Hero image: Tara Winstead / Pexels
South Korea Warns of AI Deepfake Ad Scams Targeting Consumers
South Korea’s consumer protection chief has sounded the alarm over AI-generated advertisements that use celebrities’ likenesses without consent, prompting calls for legal action and stronger oversight amid a surge in deceptive online campaigns.
In late August 2026, South Korea’s senior consumer affairs official, Lee Ji-hye, publicly warned that AI deepfake advertisements are being used to deceive consumers by impersonating well-known figures in endorsements they never made. This warning comes as digital impersonation tools become more accessible and sophisticated, raising concerns about fraud, reputational harm, and the erosion of public trust in advertising. This report synthesizes available reporting to assess the scope of the problem, the mechanisms of misuse, and the adequacy of current legal and regulatory responses.
South Korea’s Deepfake Ad Crisis: What Lee Ji-hye’s Warning Reveals
South Korea’s consumer protection landscape is confronting a rapidly evolving threat: AI-generated advertisements that leverage the voices, faces, and personas of celebrities and public figures without their consent. The warning issued by Lee Ji-hye, head of the country’s consumer affairs office, highlights a growing pattern in which scammers use generative AI to create lifelike endorsements, product promotions, and public service announcements that appear to come from trusted individuals. These deepfake ads are then distributed across social media, messaging platforms, and even legitimate advertising networks, exploiting the credibility of well-known figures to trick consumers into purchases, data sharing, or financial transfers.
While the phenomenon is not unique to South Korea, the country’s advanced digital infrastructure, high social media penetration, and strong celebrity culture make it particularly vulnerable. The scale of the issue remains difficult to quantify due to the ephemeral and often cross-border nature of digital advertising, but the official warning signals that the problem has reached a threshold requiring immediate attention from regulators, platforms, and law enforcement.
What Outlets Are Reporting: A Single-Source Snapshot and Its Limits
As of this publication, detailed, multi-outlet investigative reporting on the Lee Ji-hye warning remains limited in the English-language press. The primary source of information is Chosun Ilbo, a major South Korean newspaper, which published a report on August 30, 2026, detailing the consumer affairs chief’s call for legal action against the creators and distributors of AI deepfake advertisements. The article emphasizes the urgency of the issue and the lack of existing legal tools to address it effectively.
Notably, no other major international or domestic outlets have yet published corroborating reports, investigative features, or data-driven analyses on the scale or mechanisms of these scams. This limited coverage restricts the ability to independently verify the claims or assess their broader context. However, the Chosun Ilbo report provides a foundational account of the warning, the legal gaps identified, and the proposed remedies, which this synthesis uses as a starting point for analysis.
What Chosun Ilbo Reports
According to Chosun Ilbo, Lee Ji-hye has urged the government to classify AI deepfake advertisements as a form of “fraudulent misrepresentation,” a legal category that could trigger penalties for creators and platforms that fail to remove such content promptly. The article notes that current laws do not explicitly address AI-generated impersonations in advertising, leaving consumers with limited recourse when they encounter deceptive endorsements. The consumer affairs chief is quoted as saying that the proliferation of these ads threatens to undermine public trust in both digital advertising and the celebrities whose likenesses are being exploited.
Chosun Ilbo also reports that the government is considering amendments to the Act on the Consumer Protection in Electronic Commerce and the Framework Act on Telecommunications, which could introduce obligations for platforms to detect and remove deepfake ads within a specified timeframe. The article suggests that without such measures, the problem is likely to escalate as AI tools become more accessible and realistic.
The Claim: AI-Generated Ads Using Celebrity Likeness Without Consent
The central claim advanced by Lee Ji-hye and reported by Chosun Ilbo is that AI-generated advertisements are being created using the voices, faces, and personas of celebrities and public figures without their consent. These ads are designed to appear authentic, often mimicking the tone, style, and visual identity of the impersonated individual. The goal, according to the warning, is to deceive consumers into believing that the celebrity endorses a product, service, or campaign when no such endorsement exists.
