Election Commission of India Mandates Labeling of AI-Generated Content in Political Campaigns

ECI mandates label use for AI generated content in poll campaigns

The Election Commission of India (ECI) has issued an advisory to political parties directing them to label synthetic/AI-generated content used during election campaigning. Political parties must clearly label any images, videos, or audio content generated or "significantly altered" using AI technologies as "AI-generated", "digitally enhanced", or "synthetic content". Further, parties should give disclaimers wherever they include such content within campaign advertisements or promotional content disseminated online. The ECI claims such efforts would aid "fair, transparent, and responsible" usage of technologies during election campaigning. This notice comes days after the Aam Aadmi Party (AAP) faced a first information report (FIR) for allegedly posting AI-generated content of Prime Minister Narendra Modi and Home Minister Amit Shah on the party's X handle. However, this is not the first instance of deepfake usage during election campaigning. The BJP deployed a similar tactic during Maharashtra's Assembly elections when it published a deepfake audio of NCP (SP) MP Supriya Sule on X, which fact-checkers later found to be manipulated. Previously during India's General Elections 2024, the ECI issued guidelines to political parties, discouraging the use of AI-generated distorted content, and deepfakes, among other tools on social media to disseminate misinformation. Among other provisions, the guidelines noted that there must be no publishing and circulation of deepfake audio/videos from the political parties' social media handles. If published, such content should be removed within three hours of its notice and a warning should be issued to the accountable person. Taking note of the influence of AI-generated content on elections, many companies including OpenAI and other Big Tech entities sought to establish guardrails to prevent the "deceptive use" of this technology. Meta also announced plans to label AI-generated content that users post across Facebook, Instagram, and Threads. For images, the company explained its reliance on industry-standard indicators that companies like OpenAI, Midjourney, and Adobe, etc, worked on to add labels to AI tools. Later, the Mark Zuckerberg-led company expanded its efforts, introducing "Made with AI" labels on such content, now updated to "AI info". Besides Meta, OpenAI added signature watermarks to images produced using its DALL.E 3 image generator, while content generated within ChatGPT contained a manifest to indicate its origin. However, such methods are not foolproof. While speaking at MediaNama's 'Deepfakes and Democracy' event, Gautham Koorma, a researcher at UC Berkley pointed out that while watermarking could be a potential solution, 'sophisticated adversaries' could easily break it. Notably, another participant, Tarunima Prabhakar, co-founder of Tattle Civic Technologies suggested other methods like fingerprinting (adding provenance info to) deepfakes at the source to avoid future possibilities of deepfakes and AI bypassing technical detection. However, fact-checkers previously speaking at a MediaNama event on 'Fact-Checking and Combating Misinformation in Elections' expressed that despite deepfakes having a limited impact on India's General Elections 2024, they led to an erosion of trust. Kritika Goel, Head of Editorial Operations (India) at Logically Facts also noted that individuals could misuse deepfakes to claim real audios as false, leveraging "plausible deniability" or "liars dividend".

ECI Mandates Labelling of AI Generated Content in Political Campaigns

Ahead of the Delhi assembly elections, The Election Commission of India (ECI) has reinforced its directive for political parties to label and disclose AI-generated and synthetic content used in election campaigns. The advisory mandates that parties explicitly label images, videos, audio, or other materials significantly altered by AI technologies with notations such as "AI-Generated," "Digitally Enhanced," or "Synthetic Content." Additionally, disclaimers must accompany campaign advertisements or promotional materials utilising synthetic content. Chief Election Commissioner Rajiv Kumar has consistently warned about the dangers of AI and deepfakes exacerbating misinformation. In a statement, he emphasised that such technologies have the potential to undermine public trust in electoral processes. In a letter to the presidents, general secretaries, and chairpersons of all national and state-recognised political parties, ECI Joint Director Anuj Chandak highlighted how advancements in AI have enabled the creation of highly realistic synthetic content, including images, videos, and audio. Acknowledging the growing impact of AI-generated and synthetic content on public opinion, the Election Commission has urged political parties, their leaders, candidates, and star campaigners to prominently label such content when shared on social media or other platforms during campaigns. Last year, during the Lok Sabha elections, the ECI issued guidelines for the ethical and responsible use of social media platforms, further demonstrating its commitment to maintaining transparency and fairness in campaigns. The latest advisory aligns with the ECI's broader efforts to ensure a level playing field in elections, particularly through responsible use of AI and digital platforms. During the Global Election Leaders Summit (GELS) 2024, the Commission reiterated the importance of ethical practices in leveraging technology for electoral campaigns.

EC asks parties to label all AI-generated campaign content

Tech companies are turning to 'synthetic data' to train AI models - but there's a hidden costAhead of the Delhi assembly elections, the Election Commission of India (ECI) has asked all the political parties to ensure "prominent labelling" of any AI-generated/synthetic content being shared as campaign material on their social media and other platforms. The poll panel's advisory comes amid increasing instances of political parties deploying AI-generated content to target their opponents and as campaign material. The ECI, in a communication to all political parties on Thursday, has noted the "deepening influence of AI-generated/synthetic content in shaping public opinion" and has advised that their leaders, candidates and star campaigners take necessary measures to bring in prominent and easily discernible labelling of AI-generated content to ensure "responsible and transparent" poll campaigning. The poll panel has sought that any image, video, audio or other materials generated or significantly altered by AI technologies should be labelled with a notation such as "AI-Generated", "Digitally Enhanced" or " Synthetic Content". Parties are also required to include disclaimers in campaign advertisements or promotional content disseminated online or at other platforms indicating use of synthetic content is utilised. The ECI has noted that amid increasing use of advanced AI technologies in political campaigning, it is essential to maintain transparency and accountability, as AI-generated content, including images, videos and audio, has the potential to "influence voter opinion and trust". Chief Election Commissioner Rajiv Kumar has also cautioned on disinformation concerns around deep fakes and artificial intelligence (AI). Following Delhi High Court's 2024 directions in a public writ petition, the poll panel had also issued guidelines amid the Lok Sabha polls to curb wrongful use of AI content and "deep fakes" by political parties that could disturb the level playing field.

Election Commission issues advisory on use of AI in poll campaigning

The Election Commission has issued guidelines to political parties for transparent use of AI-generated content in campaigns. Parties must label AI-altered materials and include disclaimers on synthetic content in advertisements. Chief Election Commissioner Rajiv Kumar highlighted concerns over AI and deep fakes potentially spreading misinformation and undermining trust in elections.Against the backdrop of increasing use of artificial intelligence in campaigning and its potential to influence voter opinion, the Election Commission on Thursday issued an advisory to political parties asking them to promote transparency and accountability in use of AI-generated content. The advisory introduces labelling and disclosure norms requiring parties to clearly label any images, videos, audio or other materials generated or significantly altered by AI technologies with a notation such as "AI-Generated"/ "Digitally Enhanced"/"Synthetic Content". It also requires political parties to include disclaimers during dissemination of campaign advertisements or promotional content, wherever synthetic content is utilised. Chief Election Commissioner Rajiv Kumar had recently cautioned against the potential of AI and deep fakes in compounding the malaise of misinformation. He had expressed concern that deep fakes and misinformation have the potential to erode trust in electoral processes. During last year's Lok Sabha elections, the Commission had issued guidelines for responsible and ethical use of social media platforms.