Meddling with Truth in the Age of AI and Deepfakes

  • 27/10/2025
  • Category: Commentaries
No Comments
deepfakes

Author: Emira Anjani
Editor: Achmed Faiz Yudha Siregar

Recently, Indonesia’s political climate has intensified due to the sudden increase in salaries and allowances for members of the House of Representatives (DPR), juxtaposed with the nation’s worsening socio-economic conditions. Public outrage was captured in the viral hashtag #bubarkanDPR (“disband the DPR”), which spread across social media. Just a week prior to widespread demonstrations on the street, an AI-generated video carrying the #bubarkanDPR tag circulated online depicting a peaceful march by mass organizations (ormas), allegedly involving hundreds or even thousands of participants. While the video appeared to side with public demands, it was met with criticism as many viewed it as disinformation––an attempt to manipulate public perception and rally supper through fabricated content. In a different light, at the peak of public unrest, another AI-generated deepfake video surfaced featuring the so-called “brave-pink” woman, who rose as a symbolic figure of protest, in which she seemingly mocked and called for President Prabowo to be replaced by Anies––his utmost political opponent in the election. These two cases illustrate the emergence of a new political and social phenomenon in Indonesia. That is, the strategic use of AI-generated disinformation as a political weapon, both by the state against citizens and vice versa, to advance specific narratives and shape public opinion.

The #bubarkanDPR movement was not the first instance of generative AI playing a role in Indonesian politics. The actual onset of AI in political discourse began months leading up to the 2024 general election, during which the three presidential candidates utilized AI in various ways aimed at boosting public engagement.[1] The most prominent example was the AI-generated caricatures featuring the winning pair, now President Prabowo Subianto and Vice Gibran Rakabuming Raka. Fast forward to the present, the presence of AI in the political sphere remains intact, strong, and continues to grow. A large-scale, government-led initiative is now promoting the integration of AI technology and education into public institutions and administrative programs. Notably, the election-time AI caricatures have resurfaced in government promotional campaigns––a move that has drawn significant public criticism. Given the scale and visibility of the government’s AI promotion, the appearance of the two recent AI-generated videos is no longer particularly surprising or unprecedented. However, the emergence of disinformation content that redirects public perception toward specific agenda and implants predetermined narratives reflects a deeper challenge at the intersection of unregulated AI and politics: the rise of political manipulation through reality distortion, the erosion of civic dialogue, and the decline of democratic values. In this context, this article seeks to explore in depth the growing role of AI in shaping the dynamics of political communication within society.

It feels right, what’s the problem?

AI systems have enabled the creation of hyper-realistic content that is often nearly indistinguishable from reality.[2] With simple textual prompts, AI-generated deepfakes can replicate and manipulate everything down to voice and dialogue to facial expressions, content details, and even video resolution. It is no secret that generative AI, particularly in the realm of political messaging, holds the potential to significantly enhance audience targeting, adapt to evolving social dynamics, and streamline the message production process.[3] These are precisely the benefits that political actors in Indonesia often emphasize and advocate for. However, such advantages cannot be viewed in isolation from growing academic concern over the role of generative AI in perpetuating post-truth narratives for political gain.

According to Brennen and Perrault (2023), AI can be exploited in three primary ways to achieve political objectives. First, by generating personalized content tailored to specific audiences; second, by modifying existing narratives to support a particular agenda, often involving the distortion or manipulation of truth; and third, by creating entirely false information designed to promote specific political ideologies and influence public opinion. Not dissimilar to fake news or historical forgery, disinformation openly rejects the principles of objectivity and truth-seeking.[4] It shifts the foundational premise of information discourse from ‘what is right’ to ‘what feels right,’ by altering and engineering content elements to align with a particular agenda.[5]

As a case in point, many social media users interpret “brave pink” counterfeit video as an act of spreading hate instead of a legitimate peaceful demand for constructive reforms. The video is also framed as an indicator that the ongoing demonstration is inorganic and politically motivated to discredit the current government. This is an old tactic to undermine social movements; modernized by the use of AI. Disinformation drives wedges within society, prompting the polarization of the population, distorting public perspectives on the movement and stripping power from unapologetic resistance.[6] In many cases, the diffusion of social movements is accompanied by fear-mongering, which further erodes solidarity and unity. Beyond delegitimizing social movements, the broken public trusts caused by disinformation could evolve into a ‘cry wolf’ effect, diminishing the collective response to future mobilization and ultimately paralyzing participation and movement-building.[7]

In the long run, continuous use of deepfake contents to pit society against itself may deepen existing division within a nation.[8] Especially if contents were to exploit ethnic or political tensions to send a divisive message. This could either catalyze internal conflicts or sabotage reconciliation efforts, which ultimately compromising the social and political fabric of the nation. Furthermore, the true extent for which deepfakes are used has yet to be determined, meaning that it may well be used in judicial settings and propaganda disseminations. Altogether, the bending of truth and production of falsehood could significantly harm public trust on the government, each other, and the information system.[9] In this context, informational and governmental credibility have deteriorated or vanished, leaving a dysfunctional society that thrives on a sporadic emergence of narratives and counternarratives. The public have either lost their trust, a critical pillar of a democratic society, or become the perpetual victim of indeterminate news. These are not temporary consequences that fade over time, but rather deep institutional failures, rooted in ignorance and politically driven interests.

