HON 102 Padlet: Misinformation & Democracies by Abbas Dadawalla

Psychreg Journal of Psychology(Global)

aagray7 — 2025-02-19 19:16:53 UTC

This source discusses the ability of AI to mimic the voices and appearances of people to deceive audiences. On social media where misinformation is rampant, it prioritizes user engagement over truth and since misinformation is generally more interesting than true information, it gets more attention from people, which starts a cycle of misinformation. This is helpful in understanding our problem because the sophistication of AI-generated misinformation is so complex that it is hard to detect. It also talks about how a collaboration between government, tech companies, and civil society organizations would be helpful to enforce a standard to hold platforms accountable for the content they host, which could be part of a possible solution.

MIT Technology Review(National/Global)

aagray7 — 2025-02-19 19:17:36 UTC

It talks about the accessibility and affordability of AI leading to its widespread use. It allows governments to subtly censor various online platforms. It talks about Venezuela, which spreads pro-government messages from a fake news anchor from a fake English news channel. Freedom House researchers found that 22 countries created laws that required or encouraged online platforms to censor unfavorable speech. It helps show examples of how the problem is used in non-democratic countries, such as in China where chatbots are programmed to not answer questions about Tiananmen Square.

Peer-reviewed journal article(The Problem)

aagray7 — 2025-02-19 19:18:21 UTC

AI has been used to counter misinformation and AI itself by identifying bot accounts and having an automated fact checker, but it has its limitations. It struggles with sarcasm and culturally specific nuances, as well as overblocking legitimate content. It also discusses the algorithm bias built into the system itself which can do the opposite of what it was supposed to do. This shows how we can develop a solution for our problem and what has been tried to solve it already.

Think Tank Report(National/Global)

aagray7 — 2025-02-19 19:18:50 UTC

This report talks about the fact that no policy or intervention has been universally effective in combating misinformation. Solutions to this problem are dependent on the will of the political emplacement, which in countries where misinformation is a political tactic will not regulate it. This report advocates for solutions like supporting local journalism and fact-checking. This report helps with understanding the problem and possible solutions by talking about the obstacles in the path of combating misinformation, especially in countries that actively use it for political reasons. Possible solutions can be investing in psychological research on factors that make misinformation spread.

Think Tank Report(Global)

aagray7 — 2025-02-19 19:19:18 UTC

This report talks about the types of AI generators such as generative models that create fake images and videos, automated bots, and algorithms that can create misinformation and give it traction immediately. It talks about the technological imbalance, where we have incredibly complex AI generators of various types, but the technology to detect it within real information has lagged behind, which explains the problem we are having now. A possible solution the report goes over is educating the people on how AI works and how to identify it, which can not only lessen the effect of AI but also generally increase critical thinking skills. It also helps to understand the problems democratic governments have fighting it as opposed to authoritarian governments' heavy use of it to reinforce their own position.

Final Report from the U.N. Advisory Board on AI & Humanity (Global)

dadawalla_abbas — 2025-02-25 05:00:18 UTC

1) This “Governing AI for Humanity: Final Report”, formed by the U.N. high-level advisory board on AI, discusses and presents a need for a “global governance framework” to address both opportunities and challenges presented by AI. The report highlights gaps in current AI governance, particularly with coordination (AI governance efforts lacking cohesion), representation (global south in particular is excluded from AI governance and initiatives), and implementation (commitments often fail to translate into actionable outcomes). Seven recommendations are proposed regarding policy dialogue, scientific inquiry, capacity development, and establishing a U.N. AI office.

2)This source is helpful because it helps readers understand what the current risk AI poses: 1) AI is utilized to create disinformation campaigns, 2) AI systems are not easy for the public to understand making it difficult to identify when it is used maliciously, and 3) biased algorithms targeting marginalized groups have the potential to violate human rights. These concerns also help readers to understand the recommendations proposed which include: establishing a scientific panel to assess AI risks + opportunity, creation of a global fund for addressing AI disparities, as well as development of a global AI capacity network which will be necessary for providing skills, infrastructure, and equitable distribution of AI.

