A political argument does not require the humiliation of a people. That is the line crossed by an AI-generated video published by the state-run China Daily on 10 July 2026 and later condemned by the Philippine government as racist, offensive, and dehumanizing.
The video used a monkey character to represent Filipinos in a dispute over the South China Sea. According to reporting by Reuters and the Associated Press, the character wore Filipino-inspired clothing, was directed by arms representing the United States and Japan, held lyrics labeled with the South China Sea arbitration award, and was eventually thrown into the water and struck by a vessel’s water cannon.
Those images were not a neutral summary of a diplomatic dispute. They turned a nationality into an animal caricature, added physical punishment, and presented the result as political messaging. AI helped make the scene. People chose the idea, approved it, and published it.
What is verified: China Daily published the AI-generated video on Facebook. The Philippine Department of Foreign Affairs lodged a formal protest and demanded its removal. China’s foreign ministry later said the video did not represent China’s official position. These points are supported by Reuters and Associated Press reporting linked below.
What happened after publication
The Philippine response was direct. The Department of Foreign Affairs raised a firm objection with Chinese Ambassador Jing Quan. The Philippine Embassy in Beijing also wrote to China Daily’s editor-in-chief, and the government demanded that the material be taken down.
Defense Secretary Gilberto Teodoro described the video as dehumanizing. That description is important because the problem extends beyond insult. Animalizing a national or racial group removes individuality and dignity. It asks an audience to see people as lesser, ridiculous, or deserving of harm.
China’s foreign ministry spokesperson Lin Jian said the video did not represent the country’s official position and did not offer a further comment on the production. The ministry also repeated Beijing’s rejection of the 2016 South China Sea arbitral ruling.
That distinction should remain clear. China Daily is a state-run media outlet. A publication by that outlet is consequential, but it is not accurate to rewrite the foreign ministry’s response and call the video a formal statement of the Chinese government. Precision makes criticism stronger, not weaker.
AI did not remove human responsibility. It made a human editorial choice faster to produce and easier to circulate.
Why this was not harmless satire
Satire can expose hypocrisy, challenge power, or reveal the absurdity of a policy. Its sharpness does not exempt it from ethical judgment. When a message reduces Filipinos to a monkey and then depicts punishment through maritime violence, the target is no longer only a legal argument or a government decision. The target becomes a people.
The timing deepened the political meaning. The video appeared around the tenth anniversary of the 2016 arbitral award in the case brought by the Philippines under the United Nations Convention on the Law of the Sea. The tribunal rejected the legal basis for expansive historic-rights claims beyond the limits allowed by the convention. China has consistently rejected the award.
There is room for governments and commentators to argue about diplomacy, security partnerships, and maritime policy. There is no need to use racial imagery to do it. A persuasive case should be able to stand on evidence and law without degrading the identity of the people on the other side.
What AI changes in propaganda
AI-generated video lowers the cost of producing scenes that once required a studio, actors, animation teams, or extensive post-production. It can generate emotionally charged material quickly, adapt it to different platforms, and repeat a visual narrative at scale.
That efficiency also creates a dangerous temptation: to treat synthetic content as though the machine, not the publisher, is responsible for the message. But an AI system does not decide to open an official media account, select a geopolitical target, approve a racist metaphor, and press publish. Those are editorial and institutional decisions.
Responsible use therefore begins before generation. Editors should ask who is being depicted, what historical meaning the chosen imagery carries, whether the message attacks conduct or identity, and what harm could follow when it is distributed through a powerful public channel.
Before publication
Interrogate the metaphor.
If the visual depends on animalizing a nationality, race, or ethnic group, it has already abandoned serious argument.
During review
Keep a human accountable.
A named editor, not a model or software vendor, must own the decision to approve and distribute synthetic media.
After harm
Correct without evasion.
Remove the material, acknowledge why it was wrong, and explain the safeguards that will prevent a repeat.
FMB&CO. perspective · Opinion
Defend dignity without creating a new prejudice
We reject the portrayal of Filipinos as an animal caricature. We also reject the idea that one offensive publication gives anyone permission to insult Chinese people as a whole. The responsible response names the publisher, the imagery, the editorial choice, and the harm with precision.
Filipino dignity does not need exaggeration to be defended. It needs clarity, courage, and a refusal to answer dehumanization with another form of dehumanization.
The strongest public communication does not merely win attention. It protects the humanity of the people it asks the public to see.
What accountability should look like
The immediate standard is simple: the video should not remain in circulation through an official media channel. A credible response would also acknowledge the racist device at the center of the production, clarify how it passed editorial review, and establish a meaningful process for AI-assisted content.
The larger lesson is not limited to one outlet or one country. Political actors everywhere are learning how cheaply synthetic media can provoke anger, mock opponents, and bypass the discipline of evidence. Every newsroom, agency, and public institution using these tools will eventually reveal what it values through the content it is willing to publish.
Technology can manufacture an image. It cannot manufacture the ethical judgment required to decide whether that image belongs in public life.