The claim implies a dual harm: financial exploitation of consumers and reputational harm to the impersonated individuals. By leveraging the trust associated with well-known figures, scammers can increase the likelihood of engagement, clicks, and conversions, even when the underlying offer is fraudulent or misleading. The use of AI to generate these ads at scale suggests a systemic issue rather than isolated incidents, with potential for rapid proliferation across digital ecosystems.
What the Evidence Actually Shows: Pattern of Misuse and Legal Gaps
While the Chosun Ilbo report provides the most detailed account of the warning, it does not present systematic evidence such as quantified case counts, platform-specific data, or longitudinal trends. However, it does outline a clear pattern: the misuse of AI to generate endorsements that are then distributed through online channels where oversight is inconsistent. The absence of explicit legal prohibitions against AI deepfake advertising in South Korea’s current regulatory framework is a critical gap, as highlighted by the consumer affairs chief.
This legal gap is not unique to South Korea. Internationally, laws governing digital impersonation and deepfakes are often fragmented, with some jurisdictions addressing the issue through privacy, fraud, or defamation statutes, while others lack clear provisions altogether. The Chosun Ilbo report suggests that South Korea is at a crossroads, with policymakers now considering whether to treat AI deepfake ads as a distinct category of consumer harm requiring targeted intervention.
Mechanisms of Misuse
According to Chosun Ilbo, the process typically begins with the collection of publicly available data—such as interviews, speeches, or social media posts—from which AI models extract patterns to generate new content. These models can then produce audio or video clips that closely resemble the target individual’s voice or appearance. Scammers embed these generated clips into advertisements that are distributed via social media platforms, messaging apps, and even email campaigns. The ads often include calls to action such as “limited-time offers,” “exclusive deals,” or “urgent alerts,” designed to pressure consumers into responding quickly.
The report does not specify the platforms most affected, but it implies that social media networks with large user bases and algorithmic amplification are primary vectors. The lack of transparency in ad approval processes on many platforms further complicates detection and removal, as scammers can exploit loopholes or bypass safeguards by using novel accounts or cloaked domains.
Who Is Affected and How the Scams Spread Online
Consumers across demographic groups are potentially affected by AI deepfake ad scams, but the risk is likely highest among users who rely on social media for news and shopping recommendations. The Chosun Ilbo report does not provide demographic breakdowns, but it suggests that the scams target a broad audience, with a particular emphasis on products or services that benefit from perceived endorsements—such as health supplements, financial services, and luxury goods.
The scams spread rapidly due to the viral nature of digital content and the use of targeted advertising tools that allow scammers to reach specific audiences based on interests, behaviors, or location. Once a deepfake ad gains traction, it can be shared across networks, amplified by bots or fake accounts, and repurposed into new variations to evade detection. The ephemeral nature of many social media posts and the use of encrypted messaging apps further complicate efforts to track and contain these campaigns.
Potential Victims and Harm
While the Chosun Ilbo report focuses on consumer deception, the harm extends beyond financial loss. Consumers who fall for these scams may suffer reputational damage if they share the ads with their networks under the mistaken belief that the endorsement is genuine. Additionally, the impersonated celebrities may face reputational harm, loss of trust, and potential legal disputes if they are wrongly associated with fraudulent schemes. The report implies that the cumulative effect of these incidents could erode public confidence in digital advertising and celebrity endorsements as a whole.
Red Flags and a Debunking Checklist for Consumers
Given the sophistication of AI-generated content, consumers should adopt a skeptical approach to online endorsements, especially those involving celebrities or public figures. While no single red flag is definitive, a combination of warning signs can help identify potential deepfake ads. Below is a practical checklist grounded in the mechanisms described by Chosun Ilbo and general best practices for evaluating digital content.
- Unusual Endorsement Context: Be wary of ads that feature celebrities endorsing products or services outside their known expertise or typical brand affiliations. For example, a K-pop idol endorsing a financial trading platform or a politician promoting a supplement may warrant additional scrutiny.
