福利小视频

As the 2024 presidential election looms, a new threat to American democracy is emerging from an unlikely source: artificial intelligence (AI). Experts at Colorado State University 福利小视频 (CSU 福利小视频) are raising alarm bells about the growing use of computer systems that emulate aspects of human cognition to spread misinformation, with a particular focus on Hispanic communities. This trend, already observed in countries like Venezuela, could have far-reaching implications for the integrity of the upcoming election.

Roberto Mejias, associate professor of computer information systems at CSU 福利小视频, paints a sobering picture. "AI has a more sophisticated approach to affect not only Latinos, but just a wide range of American voters," he warns. "AI has been around for 30, 40 years but more capable now because of machine learning—where computers are trained to make humanlike decisions—making it far easier to manipulate information."

When polls open on November 5, it will mark the first presidential election since the rise of generative AI. Voters may find themselves exposed to a deluge of AI-created audio, images, and videos, produced by software like Midjourney, DALL-E, and ChatGPT. The sophistication of these digital propaganda tools makes it increasingly difficult for the average viewer to distinguish fact from fiction.

The Latino community, in particular, finds itself at the epicenter of this digital storm. An estimated , up from 32.3 million in 2020, according to the Pew Research Center. As the largest minority group in the country, Latinos will play a crucial role in battleground states like Florida, Arizona, Nevada, and Georgia.

What makes the Latino community especially vulnerable to AI-driven misinformation is their significant online presence. "Latinos use the internet for everything," explains Jim Quintana from CSU 福利小视频's Cyber Security Lab. "There's a sizable portion of the community that relies on YouTube or Facebook for their news."

This reliance on social media for information creates a perfect breeding ground for AI-generated misinformation. A study from New York University's Center for Social Media and Politics found that Latinos—both bilingual and predominantly Spanish-speaking—use YouTube for political news far more than non-Hispanic whites. Moreover, they tend to consume political news on YouTube channels not associated with major media outlets, where information is less likely to be verified.

The problem extends beyond traditional social media platforms. Encrypted messaging apps like WhatsApp and Telegram, popular among Latino communities, present another challenge. "These encrypted tools don't allow for monitoring and removal of content," Quintana notes. "People share whatever they want, and it also allows for a lot of transnational sharing of information."

Mejias identifies a two-pronged approach that makes Latino communities particularly vulnerable to misinformation. First, older generations consuming traditional Spanish-language media—often lacking rigorous fact-checking—influence their children's political views. Second, younger Latinos primarily get news from social media, a breeding ground for AI-driven misinformation campaigns.

This intergenerational dynamic, coupled with the unfiltered flow of information across borders through family networks, creates a perfect storm for the spread of false narratives. With only 30% of Spanish-language media employing strict fact-checking practices, bad actors can effectively target Latinos through both traditional and new media channels, exploiting familial ties as inadvertent conduits for misinformation.

Mejias stresses that Latinos are not a monolith voting group. But the perpetrators of mis- and disinformation tailor their campaigns to exploit specific cultural concerns. In 2020, a Trump campaign falsely claimed the Venezuelan government supported Joe Biden. According to the independent news nonprofit, ProPublica, the video was watched over , nine days before 2020 election day.

Ads dealing with inflation target people with Argentinian and Venezuelan heritage, playing on their experiences with economic instability. Messages about abortion and reproductive rights resonate with the predominantly Catholic Latin American population. Hints of electoral fraud strike a chord with South Americans, because countries like Honduras, Nicaragua, and Ecuador actually experience this.

Some narratives, however, transcend sub-communities to target the wider Spanish-speaking electorate. , who leads , points to the persistent claims of electoral fraud in 2020 and the so-called "illegal migrant invasion" as prominent examples. "We've often seen immigration messages, including some using false or misleading data to draw a link between increased immigration and criminality," Rojas said in a .

A  that traced the browsing habits of more than 3,300 people found Latinos — both those who are bilingual and those who predominantly speak Spanish — use YouTube for political news far more than non-Hispanic whites. The study also found Latinos consume a large amount of political news on YouTube channels that are not associated with major media outlets and where information is less likely to be verified.

The potential for AI to influence elections has already been demonstrated. During New Hampshire's primary, an AI-generated robocall mimicking President Biden's voice attempted to discourage voter participation, highlighting the technology's potential for voter suppression.

Recent incidents highlight the global reach and potential dangers of AI-driven disinformation campaigns. In Venezuela, a fictitious news channel featured AI-generated news anchor presenting distorted views of the country’s economic situation. Venezuela showcases the sophistication of these AI-generated disinformation campaigns. Videos featuring AI-created presenters "Noah" and "Daren" have gone viral on social media platforms, presenting a distorted view of the country’s economic situation. These videos, which have garnered hundreds of thousands of views, were even broadcast on state-owned television networks. 

As the election approaches, experts emphasize the importance of media literacy and critical thinking. "Voters need to be more vigilant than ever," Mejias advises. "Verifying sources, using fact-checking resources, and consuming diverse news sources are crucial steps in combating AI-driven manipulation."

CSU 福利小视频 is taking proactive steps to address these challenges, developing programs to educate students and the wider community about AI-generated misinformation and its potential impact on democratic processes. Additionally,

While AI presents significant challenges, it's important to note that it's not creating entirely new problems. "Disinformation and election interference existed long before AI," Quintana reminds us. "Our focus should be on using technology responsibly while bolstering our defenses against its misuse."

As the 2024 election approaches, the role of AI in shaping political discourse remains a critical concern. The Latino community, with its significant online presence and unique vulnerabilities, finds itself at the forefront of this digital battlefield. Their engagement and critical thinking in the face of AI-driven misinformation may well determine the integrity of our democratic process in this new technological era.