Vietnam Says Platforms Are Using AI to Target AI Misinformation as Verification Campaign Expands

Platforms and the public face a shared test

Vietnam's Ministry of Public Security says international digital platforms have begun using artificial intelligence to tackle false information and fabricated images and videos in the country, alongside a public campaign launched on July 31, 2026, to encourage people to check claims before believing them.

At an October 1 briefing on the ministry's work during the third quarter of 2026, Colonel Ha Van Bac, deputy head of its Cybersecurity Department, said Vietnam was the first country where platforms operating across national borders had applied AI to address those categories of content. The claim came from the department, which has been pressing platforms to develop specific responses to misuse of the technology.

Bac said assessing the measures would take time because the AI systems needed to learn and adapt to the technology, proposed solutions and conditions in practice. The announcement therefore pairs a claim about deployment with an acknowledgment that its effectiveness still needs assessment.

Vietnam's response has two connected parts: platforms and police are being asked to identify and address suspected violations, while broadcasters, educators and cybersecurity organisations are encouraging users to verify information themselves. The practical stakes range from protecting a person's reputation to stopping a convincing fake video from drawing someone into a financial scam.

What the platform response involves

The Cybersecurity Department said it had worked with international platforms and called for measures aimed specifically at activity involving AI. According to Bac, the companies have developed responses to violations involving the technology, including false information and AI related images and videos.

The briefing did not identify the platforms, describe the detection systems, provide performance figures or give a date for an effectiveness assessment. Those details matter to understanding how widely the measures operate and how consistently they identify fabricated content.

AI can help examine large volumes of posts, images or video, but deciding whether a claim is true also requires context. A computer generated picture may be an openly labelled illustration, while a genuine photograph can mislead if someone attaches a false description. The public verification campaign addresses that wider problem of deciding what a piece of content actually means.

For users, the immediate change is that more of the response may happen through platform moderation. The department's account also places responsibility on the companies hosting and distributing content, alongside the people who create or share it.

How police are pursuing suspected violations

The Cybersecurity Department is working with local police to verify accounts, identify account administrators and establish who posted content suspected of violating the law. This adds an investigative process to the platforms' technical response.

For accounts or individuals abroad, Vietnamese authorities ask the relevant platforms to cooperate by blocking or removing content. International cooperation is therefore a practical part of the response: the person posting material, the company hosting it and the audience seeing it may all be in different countries.

People whose names, faces or voices are used in fabricated content can face damage before an investigation is complete. Lieutenant General Le Xuan Minh, director of the department responsible for cybersecurity and preventing crimes involving advanced technology, warned at the campaign launch that widely circulated falsehoods could harm individuals, businesses and public organisations and create public anxiety.

The combined approach addresses different stages of the same problem. Platform action can restrict circulation, police can investigate suspected offences, and public verification can reduce the chance that users accept or forward a false claim in the first place.

A campaign built around checking first

The TinAI? campaign was officially launched in Hanoi on July 31, 2026, by the ministry's cybersecurity department, the Digital Trust Alliance and the Vietnam National Cybersecurity Association. Vietnam Television and Phenikaa University are also involved in the wider verification initiative.

The Ministry of Public Security's official launch announcement, published on August 1, 2026, confirms that the ceremony took place on the morning of July 31. The nationwide campaign is backed by the ministry, with financial technology group MoMo and Vietnam Television also supporting it.

The Vietnamese message, "Kiểm trước tin sau," translates as "Check first - believe later." Its central request is straightforward: examine where information came from, compare the evidence and pause before sharing or acting on it.

Minh said AI had made fabricated images, videos and voices faster and more sophisticated to produce. He argued that protecting people online now extends beyond securing computer systems, data and digital infrastructure to protecting public understanding and trust.

His explanation also set a limit on what institutions can do alone. No tool can verify every item for every user, he said, and no authority can keep pace with all the content being created and shared. The campaign consequently seeks a routine habit of verification across the public, supported by media organisations, technology companies, content creators and other participants.

Minh stressed that responsibility for information remains with people regardless of how technology changes. He described checking before believing and considering before sharing as practical steps to protect individuals and communities and build a safe, trustworthy online environment.

The launch also included a discussion titled "Xây dựng phản xạ kiểm chứng trong kỷ nguyên AI," or "Building a verification reflex in the AI era." Representatives of regulators, news organisations, technology businesses and digital platforms joined experts and content creators to discuss AI's challenges and their shared responsibility for a transparent, trustworthy information environment.

A daily television bulletin joins the effort

Vietnam Television launched the daily bulletin "Thật-Giả, kiểm chứng," translated as "Truth or Falsehood - Fact Check," with a scheduled broadcast time of 19:55. The programme gives the campaign a regular place in the television schedule rather than relying only on launch events or occasional warnings.

The five-minute bulletin airs daily on VTV1. It was unveiled at the July 31 launch by Vietnam Television's news division in coordination with the Cybersecurity and High-Tech Crime Prevention Department.