Transformation of Political Realities as an Unfolding Event

The exploitation of AI for political purposes is not limited to political figures or state actors, but is also carried out by ordinary citizens seeking to create momentum for specific political causes. Given that deepfake technology can be operated without any advanced technical skills and is (in most cases) freely available online, it is virtually accessible to anyone with a basic internet connection. Even long before the emergence of the #bubarkanDPR march videos, fabricated content created as a form of grassroots resistance or expression of dissatisfaction had already been circulating widely on social media. The majority of such content was not necessarily intended to push forward an alternative narrative, but rather to troll existing content, often in a humorous and satirical manner. Many of these instances may appear relatively harmless, and some even perceived as comedic. However, what deserves attention here is the high accessibility of deepfake technology that grants the public the same ability as the government to exploit synthetic media for political momentum and influence, especially in (re-)shaping public opinion.

This observation is not intended to play the devil’s advocate or to equate all actors as both victims and perpetrators of the same issue. Rather, it aims to underscore the extent to which AI technologies have fundamentally transformed for political realities are constructed, contested, and experienced in real time. Through AI, it becomes possible to manufacture alternate realities that compete with, rather than contradict, established facts, turning the old saying ‘there are two sides to every story’ into a literal, programmable truth. These realities can be intentionally crafted to communicate a specific message or serve a particular political agenda. More critically, there is no definitive way to predict where this phenomenon will lead. Will AI, as many proponents suggest, ultimately bring more good than harm? Or will it further exacerbate political ambivalence, disinformation, and injustice?

Various experts have expressed growing concern that the continued advancement of deepfake technology could lead to an information apocalypse or “infocalypse”, in which public trust in formal sources of information is drastically eroded, pushing people to rely solely on their inner social circle.[10] This is arguably not surprising given the increasingly sporadic and untraceable spread of fabricated content across multiple media platforms. Another significant concern among scholars involves segments of the population who, due to the digital divide, lack of awareness, or other socioeconomic factors, remain unaware of the existence and risks of deepfakes. These individuals are especially vulnerable, as they are more likely to consume and believe disinformation without scrutiny.[11] Even for those outside this vulnerable demographic, the hyperrealism of deepfakes makes it increasingly difficult to distinguish fabricated media from authentic content. While advancement in AI also produced deepfake detection tools, this creates a paradoxical scenario in which the same technology that generates the problem is being used to solve it.[12]

In conclusion, preventing a nationwide erosion of public trust, institutional legitimacy, and the reliability of information requires grounded and coordinated efforts. Identifying, confronting, and prosecuting disinformation-related offenses cannot rest solely on law enforcement mechanisms.[13] As AI technologies continue to evolve, so too will the complexity and scale of the challenges they present. A number of academic and policy-oriented works, both in Indonesia-specific contexts (Fatimah et al., 2023; Sakila et al., 2025; Judijanto et al., 2025) and from a global perspective (Olakoyenikan & Olanipekun, 2023; Busch & Ware, 2023; Romanishyn et al., 2025; Kim, 2025), have proposed mitigation efforts to curb the proliferation of disinformation. Existing regulations, such as in France and Germany, typically scrutinize the contents itself in combating disinformation, leading to approaches that mainly focus on content moderation.[14] Although content moderation is certainly valuable as a direct action to clean media from disinformation, itself as an isolated approach may be ineffective and unsustainable. Jaursch (2021) suggests a more foundational plan by incorporating a multidimensional and multistakeholder approach, such as auditing media companies and reinforcing international cooperation. Critically, though, any future regulatory efforts must not be one-sided. It must not become a political tool for governments or authorities to repress dissent or silence critique under the guise of combating disinformation. The preservation of democracy and justice requires that authorities uphold the principles they promote, rather than exploit strategic loopholes to maintain power and manipulate public perception.

Bibliography

Anna Sakila, Maulidiah Yasmin, Nidaurrohmah, and Royhan Ammar Sulthon. “Pancasila-Based AI Ethics: Preventing Digital Manipulation and Deepfakes in Society.” Khatulistiwa Journal of Artificial Intelligence 2, no. 2 (2025). https://journal.literasikhatulistiwa.org/index.php/kjarti/article/view/254.