Journal Article on a Blockchain Authentication System for Detecting Digital Misinformation (Global, but can be implemented at different scales)

dadawalla_abbas — 2025-02-25 06:10:53 UTC

1) This article by Islam et al. (2024) discusses the threats posed by AI to democracy and global elections. Through the creation of deepfakes and misinformation, AI technology may exacerbate political polarization (prioritizing sensational content and creating “echo chambers”), manipulate public perceptions, and destroy voters’ trust in the political process. A “Blockchain-based Deepfake Authenticity Verification Framework” (B-DAVF) is proposed; it is a technology which uses the blockchain system to authenticate digital content with the goal of utilizing it to improve the integrity of electoral processes (voter targeting & messaging; data analysis). The authors also provide a rather comprehensive countermeasure list which includes educational initiatives (e.g. media literacy programs on recognizing AI-generated content) and technology regulations (e.g. legislation on role of AI in political campaigns; mandating disclosure when AI is used in political advertising, etc.) to mitigate AI risks in political scenarios.

2) This source supplies readers with an understanding of the structure of a Blockchain verification system, currently considered countermeasures, a history of AI misuses, as well as example cases studies depicting the dangers that accompany AI misuse.

Report from a Non-Profit Public Policy Org. on AI & African Democracies (National & Local)

dadawalla_abbas — 2025-02-25 07:30:54 UTC

1)This report explores AI disinformation threats on African democracies during national elections. The report writers point out that there is growing trend of “foreign-sponsored disinformation” and viable countermeasures include increasing AI literacy and distributing greater responsibility to local fact-checking organizations. They call for a collaboration of many stakeholders to build a more resilient democratic framework against AI-generated misinformation. Something concerning the authors mention is that AI-generated misinformation has already contributed to coups in nations such as Nigeria, Gabon, and Burkina Faso.

2) This source is helpful in bringing attention to what specific issues that nations in the global South and underdeveloped countries face in regards to AI misuse. First, media consumers in many African countries often encounter misinformation, not only related to politics, but public health; this has been linked to the HIV/AIDS crisis. Secondly, representation (as the U.N. report mentions) has not advanced on its front given that most AI programs operate in Western languages, easing the spread of false information in African contexts. Lastly, it highlights national and local resources that African citizens must engage with or rely on and certain initiatives - these include sites like AfricaCheck or Dubawa and working to increase local media literacy.

Slovak Case - A Commentary (National)

dadawalla_abbas — 2025-02-25 08:27:21 UTC

1) This commentary analyzes the implications of deepfakes in the 2023 Slovak parliamentary elections, particularly the victory of pro-Russian candidate Robert Fico, who many claim was influenced by a deepfake depicting electoral fraud that went viral. The authors say that this raises concerns over “technology-induced misinformation” however there needs to be research into how deepfakes interact with a historical context in nations that already have long standing political distrust between citizens and government.

2) This source is unique in that it provides insight into vulnerability to AI/pro-Russian propaganda/conspiracy theories in democracies with a history of disinformation campaigns. The authors point to politicians as more than just consumers but also major disseminators of misinformation who may intentionally utilize AI to exacerbate the “liar’s dividend” where true and false information are conflated in public discourse. The authors also encourage research into encrypted messaging, a tool which initially intended to secure information, might be used as a misinformation vehicle.

Journal Article on AI and China's Authoritarian Government (National)

dadawalla_abbas — 2025-02-25 09:17:53 UTC

1) This article examines the intersection of AI and China’s authoritarian government and highlights how the CCP leverages AI for administrative efficiency as well as maintaining legitimacy and state control. The author, Jinghan Zeng, explores AI’ capacity to bolster authoritarian practices. He states that AI offers opportunity for economic advancement in China but that comes at the risk of increasing state surveillance and repression. In particular, Zeng mentions that the CCP aims to develop an AI economy worth > 1 trillion yuan by 2030, hoping to use AI to foster public support.

