Anjana Susarla, Michigan State University

Meta’s decision to change its content moderation policies by replacing centralized fact-checking teams with user-generated community labeling has stirred up a storm of reactions. But taken at face value, the changes raise the question of the effectiveness of Meta’s old policy, fact-checking, and its new one, community comments.

With billions of people worldwide accessing their services, platforms such as Meta’s Facebook and Instagram have a responsibility to ensure that users are not harmed by consumer fraud, hate speech, misinformation or other online ills. Given the scale of this problem, combating online harms is a serious societal challenge. Content moderation plays a role in addressing these online harms.

Moderating content involves three steps. The first is scanning online content – typically, social media posts – to detect potentially harmful words or images. The second is assessing whether the flagged content violates the law or the platform’s terms of service. The third is intervening in some way. Interventions include removing posts, adding warning labels to posts, and diminishing how much a post can be seen or shared.

Content moderation can range from user-driven moderation models on community-based platforms such as Wikipedia to centralized content moderation models such as those used by Instagram. Research shows that both approaches are a mixed bag.

Does fact-checking work?

Meta’s previous content moderation policy relied on third-party fact-checking organizations, which brought problematic content to the attention of Meta staff. Meta’s U.S. fact-checking organizations were AFP USA, Check Your Fact, Factcheck.org, Lead Stories, PolitiFact, Science Feedback, Reuters Fact Check, TelevisaUnivision, The Dispatch and USA TODAY.

Fact-checking relies on impartial expert review. Research shows that it can reduce the effects of misinformation but is not a cure-all. Also, fact-checking’s effectiveness depends on whether users perceive the role of fact-checkers and the nature of fact-checking organizations as trustworthy.

Crowdsourced content moderation

In his announcement, Meta CEO Mark Zuckerberg highlighted that content moderation at Meta would shift to a community notes model similar to X, formerly Twitter. X’s community notes is a crowdsourced fact-checking approach that allows users to write notes to inform others about potentially misleading posts.

Studies are mixed on the effectiveness of X-style content moderation efforts. A large-scale study found little evidence that the introduction of community notes significantly reduced engagement with misleading tweets on X. Rather, it appears that such crowd-based efforts might be too slow to effectively reduce engagement with misinformation in the early and most viral stage of its spread.

There have been some successes from quality certifications and badges on platforms. However, community-provided labels might not be effective in reducing engagement with misinformation, especially when they’re not accompanied by appropriate training about labeling for a platform’s users. Research also shows that X’s Community Notes is subject to partisan bias.

Crowdsourced initiatives such as the community-edited online reference Wikipedia depend on peer feedback and rely on having a robust system of contributors. As I have written before, a Wikipedia-style model needs strong mechanisms of community governance to ensure that individual volunteers follow consistent guidelines when they authenticate and fact-check posts. People could game the system in a coordinated manner and up-vote interesting and compelling but unverified content.

Content moderation and consumer harms

A safe and trustworthy online space is akin to a public good, but without motivated people willing to invest effort for the greater common good, the overall user experience could suffer.

Algorithms on social media platforms aim to maximize engagement. However, given that policies that encourage engagement can also result in harm, content moderation also plays a role in consumer safety and product liability.

This aspect of content moderation has implications for businesses that either use Meta for advertising or to connect with their consumers. Content moderation is also a brand safety issue because platforms have to balance their desire to keep the social media environment safer against that of greater engagement.

AI content everywhere

Content moderation is likely to be further strained by growing amounts of content generated by artificial intelligence tools. AI detection tools are flawed, and developments in generative AI are challenging people’s ability to differentiate between human-generated and AI-generated content.

In January 2023, for example, OpenAI launched a classifier that was supposed to differentiate between texts generated by humans and those generated by AI. However, the company discontinued the tool in July 2023 due to its low accuracy.

There is potential for a flood of inauthentic accounts – AI bots – that exploit algorithmic and human vulnerabilities to monetize false and harmful content. For example, they could commit fraud and manipulate opinions for economic or political gain.

Generative AI tools such as ChatGPT make it easier to create large volumes of realistic-looking social media profiles and content. AI-generated content primed for engagement can also exhibit significant biases, such as race and gender. In fact, Meta faced a backlash for its own AI-generated profiles, with commentators labeling it “AI-generated slop.”

More than moderation

Regardless of the type of content moderation, the practice alone is not effective at reducing belief in misinformation or at limiting its spread.

Ultimately, research shows that a combination of fact-checking approaches in tandem with audits of platforms and partnerships with researchers and citizen activists are important in ensuring safe and trustworthy community spaces on social media.

