Artificial Intelligence
Automated Content Moderation’s Main Problem is Subjectivity, Not Accuracy, Expert Says
With millions of pieces of content generated daily, platforms are increasingly relying on AI for moderation.

WASHINGTON, February 2, 2023 — The vast quantity of online content generated daily will likely drive platforms to increasingly rely on artificial intelligence for content moderation, making it critically important to understand the technology’s limitations, according to an industry expert.
Despite the ongoing culture war over content moderation, the practice is largely driven by financial incentives — so even companies with “a speech-maximizing set of values” will likely find some amount of moderation unavoidable, said Alex Feerst, CEO of Murmuration Labs, at a Jan. 25 American Enterprise Institute event. Murmuration Labs works with tech companies to develop online trust and safety products, policies and operations.
If a piece of online content could potentially lead to hundreds of thousands of dollars in legal fees, a company is “highly incentivized to err on the side of taking things down,” Feerst said. And even beyond legal liability, if the presence of certain content will alienate a substantial number of users and advertisers, companies have financial motivation to remove it.
However, a major challenge for content moderation is the sheer quantity of user-generated online content — which, on the average day, includes 500 million new tweets, 700 million Facebook comments and 720,000 hours of video uploaded to YouTube.
“The fully loaded cost of running a platform includes making millions of speech adjudications per day,” Feerst said.
“If you think about the enormity of that cost, very quickly you get to the point of, ‘Even if we’re doing very skillful outsourcing with great accuracy, we’re going to need automation to make the number of daily adjudications that we seem to need in order to process all of the speech that everybody is putting online and all of the disputes that are arising.’”
Automated moderation is not just a theoretical future question. In a March 2021 congressional hearing, Meta CEO Mark Zuckerberg testified that “more than 95 percent of the hate speech that we take down is done by an AI and not by a person… And I think it’s 98 or 99 percent of the terrorist content.”
Dealing with subjective content
But although AI can help manage the volume of user-generated content, it can’t solve one of the key problems of moderation: Beyond a limited amount of clearly illegal material, most decisions are subjective.
Much of the debate surrounding automated content moderation mistakenly presents subjectivity problems as accuracy problems, Feerst said.
For example, much of what is generally considered “hate speech” is not technically illegal, but many platforms’ terms of service prohibit such content. With these extrajudicial rules, there is often room for broad disagreement over whether any particular piece of content is a violation.
“AI cannot solve that human subjective disagreement problem,” Feerst said. “All it can do is more efficiently multiply this problem.”
This multiplication becomes problematic when AI models are replicating and amplifying human biases, which was the basis for the Federal Trade Commission’s June 2022 report warning Congress to avoid overreliance on AI.
“Nobody should treat AI as the solution to the spread of harmful online content,” said Samuel Levine, director of the FTC’s Bureau of Consumer Protection, in a statement announcing the report. “Combatting online harm requires a broad societal effort, not an overly optimistic belief that new technology — which can be both helpful and dangerous — will take these problems off our hands.”
The FTC’s report pointed to multiple studies revealing bias in automated hate speech detection models, often as a result of being trained on unrepresentative and discriminatory data sets.
As moderation processes become increasingly automated, Feerst predicted that the “trend of those problems being amplified and becoming less possible to discern seems very likely.”
Given those dangers, Feerst emphasized the urgency of understanding and then working to resolve AI’s limitations, noting that the demand for content moderation will not go away. To some extent, speech disputes are “just humans being human… you’re never going to get it down to zero,” he said.
Artificial Intelligence
Sen. Bennet Urges Companies to Consider ‘Alarming’ Child Safety Risks in AI Chatbot Race
Several leading tech companies have rushed to integrate their own AI-powered applications

WASHINGTON, March 22, 2023 — Sen. Michael Bennet, D-Colo., on Tuesday urged the companies behind generative artificial intelligence products to anticipate and mitigate the potential harms that AI-powered chatbots pose to underage users.
“The race to deploy generative AI cannot come at the expense of our children,” Bennet wrote in a letter to the heads of Google, OpenAI, Meta, Microsoft and Snap. “Responsible deployment requires clear policies and frameworks to promote safety, anticipate risk and mitigate harm.”
In response to the explosive popularity of OpenAI’s ChatGPT, several leading tech companies have rushed to integrate their own AI-powered applications. Microsoft recently released an AI-powered version of its Bing search engine, and Google has announced plans to make a conversational AI service “widely available to the public in the coming weeks.”
Social media platforms have followed suit, with Meta CEO Mark Zuckerberg saying the company plans to “turbocharge” its AI development the same day Snapchat launched a GPT-powered chatbot called My AI.
These chatbots have already demonstrated “alarming” interactions, Bennet wrote. In response to a researcher posing as a child, My AI gave instructions for lying to parents about an upcoming trip with a 31-year-old man and for covering up a bruise ahead of a visit from Child Protective Services.
A Snap Newsroom post announcing the chatbot acknowledged that “as with all AI-powered chatbots, My AI is prone to hallucination and can be tricked into saying just about anything.”
Bennet criticized the company for deploying My AI despite knowledge of its shortcomings, noting that 59 percent of teens aged 13 to 17 use Snapchat. “Younger users are at an earlier stage of cognitive, emotional, and intellectual development, making them more impressionable, impulsive, and less equipped to distinguish fact from fiction,” he wrote.
These concerns are compounded by an escalating youth mental health crisis, Bennet added. In 2021, more than half of teen girls reported feeling persistently sad or hopeless and one in three seriously contemplated suicide, according to a recent report from the Centers for Disease Control and Prevention.
“Against this backdrop, it is not difficult to see the risk of exposing young people to chatbots that have at times engaged in verbal abuse, encouraged deception and suggested self-harm,” the senator wrote.
Bennet’s letter comes as lawmakers from both parties are expressing growing concerns about technology’s impact on young users. Legislation aimed at safeguarding children’s online privacy has gained broad bipartisan support, and several other measures — ranging from a minimum age requirement for social media usage to a slew of regulations for tech companies — have been proposed.
Many industry experts have also called for increased AI regulation, noting that very little legislation currently governs the powerful technology.
Artificial Intelligence
Oversight Committee Members Concerned About New AI, As Witnesses Propose Some Solutions
Federal government can examine algorithms for generative AI, and coordinate with states on AI labor training.

