What if the truth were in jeopardy? The dilemma of fact-checking

Meta has recently revised its fact-checking strategy. These are measures closely linked to politics, which threaten the quality of online information. We tackled the topic with Eleonora Benecchi, Lecturer-Researcher and Bachelor Director at the Faculty of Communication, Culture and Society of Università della Svizzera italiana (USI). She explained the essential characteristics of quality fact-checking, its importance, and the risks associated with the current decisions made by various platforms.

Below is the interview conducted by the USI Institutional Communication Service.

Eleonora Benecchi, how does fact-checking take place on major social platforms?

"Fact-checking on major social media platforms is typically conducted in collaboration with independent verification organisations. For instance, Meta has established a global fact-checking programme that includes approximately 90 partners across 130 countries, many of which are affiliated with the International Fact-Checking Network (IFCN). Fact-checkers review viral content reported by users or identified by algorithms. If information is found to be false, the platform may label it, reduce its visibility, or restrict its dissemination."

Considering the size of the web and the amount of platforms and content published daily, is it still possible to carry out quality fact-checking? If so, why is it important that it is done?

"Despite the vast amount of content generated daily, it is still possible to ensure quality fact-checking through a hybrid approach combining automation with human review. However, the increasing reliance on automated moderation technologies has revealed the risks associated with inadequate control. For instance, a recent error by Meta on Instagram Reels exposed millions of users to violent and unsafe content due to a malfunction in the recommendation algorithm. This incident highlights how automated systems can fail to accurately identify content that should be filtered. The importance of fact-checking goes beyond correcting fake news: it plays a crucial role in safeguarding democratic dialogue and combating information manipulation. This manipulation has already had significant repercussions during the pandemic and elections. Therefore, it is vital to balance automated processes and human oversight to ensure the reliability of information and to combat global disinformation effectively."

How should quality fact-checking be done to ensure political and ideological neutrality?

"Quality fact-checking must respect criteria of transparency, independence and rigorous methodologies. Verification organisations must disclose their sources, follow shared standards such as those set by the IFCN and ensure a fair approach, analysing information from various political orientations without bias. Additionally, it is essential to maintain a clear distinction between fact-checking and platform moderation decisions to prevent accusations of censorship or political manipulation."

How is it possible to reconcile fact-checking and freedom of expression? How would you respond to those who compare fact-checking to censorship?

"Fact-checking does not restrict freedom of expression; it helps counter disinformation by providing additional context to assess the truthfulness of claims. Unlike censorship, which blocks the spread of information, fact-checking enriches the discourse without removing content. Through the Digital Services Act (DSA), EU regulation aims to balance these aspects by requiring platforms to collaborate with fact-checkers without arbitrarily removing content."

Recently, Mark Zuckerberg announced the end of Meta's fact-checking programme, which was launched in 2016. Does this concern you?

"Yes, the end of Meta's fact-checking programme is a worrying decision, as the platform is the main backer of information verification initiatives globally, having invested approximately 100 million dollars since 2016. Currently, about one-third of Meta's 90 partner organisations rely solely on its funding and face the possibility of closure. Many others may have to reduce their activities, leading to staff layoffs and diminished efforts to combat disinformation. In Europe, the Digital Services Act (DSA) mandates that platforms must collaborate with fact-checkers to lessen the risks associated with disinformation and ensure a "fair financial contribution" to support their efforts. However, the application of these regulations is still uncertain. The main platforms, including Meta, have shown little compliance with the commitments in the Code of Practice on Disinformation. Meta's decision could greatly diminish the ability to combat disinformation, particularly in countries where fact-checking relies almost entirely on funding from digital platforms. As a result, Meta's move not only limits the role of information verification but also raises important questions about the responsibility of these platforms to maintain a healthy and transparent information ecosystem."

Thinking about the case of another social network, X is criticised because it leaves a certain space for sensitive speeches, including hate speech that proliferated during the recent election campaign in the United States. How does this kind of freedom relate to fact-checking, and how does allowing freedom of expression while permitting hatred correlate with the spread of fake news due to ineffective fact-checking?

"Freedom of expression is a fundamental principle, but it cannot be interpreted as a total absence of regulation: without proper moderation and fact-checking, platforms can become breeding grounds for the unchecked spread of disinformation and hate speech. The situation with X (formerly Twitter) under Elon Musk's management is a striking example of this issue. The platform has significantly reduced moderation, dismantled much of its security and trust team, and withdrew from the EU Code of Practice on Disinformation. This voluntary initiative requires platforms to collaborate with fact-checkers and take measures to limit the spread of false news. Multiple studies have shown that there has been a significant rise in the spread of fake news, misleading content, and hate speech. This issue has become more pronounced during election campaigns, as the platform has permitted manipulative content designed to influence voter behaviour and polarise public opinion to circulate freely without restrictions. The European Commission has, in fact, launched a formal investigation into X under the Digital Services Act (DSA), accusing it of not taking adequate measures to counter disinformation and illegal content, especially during critical moments such as elections. X's management shows that deregulating moderation and reducing fact-checkers do not promote greater pluralism; instead, they create a more toxic and manipulable digital environment."

Could the choice not to place limits on publishable content somehow be a political move comparable, although opposed, to that of limiting visible content in one's own country? Why?

"Both strategies significantly impact the quality of information and shape how the public perceives reality, highlighting that the control—or lack thereof—of information is always a political issue. Total deregulation of information, as seen in the case of X, can be viewed as a libertarian approach often exploited for political purposes. However, this can result in an uncontrolled spread of disinformation and hate speech. Conversely, authoritarian regimes employ censorship to manage the public narrative and suppress dissent by drastically limiting the diversity of information available. In both scenarios, the management of information is used to influence public discourse, though with different objectives and methods. Meta's decision to scale back its fact-checking programme in the United States while maintaining it for the moment in the European Union to comply with the DSA regulations highlights how platforms do not adopt a single global policy but adapt to different regulatory contexts and local pressure. This flexibility shows that platforms' content moderation is not neutral but rather influenced by political, economic, and regulatory interests."