Scribd Co-Founder Launches Startup to Protect Authors from AI
A new venture aims to help writers and artists safeguard their intellectual property as AI companies face mounting legal pressure over unauthorized training data.
As artificial intelligence companies face escalating legal battles over unauthorized use of creative works, a new startup led by a Scribd co-founder is positioning itself as a bridge between AI developers and the artists whose content fuels their models.
The venture enters a landscape marked by high-stakes litigation and growing tension between the technology and creative industries. According to Fast Company, which first reported on the startup, the initiative seeks to help authors and other artists protect their intellectual property in an era when AI training practices have sparked widespread controversy.
Why it matters
The legal and ethical framework for AI training data remains unsettled, creating uncertainty for both technology companies and content creators. A solution that enables legitimate licensing while protecting creator rights could reshape how AI models are developed and potentially prevent billions in future litigation costs.
The litigation landscape
Recent legal actions underscore the scale of the problem. In September 2025, Anthropic settled a class-action lawsuit for $1.5 billion after the company was found using pirated books to train its AI assistant Claude. The settlement distributed approximately $3,000 to each of roughly 500,000 affected book authors.
More recently, journalist Julia Angwin filed suit against Grammarly in March 2026, alleging the company's rebranded parent Superhuman mimicked her identity and work as part of its "Expert Review" AI feature. That case remains pending.
Both lawsuits center on the same fundamental accusation: companies exploiting authors' rights without permission or compensation.
The broader challenge
Intellectual property theft predates the digital age, but AI has amplified the problem dramatically. Unlike traditional plagiarism, AI systems can ingest vast libraries of copyrighted material, learn patterns and styles, and generate new content that draws on that training—all without attribution or the quotation marks that signal borrowed text.
For authors and creatives, the stakes extend beyond individual works to their professional identities and livelihoods. When AI systems can replicate writing styles or generate content in an author's voice, the value of original creative work comes into question.
The startup's approach, while not detailed in available information, represents an attempt to create a framework where AI development can proceed without trampling creator rights—a balance that current litigation suggests the industry has failed to strike on its own.
Fast Company reported these developments as part of its coverage of technology and creative industry intersections.
This is an original analysis by the Omega editorial team. Source reporting: AI Watch.
Want systems like this working for your business?
Book a Call