2) his source discusses how the potential for AI to consolidate authoritarian practices raises implications about global civil liberties. This author would argue that AI advancement empowers state control and repression which, of course, is a problem for democratic norms. By 2030, this would position China as a major player in international politics and economics, and a potential antagonist against liberal democracies.

How AI Threatens Democracy (Global)

2025-02-25 17:53:43 UTC

1) This is a peer-reviewed publication by the Journal of Democracy. It explains how AI undermines democracies by undermining representation, accountability, and trust, which are the three pillars of democracy. According to the source, AI can inhibit political representation by allowing bad actors to generate many unique advocacy letters and sending them to representatives. With AI, anyone can write propaganda. Also, with so much information in the media market we may see the public wholly distrusting the democratic process even more than before.

2) Here, it is suggested to create AI detection tools, set boundaries so that AI is not used in unethical ways, and promote digital literacy. There have been pushes to implement these changes, but more work still needs to be done. For our project, this article demonstrates why AI and misinformation in general is a great problem that we must work to solve.

Artificial Intelligence: The Biggest Threat to Democracy Today? (Global)

2025-02-25 18:11:37 UTC

1) This source discusses how AI poses a threat to democracy. For one, it discusses how many generative AI models have data sets manually coded in by programmers. With these come bias. Since AI can be responsible for the production of misinformation, such technologies harm one's ability to make political decisions independent of outside influence. As a result, political trust is likely to decrease. The authors forecast the emergence of a radical populist right, the bolstering of authoritarian regimes, and algocracy (the rule of algorithms).

2) Understanding the direction that politics is headed in the age of generative AI is useful for identifying the problems we will run into and planning their solutions ahead of time.

Nie, M. (2024). Artificial Intelligence: The Biggest Threat to Democracy Today?. Proceedings of the AAAI Symposium Series, 3(1), 376-379. https://doi.org/10.1609/aaaiss.v3i1.31239

How AI and Democracy Can Fix Each Other

2025-02-25 18:28:31 UTC

1) This TED Talk demonstrates how we can overcome the challenges presented by the use of AI in democracies. She explained how she trained an AI using a constitution written by other people. As much as there will be challenges, there are opportunities to strengthen democracy with new technologies and the information output they allow for. However, this talk is overly optimistic and does not give concrete examples of AI being implemented to directly aid democracy. It does not account for the fact that AI companies profit by keeping their models undemocratic.

2) The speaker's example of training an AI model to be democratic is a glimpse into how this new technology can be harnessed for good. It is also a call for more action to see how we can implement generative AI into the functions of democracy, ethically.

An Agenda to Strengthen U.S. Democracy in the Age of AI (National/Global)

2025-02-25 18:44:16 UTC

1) This article discusses how AI was used in various ways to spread misinformation during the 2024 election, although the projected worst-case scenarios did not play out. The authors outline an agenda to minimize the dangers of AI and maximize its democratic potential. They suggest ensuring that governments have the personnel sufficient to create and maintain ethical use standards. Also, the authors call for transparency in AIs use on social media and for self-disclosure of politicians when their communications include AI content.

2) The suggestions given here are very helpful and relevant to the United States. They give action items to follow in ensuring that our elected officials use AI ethically.

Artificial Intelligence and Elections: Protecting Democracy in the Digital Era

2025-02-25 20:51:45 UTC

1) California's Legislature met to discuss how to combat the negative impacts of AI. They said that while AI will allow for voter education, outreach efforts, and election material delivery, we will need to prepare for its negative potential. The members explained that it is up to each county in California to run its election system. With decentralization, it it harder to alter election outcomes on a large scale but easier for each county to become overwhelmed by the spread of misinformation. The article explains the importance of security, impersonation checks, and communication plans.

2) This article is relevant to us as residents of California and shows some of the work being done right now by our legislators to address the negative consequences of AI. The ideas given here could be generalized to other states or the United States as a whole.

