AI's Dual Impact on Global Brand Protection: Challenges and Opportunities

In the contemporary digital marketplace, brand visibility attracts not only loyal customers but also increasingly sophisticated counterfeiters. These bad actors now operate with digital fluency and leverage the same AI tools and technologies that brands employ to expand their reach.

A notable example is Lush, the ethical beauty brand known for its packaging-free products and global presence with nearly 900 stores across 52 countries. Despite a dedicated team of IP experts and attorneys, Lush encountered a surge of counterfeit activity in 2025, with over a dozen fake storefronts appearing across Asian marketplaces and social media platforms within a single month. These counterfeiters did not merely imitate logos or brand names; they replicated entire website layouts and photographic assets to create convincing illusions of authenticity almost overnight. The minimalistic design of Lush’s product line, lacking serial numbers or barcodes, compounded enforcement difficulties, allowing counterfeiters to gain a rapid advantage.

This scenario exemplifies the broader digital landscape that brands worldwide navigate. The release of OpenAI’s GPT-4o update showcased AI’s ability to generate high-quality digital content instantly, including reimagined product visuals and near-perfect advertising creatives. While these generative AI tools enhance creative capabilities, they simultaneously lower barriers for intellectual property abuse.

Concerns about AI’s impact on creators have intensified. A widely circulated Reddit post recounted a video game artist losing his job after his employer replaced him with AI-generated 3D art created via MidJourney. The artist’s frustration stemmed from AI models being trained on artwork sourced without artists’ consent, a practice that has sparked unease across creative sectors. This issue extends beyond individual creators to a systemic brand protection crisis.

Counterfeiters now harness AI to scale their operations exponentially. Where traditional counterfeiters might generate 100 fake listings daily, AI-powered systems can produce 10,000, tailoring each to specific platforms, languages, and pricing strategies. They can also conduct A/B testing to optimize logos, product descriptions, and influencer imagery for maximum consumer deception. Additionally, counterfeiters employ AI to generate metadata that boosts fake listings’ search engine rankings above authentic brand content, complicating enforcement efforts.

Despite these escalating threats, legal frameworks remain unsettled. In the United States, copyright law lacks clear precedent on whether training AI models on scraped creative works constitutes infringement or qualifies as fair use—a doctrine that is heavily context-dependent and largely untested in this domain.

Under 17 USC Section 107, the fair use defense considers factors such as the purpose and character of the use, the nature of the copyrighted work, the amount and substantiality of the portion used, and the effect on the work’s market value. However, these criteria have yet to be definitively applied to AI training practices.

Recent guidance from the US Copyright Office marks a significant development. After extensive review, the Office affirmed that works generated entirely by machines do not qualify for copyright protection. Register of Copyrights Shira Perlmutter emphasized that granting protection to machine-determined expressive elements would undermine copyright’s constitutional goals. This clarifies that human authorship is not established merely by inputting prompts into AI systems like ChatGPT or MidJourney, delineating a clear boundary for brands and creators.

Nonetheless, the Copyright Office’s report sidestepped the contentious issue of AI training on copyrighted human works scraped from the internet without consent or compensation. This practice underpins many AI models and has prompted lawsuits from artists, authors, and media organizations alleging intellectual property misappropriation. The Office announced a forthcoming report to examine the legality of such training, licensing roles, and liability questions—an anticipated development critical for IP enforcement strategies.

Legal reform, however, lags behind AI’s rapid evolution. Generative AI introduces autonomous systems capable of judgment, analysis, and adaptation, enabling counterfeiters to train bespoke models to amplify their malicious activities.

Paradoxically, while AI facilitates more sophisticated brand impersonation, it also represents the only scalable solution to counteract these threats. IP attorneys face the arduous task of managing voluminous takedown requests, protracted litigation, and reactive enforcement that often allows infringers to profit before intervention.

Research from The 2025 State of Brand Protection reveals that 71.6% of consumers believed they were purchasing authentic items when in fact they were counterfeit—a significant increase from 50% two years prior. This indicates consumers are increasingly deceived by advanced counterfeit tactics rather than actively seeking fake goods.

As AI reshapes the counterfeiting landscape, brands and IP professionals must adapt enforcement strategies and legal approaches to safeguard intellectual property effectively in this complex, AI-driven environment.