TLDR: A new study from UC San Diego and University of Chicago highlights the significant challenges visual artists encounter in safeguarding their creations from AI crawlers, despite the existence of protective tools. The research indicates a strong desire among artists for such tools, yet a widespread lack of technical awareness and limited efficacy of current solutions hinder their efforts.
Visual artists are struggling to protect their valuable work from the pervasive reach of artificial intelligence crawlers, according to a recent study conducted by researchers at the University of California San Diego and the University of Chicago. The findings, which will be presented at the 2025 Internet Measurement Conference in October in Madison, Wisconsin, reveal a critical gap between artists’ desire for control over their digital content and the practical means to achieve it.
The study, titled ‘How Can Visual Artists Protect Their Work from AI Crawlers? It’s Complicated,’ delves into the complexities artists face as generative AI tools increasingly rely on vast datasets scraped from the internet. The research team surveyed over 200 visual artists and meticulously reviewed more than 1,100 professional artist websites to assess their control over AI-blocking mechanisms. They also evaluated the effectiveness of various processes designed to deter AI crawlers.
One of the core conclusions drawn by the researchers is that ‘content creators now wish to control how their content is used, not simply if it is accessible.’ They further elaborate, stating, ‘While such rights are typically explicit in copyright law, they are not readily expressible, let alone enforceable in today’s Internet. Instead, a series of ad hoc controls have emerged based on repurposing existing web norms and firewall capabilities, none of which match the specificity, usability, or level of enforcement that is, in fact, desired by content creators.’
The survey of 203 visual artists revealed a strong demand for protective measures: a striking 96% expressed a desire for tools that could deter AI crawlers from harvesting their data. In response to the threat, nearly 80% of the surveyed artists reported taking proactive steps to prevent their artwork from being included in AI training datasets. These measures include two-thirds (approximately 66%) utilizing tools like Glaze, a masking tool developed by the study’s co-authors at the University of Chicago, which alters artworks to make them unreadable by AI crawlers. Additionally, 60% of artists have reduced the amount of work they share online, and 51% now only share low-resolution images of their creations.
Despite these efforts and the availability of some tools, a significant hurdle remains: technical awareness. The study found that over 60% of artists were unfamiliar with robots.txt, one of the simplest and most common tools for specifying crawling restrictions. While most large AI companies reportedly respect robots.txt directives, the researchers noted that a number of AI-powered apps and crawlers, including those from ByteDance and Meta, do not adhere to these voluntary protocols.
The study also highlighted that while active crawler blocking by reverse proxies (like Cloudflare) offers stronger protections, their deployment is currently limited. Artists need to defend against three main categories of AI crawlers: those that harvest data for large language models powering chatbots, those that enhance AI-backed assistants, and those supporting AI-backed search engines.
Also Read:
- AI Search Engine Perplexity Accused by Cloudflare of Evading Web Restrictions
- Generative AI’s Growing Influence on India’s Creative Landscape: Perspectives from Five Artists
Ultimately, the research underscores the urgent need for more effective, user-friendly, and enforceable mechanisms to empower visual artists in protecting their intellectual property in the age of generative AI.
