How AI Deepfakes Are Expanding Trademark Strategy

AI has changed the economics of impersonation. A convincing fake voice line, endorsement clip, or lookalike performance no longer requires a studio, a soundalike, or a large production budget. In many cases, it now takes a public sample, a consumer-facing platform, and a few minutes.

That shift matters for trademark strategy. For celebrities, creators, and public-facing brands, the most valuable commercial signals are not always limited to a name or logo. They can also include a catchphrase, a recurring intro, a sonic cue, a stage image, a signature pose, or a voice clip that tells the public: this is official. _ Disclaimer: This article is for general information only, not legal advice. Trademark and publicity rules vary by country, industry, and the facts of each case._

Why AI Is Turning Identity Into a Trademark Question

Trademark law does not protect identity in the abstract. It protects source indicators: signs that help consumers identify who is behind goods or services, or whether a use is officially sponsored or endorsed.

That distinction is critical.

A face, voice, or phrase is not automatically a trademark just because the public recognizes it. It becomes trademark-relevant when it is used in a commercial way that signals origin, endorsement, sponsorship, or affiliation.

That is why ai and trademarks has become a more urgent issue. AI makes identity signals easier to imitate, easier to distribute, and easier to weaponize in ads, scams, fake endorsements, and misleading branded content.

Taylor Swift's 2026 Filings Show the Portfolio Is Changing

Public reporting on April 27, 2026 said Taylor Swift's company, TAS Rights Management, had filed three new trademark applications on April 24, 2026: two sound marks tied to spoken voice clips and one image tied to a recognizable stage-performance visual.

Why does that matter?

Because it shows that one of the most commercially sophisticated artists in the world is not relying only on copyright in songs, recordings, or visuals. The reported strategy appears to focus on recognizable identity signals that could help distinguish official Swift-related content from AI imitations, fake endorsements, or misleading commercial uses.

That is the real business takeaway. In the AI era, the most valuable brand asset may not be the words alone. It may be the association between the words, the voice, the image, and the source.

Taylor Swift is also an especially strong example because the risk is not hypothetical. Her likeness has already been used in AI deepfakes, including explicit fake images and false political or commercial-style associations. When a public figure starts filing around voice and stage image, it suggests the trademark portfolio is expanding to cover the signals people rely on to decide what is real.

McConaughey's Catchphrase Strategy Is Really About Control

Public reporting on January 15, 2026 said Matthew McConaughey had secured approval for multiple USPTO trademark applications covering parts of his public persona, including an audio clip of "All right, all right, all right" and visual clips associated with his image.

This example is useful because it makes a subtle point much clearer than a legal definition can.

The value of that phrase is not just the words. It is the delivery, the years of repetition, the commercial recognition, and the public association with one person. That is how a catchphrase can move from quotation to brand signal.

McConaughey's reported strategy also highlights an idea many businesses miss: this is not necessarily anti-AI. It is pro-control.

He has also been publicly linked to authorized AI voice use, which reinforces the stronger message for creators and brands: the goal is often not to stop every synthetic use. The goal is to create a clearer line between approved licensing and unauthorized imitation.

Trademark Can Help, but It Does Not Let You "Trademark Yourself"

This is where the legal framing matters most.

Under Section 2(c) of the Lanham Act, the USPTO can refuse registration of a mark that consists of or comprises the name, portrait, or signature identifying a particular living individual without written consent. That rule matters because it shows the trademark system already recognizes that personal identity can have serious commercial value.

At the same time, that does not mean every recognizable trait is automatically registrable.

For a celebrity, creator, or public-facing brand, the more accurate question is:

• Does this element function as a trademark or service mark?
• Do consumers connect it to a commercial source?
• Is it being used consistently with specific goods or services?
• Is it distinctive enough to do more than decorate or describe?

This is especially important for sound-based strategy. U.S. trademark law is broad enough to cover non-traditional signs, and the USPTO already examines sound marks. But a voice-related filing is strongest when it involves a specific, repeated, source-identifying audio signal, not just an ordinary sentence said by a famous person once.

In practical terms, stronger candidates might include:

• a recurring podcast intro phrase
• a distinctive spoken line used at the start of official videos
• a sonic logo
• a catchphrase used consistently on merchandise and promotions
• a short audio clip tied to one commercial source

The Deeper Risk Is False Endorsement and Consumer Confusion

One reason trademarks and ai now intersect so often is that AI misuse is rarely only about copying. It is often about confusing the public.

