Singapore proposes new law to ban deepfakes

Singapore has recently proposed a new law to ban deepfakes and digitally manipulated content of election candidates, aiming to safeguard the integrity of the country’s electoral process.

Introduced by the government’s Ministry of Digital Development and Information (MDDI), the ban would reportedly be applied during the election period. Singapore’s next General Election will be held in November 2025.

Proponents said the new law targets AI-generated misinformation that misrepresents candidates, recognizing the rising global threat of hyper-realistic deepfakes.

The law includes penalties for publishing or sharing such content, with steep fines for non-compliant social media platforms.

While Singapore already has legislation to combat falsehoods, this bill addresses the specific risks deepfakes pose to electoral outcomes, emphasizing the need for enhanced cybersecurity in the political landscape.

The new law can be used if it meets three conditions, according to this report. These conditions are:

  • The content is digitally generated or manipulated using AI techniques as well as non-AI techniques such as Photoshop, dubbing, and splicing.
  • The content can be “reasonably regarded” as online election advertising, where it intends to promote, procure or prejudice the electoral prospects of a party or candidate. 
  • The content must realistically show a candidate saying or doing something that they did not say or do. 

Dangers of false information

Boris Cipot, Senior Security Engineer, Synopsys Software Integrity Group, commented that the bill addresses a serious issue.

Deepfakes – realistic videos created using AI that make it appear as if the person in the video is doing or saying something they never did – are a big issue that can sway a viewer into believing false information, Cipot said

“In the past, deepfakes videos have been showcased several times to show how they can mislead the public. Popular deepfake videos of Barack Obama from 2018 and Nancy Pelosi from 2019, both showcase what this technique of video editing is capable of and how a realistic videos created by AI can influence the publics thoughts,” the executive added.

Abhishek Kumar Singh, Head, Security Engineering, Singapore, Check Point Software Technologies added: “Singapore’s proposal to ban deepfakes of candidates during elections is a proactive move, aligning with its broader strategy to protect the integrity of its democratic processes and prevent misinformation.

“Deepfakes—highly convincing yet entirely fabricated video or audio content—pose a substantial threat to electoral systems worldwide. They can be used to spread disinformation, mislead voters, and potentially damage the reputations of political candidates. As deepfakes become more sophisticated and accessible, their ability to create confusion and distrust in electoral outcomes grows, making such regulatory actions increasingly critical,” Singh said.

Singapore’s emphasis on cybersecurity and digital literacy, along with its clear regulatory measures, places it at the forefront of countries combating the misuse of AI-driven technology in politics, the executive further said.

The Philippines’ own AI law?

A lawmaker from the Philippines recently said the government should have “regulatory” control of AI, particularly when AI is used for “deepfake” videos.

Camarines Sur Rep. Luis Raymund Villafuerte has filed a bill in Congress (House Bill 10567) which also proposes monetary fines for violators.

Penalties on distributing AI without public disclosure

The proposed law penalizes individuals who will use AI to produce materials and distribute them without making a public disclosure that these are deepfakes, or AI-altered visual and/or audio presentations.

The penalties proposed in House Bill 10567 range from P2 million to P5 million for producers or distributors of deepfake materials. This is also the same penalty for those who delete or tamper without such disclosures of AI-modified voices or images.

AI to be used for next year’s elections?

Villafuerte introduced the bill just weeks after backing the proposal of Commission on Elections (Comelec) Chairman George Erwin Garcia for the electoral watchdog to prohibit the use of AI and deepfake technology in electoral campaigns, to preserve honesty and transparency in nationwide balloting in 2025 onward, the news daily Manila Standard further reported.

The lawmaker further cited a reported survey, which was posted on TikTok, supposedly reporting the results of an election survey. The survey was reportedly made by a major broadcasting firm but was found out to be fake.

Candidates face disqualification with AI, deepfake

Garcia earlier said disqualification and election offense cases to be filed against candidates who would use AI and deepfake technology in next year’s midterm polls.

“I will propose to the Commission en banc that we make it as an election offense case as well as disqualification charges,” Garcia said.

The poll body chief noted that allowing the new technology to promote one’s candidacy would be dangerous.

“The entire country could face chaos if we allow the presence of AI and deepfake because it can make it appear that a person is bad, or very good, or good-looking when it is not true,” he added.

Garcia, meanwhile, assured they were prepared to motu proprio file charges against these bets, noting that they “cannot just allow such activities without taking actions.”

He, however, said it would also be better if the public would participate and initially file charges against these candidates.

READ MORE AI NEWS.

One thought on “Singapore proposes new law to ban deepfakes

Comments are closed.