Examples of Russian disinformation and the facts (Russia-Ukraine)

aransaini17 — 2025-02-25 21:21:59 UTC

This article was written by the German government and talks about how Russia uses misinformation in regards to their Ukrainian invasion. It does so by providing specific examples of the information they released and refuting them. This article is helpful in understanding the challenge because it provides a specific example of how a government can use misinformation to disrupt democracy, and while this article may not contain all of the details it leaves a place for us to start and research.

Federal Ministry of the Interior and Community. (2023, November 2). Examples of Russian disinformation and the facts. Federal Ministry of the Interior and Community.

Can Democracy Survive the Disruptive Power of AI? (Global)

aransaini17 — 2025-02-25 21:25:53 UTC

This article was written by the think tank Carnegie Endowment for International Peace and discusses how artificial intelligence can be a threat to democracy. It provides information about how deepfakes can be used to gain a political advantage and how they make the political environment for voters even more confusing than it already is. They also discuss how AI can be a threat in many unforeseeable ways because it can advance faster than the government can regulate it if it is weaponized. It can essentially create this misleading content on its own which can outpace real information so it increases its impact with the sheer amount of content. The article also suggests certain regulations that the government can make to prevent it from getting out of hand, which includes watermarking AI content and preventing its use for certain situations. Overall this article is helpful as it highlights the threat that AI poses as well as propose certain countermeasures policymakers can take.

Csernatoni, R. (2024, December 18). Can democracy survive the disruptive power of ai?. Carnegie Endowment for International Peace. https://carnegieendowment.org/research/2024/12/can-democracy-survive-the-disruptive-power-of-ai?lang=en

Overcoming digital threats to democracy (Global)

aransaini17 — 2025-02-25 21:37:32 UTC

This article was written by the Lowy Institute which is an Australian think tank. This article both discusses the potential threat to democracy created by the internet, as well as proposing a course of action that needs to be taken. This article talks about how the power of the internet is held by a few countries and the users are no longer citizens but more like commodities. This article suggests that countries need to use more deliberative democracy, which involves publicly deliberating ideas to ensure that they are actually valid rather than the ideas that have been most efficiently spread through the internet. Not only does it ensure the validity of ideas but it also increases citizen participation in politics. The article also proposes increasing this deliberative democracy using the internet which would help reverse the current situation. This article is beneficial, because not only does it establish one potential threat to democracy that the internet has created but it also offers a potential solution.

Many Tech Experts Say Digital Disruption Will Hurt Democracy (Global)

aransaini17 — 2025-02-25 21:50:14 UTC

This article was written by the Pew Research Center which is a think tank. They discuss survey results as well as the opinions of several experts to establish how many people consider the potential threat that this advanced technology may bring to democracy. While the survey may not have been completely accurate they found that 49% of participants believe AI will be a threat in the future. The primary reason why I believe that this article will be helpful is because it contains public opinion information about what they think about AI and it as a threat to democracy. This is important because if people are concerned they are more likely to support action against the issue. Another reason why this article is beneficial is because it has the opinions of several experts and how they think AI will affect democracy, which is a prediction but it may be helpful to look back on while writing the paper.

AI And The Shadow Over Democracy: The Rising Threat To Global Elections (Global)

aransaini17 — 2025-02-25 21:56:26 UTC

This article was written by Forbes which is an American media company. While it is highly recognized the information may need to be checked using other sources, it provides some helpful background information. The article discussed how deepfakes can be detrimental to democracy as mentioned by several other articles, however they also mentioned how social media can create an illusion of consensus which may encourage voters to vote a certain way by excluding the opposition. Social media also polarizes the country because it constantly shows people opinions that they already support and does not show the opposition, because it may decrease engagement, however this simply makes them stronger supporters of their cause but increases their disdain for the other opinions as well. I believe that this article will be helpful, because it covers some of the essential issues that are caused both by artificial intelligence and social media as well as proposing some potential solutions, such as improving media awareness in order to help prevent the damage.