That confusion can show up in different ways:

• fake ads that look officially approved
• cloned voices used to sell products
• deepfake videos suggesting endorsement
• scam promotions using a famous face or catchphrase
• synthetic clips that imply affiliation with a brand, campaign, or platform

The FTC has treated impersonation fraud as a serious consumer-protection problem. In an April 4, 2025 FTC press release, the agency said impersonation scams were among its top reported fraud categories, with $2.95 billion in losses in 2024, and highlighted enforcement under its Government and Business Impersonation Rule.

That matters because the harm is wider than celebrity image management. Unauthorized AI identity use can also create:

• reputational damage
• fraud losses
• fake endorsements
• platform confusion
• licensing disputes
• pressure to respond publicly before facts are clear

Tom Hanks' public warning about AI ads using his likeness, Scarlett Johansson's dispute over a voice that sounded eerily similar to hers, and the rapid spread of AI-generated songs imitating Drake and the Weeknd all point to the same commercial problem: audiences are being asked to judge authenticity in an environment where identity is easier to simulate than ever.

Trademark Is Only One Layer of the Protection Strategy

Trademark law can be useful, but it is not the whole answer.

For many creators and brands, the broader protection stack now includes:

• trademark rights for names, slogans, sounds, and other source indicators
• right of publicity or personality-rights claims where identity is used commercially without permission
• contract clauses covering voice, likeness, scans, and digital replicas
• endorsement and licensing controls
• platform takedown processes
• domain and handle enforcement
• consumer-protection complaints where scams or fake endorsements are involved

That broader view matters because state and federal rules are still evolving. Tennessee's ELVIS Act took effect on July 1, 2024 and expressly added voice protection into that state's framework. California signed AB 2602 and AB 1836 in September 2024, restricting certain unauthorized AI digital replicas of performers. At the federal level, the NO FAKES Act of 2025 remains proposed legislation; as of July 2, 2026, Congress.gov shows it in introduced status in the Senate Judiciary Committee.

In other words, trademark strategy should not be built in isolation. It should be coordinated with publicity rights, contracts, licensing, platform enforcement, and internal approval workflows.

If you are also assessing authorship, training-data, or output-protection questions, iGERENT's guide on AI and Copyright: Who Owns AI Generated Content and What's Protected covers that separate but related issue.

What Public-Facing Brands Should Review Now

The practical response is not to file everything. It is to identify which identity signals are actually functioning as brand assets.

Start by auditing whether the public recognizes any of the following as part of the commercial identity:

• professional name or stage name
• nickname
• logo or signature
• catchphrases
• voice clips or intro lines
• sonic logos
• podcast, show, newsletter, or tour names
• recurring poses, gestures, or stage imagery
• character names or mascots
• merchandise branding
• social handles or campaign slogans when they function as source identifiers

Then test each one against real trademark logic:

• Is it distinctive?
• Is it used consistently?
• Is it tied to specific goods or services?
• Do consumers treat it as official?
• Is there evidence of recognition, licensing, promotion, or sales?

From there, the smarter next steps usually include:

• registering core names and house brands first
• considering non-traditional marks only where the facts support them
• documenting use and recognition
• tightening AI clauses in contracts
• defining approved AI partners and uses in writing
• monitoring for deepfakes, fake ads, and confusing endorsements
• preparing fast takedown templates before the first crisis

The Expert Takeaway

AI has not changed the basic rule that trademarks protect commercial source indicators. What it has changed is which signals may matter most.

For celebrities, creators, and public-facing brands, the question is no longer only whether the name and logo are protected. It is whether the voice, phrase, image, gesture, and recurring persona cues that audiences trust are being reviewed as part of the brand portfolio too.

Taylor Swift's reported filings and McConaughey's catchphrase strategy point in the same direction: when identity becomes easier to imitate, identity-based brand assets become more important to define, document, and control.

Protect the Identity Signals Your Audience Already Trusts

If your brand depends on a recognizable name, catchphrase, visual style, or public-facing persona, iGERENT's International Trademark Registration Service can help you review which elements are worth protecting and how they fit into a broader trademark strategy.

Trusted by 12,000+ businesses to register and manage trademarks in 180+ countries, iGERENT works through one dedicated specialist coordinating local counsel, with fixed quotes and clear timelines.

Prefer to ask a couple of questions first? Contact iGERENT for a free, no-obligation quote.