by Almira Louise S. Martinez, Reporter
Falsely generated videos, images, and audio powered by artificial intelligence (AI) can stir up trouble in the 2025 Philippine elections, a multinational technology company said.
“Some of the deepfake stuff that gets put on there is quite funny, but it is a real problem,” International Business Machines Corporation (IBM) APAC Vice President, Government, and Regulatory Affairs Stephen Braim said in a virtual interview on Monday.
According to Mr. Braim, deepfakes have been widely used during elections in different countries like the United States and Taiwan.
“They had a major problem with one of their neighbors using deepfakes trying to influence the election,” he said on the 2024 presidential election in Taiwan.
Global superstars and creators were common victims of deepfakes during election season. “Their deep fakes are used in political advertising and to rip them off as content creators,” Mr. Braim said.
In India, Bollywood actors Aamir Khan and Ranveer Singh were used in viral deepfake videos criticizing Prime Minister Narendra Modi during the general election.
Additionally, an old video of Home Minister Amit Shah was manipulated using a basic editing tool called ‘cheapfake’ for a political agenda against the then-ruling party.
Aside from known faces, even the general public can experience the damaging impacts of AI misuse. “There’s a lot of this stuff that’s used to trick people out of giving money over to what they think is a family member,” Mr. Braim said.
Despite the harmful effects it may cause in the political sphere, proper safeguards are still not in place to protect the people.
“There isn’t a silver bullet technology out there to handle deepfakes, whether either in the Philippines or anywhere around the world,” he told BusinessWorld.
Mr. Braim said that the government plays a role in regulating and holding users and developers of deepfakes accountable for their actions. “We think regulation should very much be around to protect elections against the use of deepfakes.”
He added that user trust is an integral part of the generative AI era. With all the negative implications of AI, Mr. Braim said it undermines people’s trust in a technology that can be useful for productivity, growth, and health outcomes.
“You know the list is endless. We just got to marginalize the bad use of it,” he said.