Anjana Susarla, Professor of Information Systems, Michigan State University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Jay Feinman, Rutgers University

The wildfires that have devastated large parts of Los Angeles County have drawn fresh attention to the struggles many Americans face insuring their homes.

Since 2022, seven of the 12 largest insurance companies have stopped issuing new policies to homeowners in California, citing increased risks due to climate change. California isn’t alone: The same thing has happened in other vulnerable states, including Louisiana and Florida. The proportion of Americans without home insurance has risen from 5% to 12% since 2019. Meanwhile, those fortunate enough to have insurance are paying more than ever: Premiums in California, like elsewhere, have increased dramatically over the past five years.

When the private insurance market fails to provide coverage, the government often comes in to fill the gap. For example, the National Flood Insurance Program was established back in the 1960s because almost all private insurers excluded flood coverage. Meanwhile, the California FAIR Plan, which serves more than 450,000 Californians, is a typical state-created insurer of last resort. Such programs, which are available in many states, offer limited coverage to people who can’t get private insurance.

But the sheer scale of need means it’s hard for public programs to stay afloat. It’s not inconceivable that the recent wildfires could exceed the reserves and reinsurance available to the California FAIR plan. Because of the way the plan is set up, that would force other insurers – and ultimately homeowners – to make up the difference.

These are tricky problems, and – speaking as an expert in insurance – I can’t say I have answers. But I do know the right questions to ask. And that’s a crucial first step if you want to find solutions.

What is insurance for, anyway?

One of the most important questions is also the most basic: What are the goals of insurance?

Insurance is a financial product that allows people to share risk – meaning that if a catastrophe strikes any one person, they won’t have to bear the costs alone. But it’s not just about money. Even if most people don’t realize it, every form of insurance embodies values and serves public policy goals. This often requires making social, political and even moral trade-offs.

What is the problem we’re trying to solve?

The first step in solving a problem is to identify it. When it comes to insurance, this isn’t always easy. For example, “Homeowners need insurance coverage that they can’t afford in the private market” might seem like a good description of the problem. But it’s not. This is because some homes in disaster-prone areas are simply too risky to insure.

Imagine a home in a coastal area that floods over and over, for example. If you were an insurer, how much would you charge for that policy? When a house is subject to repeated losses, it makes more economic sense to buy and demolish it instead.

Defining the problem carefully also helps to clarify the values at stake. For example, one value is protecting the investments of current homeowners – particularly, say, long-time, elderly residents. But another value is pricing risk correctly, so people don’t move into dangerous developments.

Put more broadly, one value is recognizing society’s collective responsibility toward people who suffer financial distress, and another is promoting fair and efficient use of social resources. These values can be in conflict.

What does the government have to do with insurance?

Back in 1881, in his classic lectures on The Common Law, Supreme Court Justice Oliver Wendell Holmes Jr. said:

The state might conceivably make itself a mutual insurance company against accidents and distribute the burden of its citizens’ mishaps among all its members. There might be a pension for paralytics, and state aid for those who suffered in person or estate from tempest or wild beasts.

Holmes’ own position was clear: “The state does none of these things,” he wrote – and it should not. This strain of individualism has remained strong in U.S. politics: Individual liberty, personal responsibility and economic opportunity are the foundations of American life, individualists say, so each person should win or lose on their own.

Under this approach, the private insurance market bases its pooling, risk classification and pricing mostly on how much risk each policyholder presents, so that homes in wildfire-prone areas are charged higher premiums. In theory, this is both morally sound and economically efficient, since each policyholder bears the cost of their own risks. But when the private market fails – as happened with flood insurance – the government has a strong incentive to step in.

Today, as an empirical matter, Holmes’ statement couldn’t be more wrong. The state does, in fact, make itself “a mutual insurance company against accidents” and provides a “pension for paralytics,” through Medicaid, Social Security Disability Insurance and other programs. And in California, as elsewhere, the government does provide aid for those who “suffered in estate … from tempest,” through the Federal Emergency Management Agency and other entities.

Since at least the New Deal, there has been broad recognition that some level of collective responsibility is essential; the only questions are where and how much. In the health insurance realm, for example, the Affordable Care Act provides subsidized health insurance for many Americans, and changing Medicare is a political third rail.

Public policy on disaster losses is situated between the two extremes of letting losses lie and having the state assume all of the burdens of those losses. Often policymakers and researchers see insurance or insurance-like plans as solutions – whether provided by a public entity or involving a mixed public-private program. FEMA, for example, operates the National Flood Insurance Program in cooperation with private insurers and also gives direct grants for mitigation of flood damage.

What should a public insurance solution look like?

Sometimes one question leads to another, and that’s the case here. In my research, I’ve identified more than a dozen questions that policymakers must answer in order to design an effective public solution to disaster insurance. Three questions are most important:

• What are the goals of the insurance?

