, but it’s a problem when their terms of service or CEOs publicly state they are neutral, but then moderate content from a non-neutral standpoint, he said.

Lee expressed hesitation about repealing or changing Section 230. “If you just repealed it altogether, it would give, in my view, an undo advantage to big market incumbents,” he said. One solution is supplementing Section 230 with additional clarifying language or new legislation, he said.

That’s why he came up with the Promise Act, legislation he introduced on February 24 that would require the disclosure of rules for content moderation, and permit the Federal Trade Commission to take corrective action against companies who violate those disclosed rules. “I don’t mean it to be an exclusive solution, but I think it is a reasonably achievable step toward some type of sanity in this area,” he said.

Senator Amy Klobuchar, D-Minn., and a couple of her colleagues also drafted Section 230 legislation that would maintain the spirit of the liability provision, but would remove it for paid content.

Continue Reading

Section 230

Pressed by Congress, Big Tech Defends Itself and Offers Few Solutions After Capitol Riot

Tim White



Photo of Google CEO Sundar Pichai from a December 2018 hearing before the House Judiciary Committee by Drew Clark

March 26, 2021 – The heads of the largest social media companies largely defended their platforms, reiterated what they’ve done, and offered few solutions to the problems that ail them during a congressional hearing Thursday.

But, under harsh questioning from the House Energy and Commerce Committee, none of the CEOs of Google, Facebook or Twitter were given chance to respond to questions for more than 30 to 60 seconds on a given topic.

The hearing was about misinformation on social media in the fallout of the January 6 Capitol riot. The CEOs said dealing with the problem of dis- and misinformation on their platforms is more difficult than people think.

“The responsibility here lies with the people who took the actions to break the law and do the insurrection,” Facebook CEO Mark Zuckerberg said in response to a question about whether the platforms were to blame for the riot.

“Secondarily, also, the people who spread that content, including the president, but others as well, with repeated rhetoric over time, saying that the election was rigged and encouraging people to organize. I think those people bare the primary responsibility as well,” Zuckerberg said.

Zuckerberg added that “polarization was rising in America long before social networks were even invented,” he said. He blamed the “political and media environment that drives Americans apart.”

A ‘complex question’ of fault

Google CEO Sundar Pichai said it’s a “complex question” in response to the question of who’s at fault for the riot. Twitter CEO Jack Dorsey, however, was more direct: “Yes, but you also have to take into consideration a broader ecosystem; it’s not just about the technology platforms we use,” he said.

It was the first time Zuckerberg, Dorsey and Pichai appeared on Capitol Hill since the January 6 insurrection at the U.S. Capitol. The hearing was spurred by the riot and the turbulent presidential election that concluded in Joe Biden’s win and Donald Trump’s ban from Twitter and Facebook. Congress has turned their eye toward the social media companies for several months on possible Section 230 reform to address the alleged problems in the tech industry.

“Our nation is drowning in misinformation driven by social media. Platforms that were once used to share kids with grandparents are all-too-often havens of hate, harassment and division,” said Rep. Mike Doyle, D-Penn., chairman of the Communications and Technology subcommittee, who led the hearing. Doyle alleged the platforms “supercharged” the riot.

Both Democratic and Republican members of the committee laid out a variety of grievances during the five-hour meeting, and while they didn’t all share the same concerns, all agreed that something needs to be done.

“I hope you can take away from this hearing how serious we are, on both sides of the aisle, to see many of these issues that trouble Americans addressed,” Doyle said.

Congressional concerns

On the left side of the political aisle the main criticism against the tech giants was the spread of misinformation and extremism, including COVID-19 vaccines, climate change and the 2020 presidential election that Trump alleged was rigged against him.

“It is not an exaggeration to say that your companies have fundamentally and permanently transformed our very culture, and our understanding of the world,” said Rep. Jan Schakowsky, D-Illinois. “Much of this is for good, but it is also true that our country, our democracy, even our understanding of what is ‘truth’ has been harmed by the proliferation and dissemination of misinformation and extremism,” she said.

“Unfortunately, this disinformation and extremism doesn’t just stay online, it has real-world, often dangerous and even violent consequences, and the time has come to hold online platforms accountable,” said Rep. Frank Pallone, D-N.J.

From the right, Republican members voiced concerns about too much censorship, easy access to opioids, and the harm on children they said social media has.

