Artificial Intelligence? Genuine Stupidity. Scroll down to learn how content “cannot be approved.”

Synthesia AI Rejected ThiS Book

A major AI platform can't tell the difference between a word that symbolizes pride and an outdated insult.

"Queers in History: The Encyclopedia of Historic LGBT+" was written without the use of "Artificial Intelligence" (AI). However, we have been experimenting with AI tools like Copilot, Clyde, and ChatGPT to create marketing materials. Recently, we submitted our book’s standard description to Synthesia AI—billed as “The #1 AI Video Platform for Business”—in hopes of generating a rough draft of a video presentation. After we spent a significant amount of time preparing the materials, our project was rejected for “violating guidelines.”  Our work was summarily deleted.

Why were we rejected?

No reason was provided other than that the content had been “automatically” moderated. We were told to expect an appeal link via email, but when the email arrived, there was no such link. Customer service? That gets an “F” for failure.

We're guessing that the rejection stems from a blanket exclusion of the word “queer,” despite its decades-long use as both a symbol of pride and, much less frequently, a slur. The word appears in the titles of countless academic papers, articles, and books that affirm the LGBT+ community. If the use of an appropriated term is forbidden, it will continue to hold its power as an insult.
Given the sophistication of language-based AI models today, you’d think their moderation system could distinguish between respectful usage and problematic content. Sadly, their model earns an “F” as well.  

It's not the first time this has happened …

Back in 2009, we faced a similar issue when Google refused to run our ads—until Time Magazine covered the situation, prompting Google to reverse its decision. We didn’t expect to encounter resistance 16 years later, this time from Synthesia AI.


Our encyclopedia's description is publicly available on platforms like Amazon, Barnes & Noble, and Google.  Synthesia AI rewrote our description (original below) and then rejected their rewrite! Here's the dangerous text that had to be suppressed:

Queers in History: Unveiling the Unsung Heroes

Explore the transformative stories of LGBTQ+ figures who have shaped history and inspired generations.

History. It’s more than simply facts about people who lived long ago. It’s about you—how you live your life today. Armed with facts, you become the better version of yourself. As you interact with family, friends and others, the lessons of history give you the confidence and ammunition you (or someone you love) will need to deal with misinformation, prejudice and ignorance.

This book is filled with stories of LGBTQ+ figures throughout history—artists, leaders, revolutionaries, thinkers—people who defied expectations and redefined social norms.

With each page turned, you'll feel a spark of pride in their accomplishments, and your own potential. Learn about Alan Turing, the brilliant mathematician who helped break the Nazi codes and essentially shaped modern computing. Audre Lorde, whose poetry gave voice to resistance and empowerment. These aren’t just names; they're proof that LGBTQ+ people have always been part of history—creating, changing, thriving.

Identity isn't something to shrink or disguise. These stories prove that. There are a few tales of the shame of those who betrayed their own identity. Most are narratives of triumph.

People may admire knowledge, but more than that, they respect authenticity.

History has power—not just to inform, but to transform. Queers In History is more than a collection of stories about past figures; it’s a testament to the resilience and brilliance of LGBTQ+ individuals, offering a foundation for self-image and social confidence.

The problem is not AI  itself; it's the way people can misuse it

Our project was rejected by a company, "Synthesia AI", operated by humans who apparently misapplied AI by creating a list of "forbidden" words and then used that list to create a standard for moderation. If they properly used AI to examine the full context of the actual text, they might get an answer similar to what Copilot said when we asked, "Should anything in the following text be considered denigrating to other people?":

The text you provided has a strong affirming and empowering tone. It centers on the importance of historical awareness, identity, and the positive impact of LGBTQ+ figures throughout history. There’s nothing in it that appears denigrating to any group of people—in fact, it aims to counter prejudice, misinformation, and shame by highlighting authenticity, resilience, and historical inclusion. The reference to individuals who "betrayed their own identity" could invite different interpretations depending on how it's expanded in the broader context, but as written, it doesn’t express contempt—it seems to underscore how difficult and complex living authentically can be, especially under social pressure. Overall, the message is one of pride, empowerment, and transformation.
- Copilot
Scroll to Top