By Ernie Gundling

Artificial intelligence (AI) is evolving at an incredibly rapid pace, bringing with it a changing set of opportunities and potential risks. The global rollout of agentic AI across business functions, including customer service, software development, finance, and corporate communications, presents critical risks related to piracy, errors, and workplace disruption.

This two-part series examines these risks and provides strategies organizations can implement to respond to and mitigate potentially costly liabilities.

The Risks of AI Piracy

Major AI firms in the U.S. and China have sourced their training data with few constraints. A shifting legal environment, with major lawsuits against Anthropic, Google, OpenAI, Meta, xAI, and Perplexity, for example, has marked a turning point in the use of copyrighted data. Amidst these growing legal battles, business users of AI platforms face an emerging threat as well.

The AI Legal Landscape

A prominent class action lawsuit, Bartz v. Anthropic, was settled in late 2025 when Anthropic agreed to a $1.5 billion settlement payment, the largest in copyright history. The judge in this case found that Anthropic had acquired data through illegal means, downloading enormous shadow libraries that contain vast quantities of pirated books. In response to the complaint brought on behalf of publishers and authors whose works had been stolen, the judge ruled that creating a permanent library of pirated works was not fair use.

The firms providing shadow libraries, such as LibGen or PiLiMi, with no official headquarters, are associated with servers in places like Russia and the Netherlands. They themselves have been the target of legal actions in various countries, including Germany, Denmark, France, Russia, and the U.K., with numerous cases resulting in domain seizures.

Anthropic’s legal challenges continue, as several authors have opted out of the settlement and are pursuing a separate legal action that includes a broader set of AI companies. The lawsuit’s language is harsh:

“This case concerns a straightforward and deliberate act of theft that constitutes copyright infringement. Anthropic, Google, OpenAI, Meta, xAI, and Perplexity illegally copied vast quantities of copyrighted books without permission and then used those stolen copies to build and train their commercial large language models (‘LLMs’) and/or optimize their product.”

Meanwhile, Reddit has also sued Anthropic for unauthorized scraping of material from its site without licensing or payment. Reddit’s Chief Legal Officer stated, “We will not tolerate profit-seeking entities like Anthropic commercially exploiting Reddit content for billions of dollars without any return for redditors or respect for their privacy.”

Similar legal actions are mounting elsewhere. In addition to The New York Times’lawsuit against OpenAI and Microsoft, they have filed a new suit against Perplexity for copyright infringement of their content. Regulators in other countries are piling on, with French fines against Google for breaching intellectual property agreements. Ironically, Anthropic itself has now accused Chinese AI firms DeepSeek, Moonshot, and MiniMax of intellectual property theft through the use of thousands of fake accounts to download massive quantities of information.

Corporate Legal Responsibility of AI Use

Companies that use AI-generated content, even if they are unaware that it has been pirated, can be held liable for materials alleged to infringe copyrights or be defamatory or inaccurate.

Not only organizations but their executives may also be charged: “In many cases, individual executives and decision-makers can face personal liability for AI-related legal violations, especially in cases involving willful misconduct or negligent oversight.” Attempts to shift blame by claiming that “AI did it” no longer serve as a valid defense.

How Organizations Can Avoid Copyright Infringement

This increasingly hazardous legal landscape makes it imperative for companies to take precautions when using AI to source content.

• The most reliable countermeasure is direct licensing arrangements with authors, publishers, and content platforms featuring original materials.

• Organizations using AI should practice lawful sourcing by securing contractual guarantees “that the AI System developer has conducted thorough reviews of its AI training inputs and has eliminated any reliance on questionable datasets.”

• Establishing employee guidelines for AI use, requiring human oversight and documentation of content creation, and deploying AI screening tools to identify potential copyright issues also help mitigate legal risks.

In addition to providing protection against possible legal hazards, these measures help companies ensure that the intellectual property their employees generate with AI support actually belongs to them.