• Who is being insured?

• How are policyholders and their risks classified?

Let’s start with the first question: What are the goals of the insurance? As I mentioned earlier, any form of insurance faces trade-offs and limits.

When an insurance solution has been adopted rather than some other form of intervention, a primary goal is to compensate the policyholder for a loss. But that’s not the only goal. For example, insurance often aims to reduce losses in addition to paying if they occur. Insurers have many ways to shape behavior, such as charging lower premiums for homeowners who keep their property free of flammable brush. Because many of these behaviors affect other people as well, they generate a social benefit. And since insurance has social benefits, how those benefits are distributed – along race, gender, class and other lines – is also important.

That leads to the second key question: Who is being insured?

Insurance involves transferring risk from an individual to a larger group of people who can share the risk. Insurance experts call this “risk pooling.” Pools that are too small will struggle because there aren’t enough people to share the burden.

In public solutions to catastrophe problems, getting more people in the pool could be especially useful in expanding coverage. For example, the National Flood Insurance Program brings many homeowners across the country into a pool, but it also excludes some, such as those who suffer damage from wind during a hurricane. In contrast, the proposed INSURE Act, introduced in the last Congress, would effectively put the entire nation in a pool to cover a variety of catastrophic risks, including flood, wildfire, earthquake and others.

Still, just because you’re in the same pool as someone else doesn’t mean you’ll be treated the same – people with the same insurance can be charged different premiums and receive different amounts of coverage.

That leads to the third question: How are policyholders and their risks classified?

If insurers treated everybody exactly the same, they would quickly go out of business. That’s why they analyze huge amounts of information about past losses, current conditions and future predictions, trying to determine the risks posed by each member. This work is done by actuaries and underwriters, but it’s not just a matter of math: Insurers classify policyholders in ways that reflect the goals and values of the insurance, which typically include balancing widespread availability, broad coverage and affordable pricing, and the social benefits the insurance generates.

One view of this process is that more precise risk classification and pricing are good. Because insurance involves risk transfer, the more accurately risks can be calculated and priced, the better the process works.

But there’s a deeper problem, which has to do with values. Sometimes accuracy in underwriting can conflict with larger social goals. With catastrophes in particular, broad coverage may be a top priority, since many people believe the state has a responsibility to protect its people. Moreover, protecting people’s investments in their homes is important, and suddenly raising the premiums of homeowners at high risk would threaten their investments. Disasters also cause communal responses – many unaffected Americans donate to the Red Cross and other nonprofits to support victims – and a strict focus on accuracy in underwriting could undermine that sense of community.

As floods, storms, wildfires and other catastrophes become increasingly common, the availability and affordability of property insurance has become a high-profile political issue. Politics involve choices. Asking better questions will help politicians – and the rest of us – make better choices.

Jay Feinman, Distinguished Professor of Law Emeritus, Rutgers University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Mark Zachary Taylor, Georgia Institute of Technology

In evaluating the outgoing Biden administration, much news has focused on inflation, immigration or Hunter’s laptop. But as an expert on national competitiveness in science and technology, I have a somewhat different emphasis. My research shows that U.S. prosperity and security depend heavily on the country’s ability to produce cutting-edge science and tech.

So, how did the Biden administration perform along these lines?

Advancing pandemic science and tech

President Joe Biden’s immediate challenge after inauguration was to end the COVID-19 pandemic and then shift the economy back to normal operations.

First, he threw the weight of his administration behind vaccine production and distribution. Thanks to President Donald Trump’s Operation Warp Speed, inoculations had begun mid-December 2020. But there had been no national rollout, and no plans existed for one. When Biden took office, only about 5% of Americans had been vaccinated.

The Biden administration collaborated with private retail chains to build up cold storage and distribution capacity. To ensure adequate vaccine supply, Biden worked to support the major pharmaceutical manufacturers. And throughout, Biden conducted a public relations campaign to inform, educate and motivate Americans to get vaccinated.

Within the first 10 weeks of Biden’s presidency, one-third of the U.S. population had received at least one dose, half by the end of May, and over 70% by year’s end. And as Americans got vaccinated, travel bans were lifted, schools came back into session, and business gradually returned to normal.

A later study found that Biden’s vaccination program prevented more than 3.2 million American deaths and 18.5 million hospitalizations, and saved US$1.15 trillion in medical costs and lost economic output.

In the wake of the economic distress caused by the COVID-19 pandemic, Biden signed two bills with direct and widespread impacts on science and technology. Previous administrations had promised infrastructure investments, but Biden delivered. The Infrastructure Investment and Jobs Act, passed with bipartisan support during late 2021, provided $1.2 trillion for infrastructure of all types.

