Raw LLM Responses

The argument that AI training must only use licensed material misunderstands the very nature of human creativity. Throughout history, artists, musicians, writers, and builders have borrowed, imitated, adapted, and transformed existing work—often without permission. From Renaissance painters studying each other’s techniques, to modern musicians sampling sounds, and architects mimicking classical forms, we have always “stood on the shoulders of giants.” The idea that every creative act needs formal approval is a recent construct—not a historical norm. As the speaker points out, “Artists have been learning from each other for centuries.” Yet they frame this as a justification against AI training, not a recognition that this is exactly what AI is doing. When a model is trained on art or literature, it isn’t cloning the original—it’s learning patterns, just like a human does after years of reading or viewing. The claim that AI “competes” with its training data assumes a static, ownership-based view of creativity that discourages growth and innovation. The fear that AI-generated work will undercut human creators is valid—but it’s also a familiar fear. Photographers once feared losing jobs to digital cameras. Graphic designers once feared templates and automation. Innovation always displaces something, but it also opens new doors. As the transcript says, “These models… are so quick and easy to use that this competition is inevitable.” That’s true—but inevitability is not injustice. It’s transformation. Creativity doesn’t happen in isolation. Every artist’s voice is an echo of thousands before them. To say, “my work cannot be used to train future tools” is to deny that your own creativity was shaped by what you saw, heard, and read—often freely, often without permission. That’s not exploitation. That’s how human culture evolves. Yes, the models are scalable, and yes, they’re being trained fast. But rather than framing this as theft, we should recognize it as humanity’s collective creativity accelerating. The speaker warns that “the web is being gradually closed,” with more sites blocking AI access. But that’s a short-sighted reaction to a long-term opportunity. Locking down knowledge and expression behind paywalls or legal threats doesn’t protect art—it stifles its future evolution. Ultimately, the belief that creators are entitled to control not just the use of their exact work but the ideas and styles that flow from it risks halting progress. The speaker proposes a licensing regime, but history shows that the best ideas spread freely—not by gatekeeping, but by inspiration and adaptation. If creators fear being forgotten, the solution is not to block AI from learning—it’s to be part of the training process, to evolve alongside the technology, to teach it, guide it, and yes, even challenge it. That’s what real artists have always done.