- Urgency and Pressure Tactics: Deepfake ads often use high-pressure language, such as “limited-time offer,” “exclusive deal,” or “act now,” to prevent consumers from verifying the claim. Legitimate endorsements rarely rely on urgency as a primary selling point.
- Inconsistent Audio or Visual Quality: While AI tools are improving, subtle artifacts such as unnatural blinking, lip-sync errors, or distorted audio may indicate a deepfake. Pay attention to inconsistencies in lighting, shadows, or facial expressions.
- Unverified Source Links: Check the destination of any links in the ad. Scammers often use shortened URLs or cloaked domains that do not match the claimed brand or celebrity. Hover over links (without clicking) to preview the actual destination.
- Lack of Verification on Official Channels: Search the celebrity’s official social media accounts, website, or news coverage to see if they have publicly endorsed the product or service. If no such announcement exists, treat the ad as suspicious.
- Request for Personal or Financial Information: Be cautious of any ad that asks for sensitive data, such as credit card numbers, passwords, or government IDs, even if it appears to come from a trusted figure. Legitimate endorsements typically do not require direct data entry from consumers.
- Unusual Distribution Channels: Deepfake ads may appear in unexpected places, such as private messaging groups, obscure websites, or social media accounts with few followers. Be especially cautious of ads shared by unfamiliar accounts or bots.
Institutional Response: Legal Action and Regulatory Gaps
The Chosun Ilbo report highlights a proactive but reactive stance from South Korean authorities. Lee Ji-hye’s call for legal action reflects an acknowledgment that existing consumer protection laws are insufficient to address the unique challenges posed by AI deepfakes. The proposed amendments to the Act on the Consumer Protection in Electronic Commerce and the Framework Act on Telecommunications could introduce obligations for platforms to monitor, detect, and remove deepfake ads within a defined timeframe. However, the report does not specify whether these amendments would include penalties for non-compliance or mechanisms for consumer redress.
The lack of explicit legal definitions for AI deepfake advertising is a critical gap. Without clear prohibitions, platforms may struggle to justify proactive removals, and law enforcement may face difficulties in prosecuting creators. The Chosun Ilbo report suggests that the government is considering whether to classify these ads as a form of fraudulent misrepresentation, which could provide a legal basis for action. However, the effectiveness of such measures will depend on the speed of legislative action and the resources allocated to enforcement.
Platform Responsibilities
While the Chosun Ilbo report does not detail the role of digital platforms, it implies that they bear some responsibility for detecting and removing deepfake ads. Many platforms already have policies against impersonation and fraudulent content, but the use of AI-generated media complicates enforcement. Platforms may need to invest in AI-driven detection tools, human review teams, and user reporting mechanisms to identify and remove deepfake ads at scale. The report suggests that regulatory pressure could accelerate these efforts, particularly if new laws impose timelines for content removal.
Original Analysis: Why This Is a Global Warning Sign for AI Trust
Taken together, the warning from Lee Ji-hye and the mechanisms described by Chosun Ilbo suggest that South Korea is experiencing an early wave of a broader global phenomenon: the weaponization of AI-generated media to exploit consumer trust at scale. While the immediate harm is localized to South Korean consumers and celebrities, the underlying dynamics are not confined by borders. As AI tools become more accessible and realistic, the risk of deepfake advertising will likely spread to other markets with strong celebrity cultures and high digital engagement, including the United States, Japan, and parts of Europe.
This development is significant because it represents a shift from isolated deepfake incidents—such as political disinformation or personal revenge porn—to a systemic threat to commercial integrity. Unlike traditional advertising fraud, which relies on fake accounts or bot networks, deepfake ads leverage the credibility of real individuals, making them more persuasive and harder to detect. The absence of clear legal frameworks in many jurisdictions creates a permissive environment for scammers, while platforms struggle to balance detection with user privacy and free expression concerns.