The bulletin is intended to verify claims attracting public attention, explain tactics used to create and spread misinformation, and present verification skills in a short, accessible format. It will also cover emerging scams and fabricated content produced with AI and other technologies.

VTV Director General Nguyen Thanh Lam said verification should protect positive values as well as expose falsehoods. That gives the programme a broader purpose: helping audiences find reliable information while explaining why deceptive claims fail scrutiny.

Lam, who is also deputy head of the Central Commission for Information, Education and Mass Mobilisation, identified four sources of misinformation: AI, social media platform algorithms, users lacking verification skills, and organisations or individuals deliberately spreading falsehoods for their own motives. He said each group required a different response rather than being treated alike.

He also warned that misinformation extends beyond deepfake scams, fake banking websites and calls impersonating authorities. Information taken out of context, including distorted policies or leaders' statements, can completely change the original meaning and spread rapidly before authorities respond.

The wider campaign plans to distribute videos, infographics and verification bulletins through traditional media, social media and digital platforms. Using several channels gives the message a route to people who may encounter a false claim on a phone rather than during a television broadcast.

Why convincing AI answers still need evidence

Associate Professor Le Anh Son, Phenikaa University's director of innovation, said AI is never fully accurate, that answers should be supported by cited sources, and that people remain responsible for verification. His comments connect the campaign to everyday use of generative tools, including asking questions and producing written explanations.

Generative AI creates text, images, audio or video in response to instructions. Its output can look polished or sound confident even when details are wrong. A citation gives a reader somewhere to check a claim, although the reader still needs to examine whether the cited material supports it.

A deepfake is an image, recording or video that uses AI to imitate a person or make that person appear to say or do something. The danger becomes concrete when a familiar face or authoritative voice persuades someone to disclose information, transfer money or share an accusation.

Vietnam's education campaign and platform measures address different needs here. Automated systems can help process content at scale, while people still need to check identities, sources and context before making decisions that affect their money, reputation or safety.

Warnings abroad show several forms of harm

In the United States, the FBI's Internet Crime Complaint Center warned on July 20, 2026, that scammers were using AI generated videos to impersonate FBI employees and leaders and promote fake complaint websites. The videos appeared on social media to gain trust and draw people into providing personal and financial information.

The scam exploits the authority of the person being impersonated. Someone seeking help with a crime could instead be directed to a fraudulent service that appears to carry official backing. The deceptive video supplies the introduction, while the fake website collects the information.

China has pursued enforcement alongside rules requiring identification of content generated or altered with AI. In September 2026, the Cyberspace Administration of China said a campaign targeting AI misuse had removed more than 5.61 million violating items and taken action against more than 49,000 accounts and 2,400 websites and applications. The violations included spreading false information, impersonating others and distributing harmful content.

Japan has approached AI misinformation as a matter of information security and public trust, strengthening coordination among government agencies to collect, analyse and respond to information manipulation. It has also warned that generative AI could increase the spread of misinformation and influence public perceptions.

Germany's Bundestag Office of Technology Assessment said in a March 2026 report that the country lacked comprehensive rules for handling AI generated content. Responses relied largely on existing civil and criminal law, with enforcement remaining difficult. The German government has also acknowledged the use of deepfakes and other AI material in campaigns intended to deepen divisions and weaken trust.

How the response developed during 2026

The year's developments connect legal concerns, fraud warnings, public education and platform enforcement. Vietnam's October announcement followed its July campaign launch, while other countries were addressing impersonation, large volumes of prohibited content and attempts to manipulate public opinion.

• March 2026: Germany's Bundestag Office of Technology Assessment described gaps in comprehensive rules for AI generated content and difficulties with enforcement.
• July 20, 2026: The FBI's Internet Crime Complaint Center warned about fabricated videos impersonating FBI personnel and directing users to fake complaint websites.
• July 31, 2026: Vietnam launched TinAI? in Hanoi alongside a daily television verification bulletin.
• September 2026: China's internet regulator announced millions of content removals and action against accounts, websites and applications over AI misuse.
• October 1, 2026: Vietnam's Ministry of Public Security described platform measures using AI and said their effectiveness would take time to assess.

Vietnam's next practical test is whether the platform measures reduce harmful circulation while the campaign builds verification habits. The ministry has described the work underway; its assessment of results is still to come.

Key Points

• Vietnam's cybersecurity department says international platforms are using AI in the country to address false information and fabricated media.
• Colonel Ha Van Bac attributed a first country claim to the deployment and said effectiveness would require time to assess.
• Police are verifying suspected accounts and administrators, with platform cooperation sought for content involving people abroad.
• The TinAI? campaign launched on July 31, 2026, with a message to check information before believing it.
• A five-minute daily VTV1 bulletin scheduled for 19:55 supports verification education alongside videos, infographics and digital distribution.
• Developments in the United States, China, Japan and Germany show responses spanning fraud prevention, enforcement, public security and legal rules.