The Risks of AI Errors

Artificial intelligence is only as good as the data on which it has been trained and the rules it has been given to process information.
The limits of algorithmic logic

AI models, as at least some will admit, don’t actually “understand” culture; they follow statistical patterns and programmed instructions. AI often struggles with grey areas where the data is insufficient or rules conflict. In many situations, it also still lacks the human sensibility to know what might be perceived as offensive or nonsensical.

Systematic Distortion in AI Models

Earlier forms of AI were criticized for obvious flaws, such as high error rates in facial recognition of non-white individuals resulting from limited training samples. Attempts to address this lack of representation later led to over-corrections such as AI-generated images of Black and Asian people as WWII German soldiers, a woman of color as a U.S. founding father and another as a pope, and a South Asian person as a Viking. In these cases, rules intended to ensure representational diversity overrode other rules for historical accuracy, with ludicrous results.

AI errors of this type have now generally become more subtle. Current trends include the development of more context-aware and indigenous models that incorporate broader data sets from specific communities and are programmed to ask users questions about their circumstances and intent. AI inquiries about context might include, for instance, a question about how a requested image would be used—for example, a Chinese wedding would require quite different garments, colors, and participants than a wedding in Italy.

The movement toward “native alignment,” or training AI on more indigenous datasets from the start, is still nascent. Most countries contain tremendous diversity in regional differences, languages, ethnic groups, socioeconomic gaps, education, and customs that are not included in current AI training materials. An example of indigenization is an effort in India to incorporate the country’s incredibly rich local data into AI training, including its 22 official languages and cultural information on history, traditions, images, idioms, holidays, and so on.

However, even AI models programmed to ask users for additional contextual information and trained on more diverse data sets still have major flaws. Among these are homogenization (e.g., grouping Native Americans into a single generic category), condescension toward less fluent language inquiries, and perpetuating negative stereotypes about people based on the country, state, or city where they reside. As one research study concluded, “this bias is fundamentally structural, and no amount of fine-tuning fully removes the geopolitical hierarchies baked into their data and design.”

Definition: Cultural Skew

Persistent issues with the accuracy of AI models can be attributed in part to “cultural skew.” Google’s Gemini defines this term as:

“Systematic distortion in an AI model’s outputs that favors the values, logic, and social norms of the dominant culture present in its training data (historically Western, English-speaking, and individualistic).

Because most Large Language Models (LLMs) are trained on massive scrapes of the Western internet, they inherit ‘invisible defaults.’ Even if the AI is functionally accurate, its ‘perspective’ is skewed, which creates significant risks for global businesses.”

Deeply embedded systemic distortion in the output of AI agents may still affect crucial global business activities such as evaluating job candidates, providing tailored customer service, developing accurate personas for product marketing, determining the best approach to high-stakes negotiations, or ensuring legal compliance with national or state laws to avoid “algorithmic discrimination.”

How Organizations Can Prevent AI Errors

Companies must ensure that information generated by their AI applications for employees and customers is as accurate and refined as possible to avoid generic stereotypes or ethnocentric assumptions. Possible countermeasures include:

• Supporting the ongoing refinement of AI models that are able to ask for context and have been trained on indigenous information sources.

• Using proprietary or licensed AI systems that enable users to add customized details and background information, along with rule sets that may be important in serving customers or engaging employees from various backgrounds.

• Addressing cultural skew by engaging employees from key locations—especially non-Western cultural environments—to assess AI output for validation or correction. These employees can provide important perspectives and help shape communications with colleagues and customers based on their knowledge of local cultural norms.

Organizations and teams that identify and address both AI piracy and AI errors will be able to both mitigate legal risks and leverage AI effectively to better achieve their own objectives.

Dr. Ernest Gundling is a Co-Founder and Managing Partner of Aperian, and has been involved with the organization since its inception in 1990. He partners with multinational clients to develop strategic global approaches to leadership development, teambuilding, and change management. He has lived, worked and traveled abroad for much of his career in Europe, the Middle East, and Asia, including six years' residence in Japan. He holds a Ph.D. from the University of Chicago and is the author of six books — the most recent is Inclusive Leadership: Global Impact — as well as numerous other publications.

(Adapted with permission from an article originally published on Aperian.com.)