“I’m deeply concerned by your decisions to operate your companies in a vague and biased manner, with little to no accountability, while using Section 230 as a shield for your actions and their real-world consequences,” said Rep. Bob Latta, R-Ohio. “Your companies had the power to silence the president of the United States, shut off legitimate journalism in Australia, shut down legitimate scientific debate on a variety of issues, dictate which articles or websites are seen by Americans when they search the internet,” he said.

“Your platforms are my biggest fear as a parent,” said Rep. Cathy McMorris Rodgers, R-Washington, expressing frustration over the impact that social media has on children. “It’s a battle for their development, a battle for their mental health, and ultimately, a battle for their safety,” she said, citing a rise of teen suicides since 2011. “I do not want you defining what is true for them, I do not want their future manipulated by your algorithms,” she said.

Platforms say it’s challenging, reiterate initiatives

In response to the many criticisms, Zuckerberg made it clear that while moderating content is central to address misinformation, it is important to protect speech as much as possible while taking down illegal content, which he said can be a huge challenge. As an example, bullying hurts the victim but there’s not a clear line where we can just censor speech, he said.

Pichai said that Google’s mission is about organizing and delivering information to the world and allowing free expression while also combatting misinformation. But it’s an evolving challenge, he said, because approximately 15 percent of google searches each day are new, and 500 hours of video are uploaded to YouTube every minute. To reinforce that point, he cited the fact that 18 months ago no one had heard of COVID-19, and in 2020 ‘coronavirus’ was the most trending search.

Dorsey expressed a similar sentiment about the evolving challenge of balancing freedom of expression with content moderation. “We observe what’s happening on our service, we work to understand the ramifications, and we use that understanding to strengthen our operations. We push ourselves to improve based on the best information we have,” he said.

The best way to face new challenges is to narrow down the problem to have the greatest impact, Dorsey said. For example, disinformation is a broad concept, and we focused on disinformation leading to offline harm, he said. Twitter worked on three specific categories, he said, these included manipulated media, public health and civic integrity.

“Ultimately, we’re running a business, and a business wants to grow the number of customers it serves. Enforcing a policy is a business decision,” Dorsey said.

Dorsey noted Twitter’s new Bluesky project, a decentralized internet protocol that various social media companies would be able to utilize, rather than being owned by a single company. It will improve the social media environment by increasing innovation around business models, recommended algorithms, and moderation controls in the hands of individuals instead of private companies, he said. But others already working in a similar technology space say the project is not without its problems.

On Section 230 reform

On the question of changing Section 230 of the Telecommunications Act, which grants social media companies immunity from liability for user-generated content, Zuckerberg suggested two specific changes: Platforms need to issue transparency reports about harmful content, and need better moderation for content that is clearly illegal. These changes should only affect large social media platforms, he said, but did not specify the difference between a large and small platform.

Dorsey said those may be good ideas, but it could be difficult to determine what is a large and small platform, and having those stipulations may incentivize the wrong things.

When asked about Instagram’s new version for children, Zuckerberg confirmed it was in the planning stage and many details were still being worked out.

Several Democrats raised concerns about minority populations, citing as one example the March 16 shooting in Atlanta that killed eight people including several Asian American women. Rep. Doris Matsui, D-Cal., asked why various hashtags such as #kungflu and #chinavirus were not removed from Twitter.

Dorsey responded that Twitter does take action against hate speech, but it can also be a challenge because it’s not always simple to distinguish between content that supports an idea and counter speech that condemns the support of that idea.

The tech leaders were asked by multiple members about the platform algorithms failing to catch specific instances of content moderation. Democrats referred to examples of posts containing misinformation or hate speech, while Republicans used examples of conservative-based content being removed.

Both Zuckerberg and Dorsey said that their systems are not perfect and it’s not realistic to expect perfection. Some content will always slip by our radars that we have to address individually, Zuckerberg said.

In response to Rep. Steve Scalise’s reference to a 2020 New York Post story about Hunter Biden that was taken down, Dorsey said we have made mistakes in some instances.

Editor’s Note: This story has been revised to add in a second paragraph that more accurately captured the fact that, while the tech executives offered few solutions, they were given little opportunity to do so by members of Congress. Additionally, the word “secondarily” was added back into Facebook CEO Mark Zuckerberg’s statement about who bore responsibility for the insurrection.

Continue Reading


Signup for Broadband Breakfast

Get twice-weekly Breakfast Media news alerts.
* = required field