WASHINGTON, March 14, 2023 – In response to lawmakers’ concerns over the impacts on certain artificial intelligence technologies, experts said at an oversight subcommittee hearing on Wednesday that more government regulation would be necessary to stem their negative impacts.
Relatively new machine learning technology known as generative AI, which is designed to create content on its own, has taken the world by storm. Specific applications such as the recently surfaced ChatGPT, which can write out entire novels from basic user inputs, has drawn both marvel and concern.
Such AI technology can be used to encourage cheating behaviors in academia as well as harm people through the use of deep fakes, which uses AI to superimpose a user in a video. Such AI can be used to produce “revenge pornography” to harass, silence and blackmail victims.
Aleksander Mądry, professor of Cadence Design Systems of Massachusetts Institute of Technology, told the subcommittee that AI is a very fast moving technology, meaning the government needs to step in to confirm the objectives of the companies and whether the algorithms match the societal benefits and values. These generative AI technologies are often limited to their human programming and can also display biases.
Rep. Marjorie Taylor Greene, R-Georgia, raised concerns about this type of AI replacing human jobs. Eric Schmidt, former Google CEO and now chair of the AI development initiative known as the Special Competitive Studies Project, said that if this AI can be well-directed, it can aid people in obtaining higher incomes and actually creating more jobs.
To that point, Rep. Stephen Lynch, D-Massachusetts., raised the question of how much progress the government has made or still needs in AI development.
Schmidt said governments across the country need to look at bolstering the labor force to keep up.
“I just don’t see the progress in government to reform the way of hiring and promoting technical people,” he said. “This technology is too new. You need new students, new ideas, new invention – I think that’s the fastest way.
“On the federal level, the easiest thing to do is to come up with some program that’s ministered by the state or by leading universities and getting them money so that they can build these programs.”
Schmidt urged lawmakers last year to create a digital service academy to train more young American students on AI, cybersecurity and cryptocurrency, reported Axios.
Artificial Intelligence
Congress Should Focus on Tech Regulation, Said Former Tech Industry Lobbyist
Congress should shift focus from speech debates to regulation on emerging technologies, says expert.

WASHINGTON, March 9, 2023 – Congress should focus on technology regulation, particularly for emerging technology, rather than speech debates, said Adam Conner, vice president of technology policy at American Progress at Broadband Breakfast’s Big Tech and Speech Summit Thursday.
Conner challenged the view of many in industry who assume that any change to current laws, including section 230, would only make the internet worse.
Conner, who aims to build a progressive technology policy platform and agenda, spent the past 15 years working as a Washington employee for several Silicon Valley companies, including Slack Technologies and Brigade. In 2007, Conner founded Facebook’s Washington office.
Instead, Conner argues that this mindset traps industry leaders in the assumption that the internet is currently the best it could ever be. This is a fallacy, he claims. To avoid this mindset, Conner suggests that the industry focus on regulation for new and emerging technology like artificial intelligence.
Recent AI innovations, like ChatGPT, create the most human readable AI experience ever made through text, images, and videos, Conner said. The penetration of AI will completely change the discussion about protecting free speech, he said, urging Congress to draft laws now to ensure its safe use in the United States.
Congress should start its AI regulation with privacy, anti-trust, and child safety laws, he said. Doing so will prove to American citizens that the internet can, in fact, be better than it is now and will promote future policy amendments, he said.
To watch the full videos join the Broadband Breakfast Club below. We are currently offering a Free 30-Day Trial: No credit card required!
-
Fiber3 weeks ago
‘Not a Great Product’: AT&T Not Looking to Invest Heavily in Fixed Wireless
-
Broadband Roundup2 weeks ago
AT&T Floats BEAD in USF Areas, Counties Concerned About FCC Map, Alabama’s $25M for Broadband
-
Big Tech3 weeks ago
House Innovation, Data, and Commerce Chairman Gus Bilirakis to Keynote Big Tech & Speech Summit
-
Big Tech2 weeks ago
Watch the Webinar of Big Tech & Speech Summit for $9 and Receive Our Breakfast Club Report
-
Big Tech2 weeks ago
Preview the Start of Broadband Breakfast’s Big Tech & Speech Summit
-
#broadbandlive2 weeks ago
Broadband Breakfast on March 8: A Status Update on Tribal Broadband
-
WISP4 weeks ago
Starry Group Files for Chapter 11 Bankruptcy
-
Broadband's Impact4 weeks ago
Community Engagement is Key to BEAD Grant Planning Process, Experts Say