Arifin, Moh. Samsul. “ARTIFICIAL INTELEGENCE (AI) DAN POLITIK: ANALISIS TERHADAP PERAN DAN SIGNIFIKANSI AI DALAM KOMUNIKASI POLITIK 2024 DI INDONESIA.” Ilmu Sosial Dan Humaniora 6, no. 1 (2025). http://dx.doi.org/10.28944/hudanlinnaas.v6i1.1860.

Battista, Daniele. “Communication and Politics in the Age of Artificial Intelligence: An Overview of Deepfakes and Their Implications.” Society Register 8, no. 2 (June 30, 2024): 7–24. https://doi.org/10.14746/sr.2024.8.2.01.

Bennett, W Lance, and Steven Livingston. “The Disinformation Order: Disruptive Communication and the Decline of Democratic Institutions.” European Journal of Communication 33, no. 2 (April 2018): 122–39. https://doi.org/10.1177/0267323118760317.

Brennen, Scott Babwah, and Matt Perault. “The New Political Ad Machine: Policy Frameworks for Political Ads in an Age of AI ,” 2023. https://techpolicy.unc.edu/wp-content/uploads/2023/11/GAI-and-political-ads.pdf..

Busch, Ella, and Jacob Ware. “The Weaponisation of Deepfakes Digital Deception by the Far-Right,” 2023. https://icct.nl/sites/default/files/2023-12/The%20Weaponisation%20of%20Deepfakes.pdf.

Chenoweth, Erica. Social Movements in the Age of Fake News with Erica Chenoweth. Interview by Mari Megias, 2020. https://www.hks.harvard.edu/wiener-conference-calls/erica-chenoweth#transcript-1766694.

Crawford, Kate. Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence. Yale University Press, 2021.

Harsin, Jayson. “Regimes of Posttruth, Postpolitics, and Attention Economies.” Communication, Culture & Critique 8, no. 2 (February 24, 2015): 327–33. https://doi.org/10.1111/cccr.12097.

Jaursch, Julian, Théophile Lenoir, Bret Schafer, and Etienne Soula. “Tackling Disinformation : Going beyond Content Moderation.” Institut Montaigne, 2021. https://www.institutmontaigne.org/en/expressions/tackling-disinformation-going-beyond-content-moderation.

Judijanto, Loso, Andrew Shandy Utama, and Heri Setiyawan. “Implementation of Ethical Artificial Intelligence Law to Prevent the Use of AI in Spreading False Information (Deepfake) in Indonesia.” The Easta Journal Law and Human Rights 3, no. 02 (February 28, 2025): 101–9. https://doi.org/10.58812/eslhr.v3i02.470.

Kedar, Hadas Emma. “Fake News in Media Art: Fake News as a Media Art Practice vs. Fake News in Politics.” Postdigital Science and Education 2 (July 13, 2019). https://doi.org/10.1007/s42438-019-00053-y.

Kim, Minsun. “Deepfakes and Democracies: Building Resilience against a Common Threat,” 2025. https://theyonseijournal.com/wp-content/uploads/2025/07/Deepfakes-and-Democracies-Building-Resilience-against-a-Common-Threat.pdf.

Oluwasegun Olakoyenikan, and Samson Olufemi Olanipekun. “LEVERAGING ARTIFICIAL INTELLIGENCE to COMBAT DISINFORMATION and DEEPFAKES in BROADCAST JOURNALISM and GLOBAL MEDIA COMMUNICATIONS.” International Journal of Engineering Technology Research & Management Volume 7, no. Issue 12 (December 14, 2023): 387–404. https://doi.org/10.5281/zenodo.15636618.

Romanishyn, Alexander, Olena Malytska, and Vitaliy Goncharuk. “AI-Driven Disinformation: Policy Recommendations for Democratic Resilience.” Frontiers in Artificial Intelligence 8 (July 31, 2025). https://doi.org/10.3389/frai.2025.1569115.

Twomey, John G, Didier Ching, Matthew P Aylett, Michael Quayle, Conor Linehan, and Gillian Murphy. “Do Deepfake Videos Undermine Our Epistemic Trust? A Thematic Analysis of Tweets That Discuss Deepfakes in the Russian Invasion of Ukraine.” PubMed 18, no. 10 (January 1, 2023): e0291668–68. https://doi.org/10.1371/journal.pone.0291668.

Vasist, Pramukh Nanjundaswamy, Debashis Chatterjee, and Satish Krishnan. “The Polarizing Impact of Political Disinformation and Hate Speech: A Cross-Country Configural Narrative.” Information Systems Frontiers 26, no. 1 (April 17, 2023): 1–26. https://doi.org/10.1007/s10796-023-10390-w.