Moreover, the South Korean case underscores a broader challenge: the need for proactive regulation that anticipates technological misuse rather than reacting to it after harm has occurred. The proposed amendments to South Korean law, if enacted, could serve as a model for other countries grappling with similar issues. However, the effectiveness of such measures will depend on international cooperation, as deepfake ads can originate from anywhere and target consumers globally. Without coordinated action, the proliferation of these scams could erode trust in digital advertising, celebrity endorsements, and even the concept of authenticity in online media.
In this context, South Korea’s warning is not merely a domestic alert but a global signal of a coming crisis in AI trust. The tools to create and distribute deepfake ads are already in the hands of bad actors; the question is whether regulators, platforms, and civil society can develop defenses before the problem reaches pandemic proportions.
What to Do If You Encounter a Deepfake Ad or Endorsement
If you suspect you have encountered an AI deepfake advertisement or endorsement, take the following steps to protect yourself and others:
- Do not engage: Avoid clicking on links, downloading files, or responding to messages in the ad. Scammers often use engagement as a way to collect data or deploy further attacks.
- Verify the source: Search for the celebrity or public figure’s official social media accounts, website, or recent news coverage to confirm whether they have endorsed the product or service. Look for verification badges or official statements.
- Report the content: Use the reporting tools provided by the platform where you encountered the ad. Most social media networks and messaging apps have options to flag fraudulent or impersonating content.
- Warn others: Share your concerns with friends, family, or online communities to prevent them from falling for the same scam. Be specific about the red flags you observed.
- Document the evidence: Take screenshots or record the ad (if possible) as evidence. This can be useful if you decide to file a complaint with consumer protection agencies or law enforcement.
- File a complaint: In South Korea, you can report deepfake ad scams to the Korea Consumer Agency or the Korea Communications Commission. Internationally, contact your local consumer protection authority or cybercrime unit.
- Check your accounts: If you entered any personal or financial information in response to the ad, monitor your accounts for unauthorized transactions and consider changing passwords or enabling two-factor authentication.
FAQ: AI Deepfakes, Celebrity Likeness, and Consumer Rights
What is an AI deepfake advertisement?
An AI deepfake advertisement is a digitally manipulated piece of content—such as a video, audio clip, or image—that uses artificial intelligence to impersonate a real person, typically a celebrity or public figure, without their consent. These ads are designed to appear authentic and are used to deceive consumers into believing the impersonated individual endorses a product, service, or campaign.
Is it illegal to create or share AI deepfake ads in South Korea?
As of August 2026, South Korea does not have a specific law that explicitly prohibits the creation or sharing of AI deepfake advertisements. However, consumer protection officials, including Lee Ji-hye, have called for such ads to be classified as fraudulent misrepresentation, which could make them illegal under existing consumer protection laws. The government is considering amendments to address this gap.
How can I tell if an endorsement is a deepfake?
Look for inconsistencies in audio or video quality, such as unnatural blinking, lip-sync errors, or distorted speech. Check whether the endorsement aligns with the celebrity’s known brand affiliations or expertise. Verify the source by searching the celebrity’s official channels for any mention of the product or service. Be cautious of ads that use urgency or pressure tactics to prompt immediate action.
What should I do if I fall for a deepfake ad scam?
If you have shared personal or financial information, monitor your accounts for unauthorized transactions and consider reporting the incident to your bank or credit card provider. You can also file a complaint with consumer protection agencies, such as the Korea Consumer Agency, or report the scam to the platform where you encountered the ad. Documenting the evidence, such as screenshots, can support your complaint.
Are social media platforms responsible for removing deepfake ads?
Social media platforms have policies against impersonation and fraudulent content, but the use of AI-generated media complicates enforcement. The Chosun Ilbo report suggests that regulators may impose obligations on platforms to detect and remove deepfake ads within a defined timeframe. However, the effectiveness of these measures will depend on the platforms’ investment in detection tools and the clarity of legal requirements.