This is a summary of the book titled “Creative Machines: AI, Art & Us” written by Maya Ackerman and published by Wiley, 2025. This book explores how artificial intelligence can deepen, rather than diminish, human creativity. Long before AI became a mainstream topic, Ackerman—a computer science professor and lifelong musician—was already investigating how machines might inspire people to become better creators. Drawing on both academic research and her experience as the CEO of a generative AI company, she argues that the most meaningful creative technologies are not those that replace human effort, but those that work alongside people as supportive collaborators. Throughout the book, she introduces the idea of “humble creative machines,” tools designed to empower users, foster growth, and keep humans at the center of the creative process.

Ackerman grounds her argument in a clear definition of creativity. Creativity, she explains, is best understood through its products—objects or experiences that are both novel and valuable. Novelty excludes mere variations on what already exists, while value implies intention and purpose. A creative work may be valuable because it is useful, or because it provides enjoyment, meaning, or emotional resonance, as art and music do. As Ackerman puts it, creativity is not simply about being first, but about making something that matters. Importantly, this definition does not restrict creativity to humans. Many animals and even plants exhibit creative behaviors, such as the bowerbird’s carefully constructed displays made from colorful objects to attract mates. By focusing on novelty and value rather than on the identity of the creator, Ackerman opens the door to recognizing how machines, too, can participate in creative processes.

She cautions, however, that machine creativity does not need to look like human creativity. One common way of evaluating creative machines is the Discrimination Test, which asks whether people can distinguish machine-generated works from those made by humans. Ackerman recounts a 1997 experiment in which a computer program composed a piece of music in the style of Bach, and most listeners—including trained musicians—mistook it for the original composer’s work. While this test is useful because it forces audiences to judge the artifact rather than the creator, Ackerman notes its limitations. Comparing machines to humans risks overlooking the fact that machines can be creative in their own right. For ethical and intellectual property reasons, she argues that AI should not imitate living artists. Instead, creative machines should be treated as entirely new entities—tools that expand human potential rather than mimic it.

Creativity depends on the balance between divergent and convergent thinking. Divergent thinking involves curiosity, exploration, and the willingness to generate many ideas, including strange or impractical ones. It is essential for innovation and often benefits from stepping away from problems and returning to them with fresh perspectives. Convergent thinking, by contrast, is the process of refining ideas, making decisions, and turning possibilities into finished work. Ackerman emphasizes that creators must move back and forth between these modes and resist judging ideas too early. Great creative output, she explains, often comes from producing a large volume of work, much of which will never be seen. Picasso’s thousands of artworks serve as an example of how brilliance emerges from persistence rather than perfection. This same principle applies to machines: if AI systems are limited to producing only the most likely or “correct” answers, they lose their ability to help humans discover something genuinely new.

Ackerman encourages readers to think of creative machines as collaborators rather than replacements. She compares working with AI to collaborating with a talented friend. The value lies not in having the work done for you, but in being supported, challenged, and inspired as you develop your own ideas. Many creative AI tools, she observes, focus on showing off impressive results without inviting users into the process. The most impactful tools, by contrast, enable co-creativity, where humans and machines interact dynamically to shape a final outcome. Used in this way, AI becomes a partner with unique strengths—always available, endlessly patient, and capable of offering fresh perspectives.

Her own journey illustrates this approach. After years of frustration with songwriting, Ackerman collaborated with an AI system called ALYSIA, which generated melodies based on lyrics. Rather than replacing her creativity, the tool sparked new ideas and helped her break out of repetitive patterns. Over time, working with AI gave her the confidence to compose independently. This experience informed her advocacy for “humble creative machines,” a concept she helped formalize with other researchers. These tools are flexible, allowing users to decide how much assistance they want; supportive, acting as coaches rather than crutches; genuinely creative, offering novel suggestions; and easy to use, fitting seamlessly into existing workflows. When users approach AI as a responsive tool rather than a performer meant to impress, the quality of the work improves and ownership remains with the human creator.

Ackerman points to conversational AI tools such as ChatGPT as practical examples of this philosophy in action. These systems adapt to user input, encourage iteration, and leave room for human judgment and editing. However, she stresses that whether an AI functions as a humble creative machine depends as much on the user’s mindset as on the technology itself. When people allow AI to do all the work, they miss valuable learning opportunities. When they engage with it thoughtfully—using feedback, critique, and iteration—they develop stronger creative skills over time.

The book also addresses the business and design implications of creative AI. Tools should not aim to retain users by fostering dependence or addiction. Instead, they should support long-term creative growth, even as users become more skilled. Ackerman describes how her own company learned this lesson when an early version of their songwriting app failed to gain traction because it tried to do too much. By refocusing on a simpler tool that offered suggestions rather than complete solutions, they attracted more users who returned consistently. Sustainable success, she argues, comes from empowering people rather than dazzling them once.

Ackerman does not shy away from the ethical challenges of creative machines, particularly the issue of bias. Because AI systems are trained on human-generated data, they inevitably reflect human stereotypes and blind spots. Studies have shown that image generators often reproduce racial, gender, and cultural biases, misrepresenting marginalized groups and oversimplifying non-Western cultures. These failures do not reflect machine intent, but rather the limitations of the data on which they are trained. Addressing this problem requires more diverse voices in AI development and research, as well as deliberate efforts to improve datasets and representations. Projects like narrative-generating systems designed to preserve Indigenous stories demonstrate how creative machines can also be used to counter erasure and support cultural understanding.

Ackerman argues that creative machines hold up a mirror to humanity. By exposing our biases, assumptions, and unexamined patterns, AI offers an opportunity for deeper self-reflection. Drawing on ideas from psychology, she suggests that confronting what machines reveal about us—both the negative and the positive—can lead to meaningful change. Technology alone cannot fix human problems, but it can help make them visible. When used thoughtfully, creative machines can support not only better art and innovation, but also greater awareness, responsibility, and human flourishing.