Westerlund, Mika. “The Emergence of Deepfake Technology: A Review.” Technology Innovation Management Review 9, no. 11 (November 2019): 39–52. https://timreview.ca/article/1282.

  1. Arifin, Moh. Samsul. “Artificial Intelligence (AI) Dan Politik: Analisis Terhadap Peran Dan Signifikansi Ai Dalam Komunikasi Politik 2024 Di Indonesia.” Ilmu Sosial Dan Humaniora 6a no. 1 (2025). http://dx.doi.org/10.28944/hudanlinnaas.v6i1.1860. ↑
  2. Battista, Daniele. “Communication and Politics in the Age of Artificial Intelligence: An Overview of Deepfakes and Their Implications.” Society Register 8, no. 2 (June 30, 2024): 7–24. https://doi.org/10.14746/sr.2024.8.2.01. ↑
  3. Crawford, Kate. Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence. Yale University Press, 2021; Brennen, Scott Babwah, and Matt Perault. “The New Political Ad Machine: Policy Frameworks for Political Ads in an Age of AI ,” 2023. https://techpolicy.unc.edu/wp-content/uploads/2023/11/GAI-and-political-ads.pdf.. ↑
  4. Kedar, Hadas Emma. “Fake News in Media Art: Fake News as a Media Art Practice vs. Fake News in Politics.” Postdigital Science and Education 2 (July 13, 2019). https://doi.org/10.1007/s42438-019-00053-y. ↑
  5. Ibid.; Bennett, W Lance, and Steven Livingston. “The Disinformation Order: Disruptive Communication and the Decline of Democratic Institutions.” European Journal of Communication 33, no. 2 (April 2018): 122–39. https://doi.org/10.1177/0267323118760317. ↑
  6. Chenoweth, Erica. Social Movements in the Age of Fake News with Erica Chenoweth. Interview by Mari Megias, 2020. https://www.hks.harvard.edu/wiener-conference-calls/erica-chenoweth#transcript-1766694. ↑
  7. Vasist, Pramukh Nanjundaswamy, Debashis Chatterjee, and Satish Krishnan. “The Polarizing Impact of Political Disinformation and Hate Speech: A Cross-Country Configural Narrative.” Information Systems Frontiers 26, no. 1 (April 17, 2023): 1–26. https://doi.org/10.1007/s10796-023-10390-w. ↑
  8. Battista, Daniele. “Communication and Politics in the Age of Artificial Intelligence: An Overview of Deepfakes and Their Implications.” Society Register 8, no. 2 (June 30, 2024): 7–24. https://doi.org/10.14746/sr.2024.8.2.01. ↑
  9. Ibid; Harsin, Jayson. “Regimes of Posttruth, Postpolitics, and Attention Economies.” Communication, Culture & Critique 8, no. 2 (February 24, 2015): 327–33. https://doi.org/10.1111/cccr.12097; Twomey, John G, Didier Ching, Matthew P Aylett, Michael Quayle, Conor Linehan, and Gillian Murphy. “Do Deepfake Videos Undermine Our Epistemic Trust? A Thematic Analysis of Tweets That Discuss Deepfakes in the Russian Invasion of Ukraine.” PubMed 18, no. 10 (January 1, 2023): e0291668–68. https://doi.org/10.1371/journal.pone.0291668. ↑
  10. Westerlund, Mika. “The Emergence of Deepfake Technology: A Review.” Technology Innovation Management Review 9, no. 11 (November 2019): 39–52. https://timreview.ca/article/1282. ↑
  11. Busch, Ella, and Jacob Ware. “The Weaponisation of Deepfakes Digital Deception by the Far-Right,” 2023. https://icct.nl/sites/default/files/2023-12/The%20Weaponisation%20of%20Deepfakes.pdf. ↑
  12. Battista, Daniele. “Communication and Politics in the Age of Artificial Intelligence: An Overview of Deepfakes and Their Implications.” Society Register 8, no. 2 (June 30, 2024): 7–24. https://doi.org/10.14746/sr.2024.8.2.01. ↑
  13. Busch, Ella, and Jacob Ware. “The Weaponisation of Deepfakes Digital Deception by the Far-Right,” 2023. https://icct.nl/sites/default/files/2023-12/The%20Weaponisation%20of%20Deepfakes.pdf. ↑
  14. Jaursch, Julian, Théophile Lenoir, Bret Schafer, and Etienne Soula. “Tackling Disinformation : Going beyond Content Moderation.” Institut Montaigne, 2021. https://www.institutmontaigne.org/en/expressions/tackling-disinformation-going-beyond-content-moderation. ↑