Rather than just rebuilding, the act prioritized technological upgrades: clean water, clean energy, rural high-speed internet, modernization of public transit and airports, and electric grid reliability.

In August 2022, Biden signed the Inflation Reduction Act, totaling $739 billion in tax credits and direct expenditures. This was the largest climate change legislation in U.S. history. It implemented a vast panoply of subsidies and incentives to develop and distribute the science and tech necessary for clean and renewable energy, environmental conservation and to address climate change.

Science and tech marquees and sleepers

Some Biden administration science and technology achievements have been fairly obvious. For example, Biden successfully pushed for increased federal research and development funding. Federal R&D dollars jumped by 25% from 2021 to 2024. Recipients included the National Science Foundation, Department of Energy, NASA and the Department of Defense. In addition, Biden oversaw investment in emerging technologies, such as AI, and their responsible governance.

Biden also retained or raised Trump’s tariffs and continued his predecessor’s skepticism of new free-trade agreements, thereby cementing a protectionist turn in American trade policy. Biden’s addition was to add protectionist industrial policy – subsidies for domestic manufacturing and innovation, as well as “buy-American” mandates.

Other accomplishments have been more under the radar. For example, within the National Science Foundation, Biden created a Directorate for Technology, Innovation and Partnerships to improve U.S. economic competitiveness. Its tasks are to speed the development of breakthrough technologies, to accelerate their transition into the marketplace, and to reskill and upskill American workers into high-quality jobs with better wages.

Biden implemented policies aimed at strengthening and improving federal scientific integrity to help citizens feel they can trust federally funded science and its use. He also advanced new measures to improve research security, aimed at keeping federally funded research from being improperly obtained by foreign entities.

The CHIPS & Science Act

The jewel in the crown of Biden’s science and tech agenda was the bipartisan Creating Helpful Incentives to Produce Semiconductors (CHIPS) and Science Act, meant to strengthen U.S. manufacturing capabilities in advanced semiconductor chips. It has awarded about $40 billion to American chip producers, prompting an additional $450 billion in private investment in over 90 new manufacturing projects across 28 states.

Directed at everything from advanced packaging to memory chips, the CHIPS Act’s subsidies have reduced the private costs of domestic semiconductor production. CHIPS also pushes for these new manufacturing jobs to go to American workers at good pay. Whereas the U.S. manufactured few of the most advanced chips just two years ago, the industry expects the United States to possess 28% of global capacity by 2032.

Less well known are the “science” parts of the CHIPS Act. For example, it invested half a billion dollars in dozens of regional innovation and technology hubs across the country. These hubs focus on a broad range of strategic sectors, including critical materials, sustainable polymers, precision medicine and medical devices. Over 30 tech hubs have already been designated, such as the Elevate Quantum Tech Hub in Denver and the Wisconsin Biohealth Tech Hub.

The CHIPS Act also aims to broaden participation in science. It does so by improving the tracking and funding of research and STEM education to hitherto underrepresented Americans – by district, occupation, ethnicity, gender, institution and socioeconomic background. It also attempts to extend the impact of federally funded research to tackle global challenges, such as supply chain disruptions, resource waste and energy security.

Missed opportunities and future possibilities

Despite these achievements, the Biden administration has faced criticism on the science and tech front. Some critics allege that U.S. research security is still not properly defending American science and technology against theft or counterfeit by rivals.

Others insist that federal R&D spending remains too low. In particular, they call for more investment in U.S. research infrastructure – such as up-to-date laboratories and data systems – and emerging technologies.

The administration’s government-centered approach to AI has also drawn criticism as stifling and wrong-headed.

Personally, I am agnostic on these issues, but they are legitimate concerns. In my opinion, science and technology investments take considerable time to pan out, so early judgments of Biden’s success or failure are probably premature.

Nevertheless, the next administration has its work cut out for it. International cooperation will likely be key. The most vexing global problems require science and technology advances that are beyond the ability of any single country. The challenge is for the United States to collaborate in ways that complement American competitiveness.

National priorities will likely include the development of productive and ethical AI that helps the U.S. to be more competitive, as well as a new quantum computing industry. Neuroscience and “healthspan” research also hold considerable promise for improving U.S. competitiveness while transforming Americans’ life satisfaction.

Keeping the whole American science and technology enterprise rigorous will require two elements from the federal government: more resources and a competitive environment. American greatness will depend on President-elect Trump’s ability to deliver them.

Mark Zachary Taylor, Associate Professor of Public Policy, Georgia Institute of Technology

This article is republished from The Conversation under a Creative Commons license. Read the original article.