If you’ve been listening to the show for a long time, you may already know that Kurt Vonnegut is far and away my favorite fiction author. I’d read all of his most popular novels and a few other random ones here and there, but I recently decided to make my way through all of his works sequentially, in the order they were published.
Well, was I in for a trippy surprise when I read Vonnegut’s first work, a novel called Player Piano that was published in 1952. Despite being written seven decades ago, it could not be more relevant to the AI revolution that’s accelerated dramatically in the past year.
More specifically, the book explores themes that are becoming increasingly relevant in the context of today’s advancements in machine learning. The novel is set in a dystopian future where machines have replaced most human labor, completely replacing the need for humans to do manual tasks.
The story unfolds in an America where automation, controlled by a vast and impersonal corporate and governmental machine, has created a sharply divided society: on one side, there are the scientists, engineers and managers who keep the machines running, and on the other, the vast majority who are rendered obsolete by these machines. As a listener to this podcast, you’re likely to identify as a scientist, an engineer or a manager, making you a member of the elite minority that runs the machines and so reap disproportionate rewards. In the book, the protagonist is named Dr. Paul Proteus; he’s an engineer who begins to question the ethics and consequences of this division between the elite few and everyone else, leading to a broader critique of a society that values efficiency and productivity over human connection and meaningful work.
For me personally, the most striking moment in the book – this isn’t really a spoiler, don’t worry, because it happens relatively early on — is the introduction of a backgammon machine. Unlike the machines that are already prevalent in the book for automating manual labor, this backgammon machine, which appears to be able to crush Dr. Proteus (a backgammon expert himself) at backgammon, heralding the encroachment of machines into realms of cognitive, strategic thinking previously believed to be the exclusive domain of humans. Is this starting to sound eerily relevant to our situation today? In the real world, we’ve had AI systems that could compete against world champion backgammon players since 1998, but thanks to deep learning and, more recently, transformer architectures, the cognitive capabilities of machines have accelerated wildly rapidly in recent years, for example across Go, complex negotiation-based board games like Diplomacy, and now the likes of Sora for stunning video generation that requires a sophisticated internal model of the world to be so effective. When you use ChatGPT’s built-in code interpreter with its GPT-4 model, it is not hard to imagine that most or all of the coding we do as data scientists and software engineers might be replaceable by machines in the near future. While historically all automation has led to better opportunities for human labor — less repetition, more comfort, and more interesting work — the present AI revolution will at least be different from those historical automation events because this time it’s cognitive work, not manual labor, that is being overtaken by machines.
All I can say is, there isn’t a novel that I could recommend to the particular audience that listens to this podcast at this particular point in human history. For machine learning professionals, Player Piano serves as a cautionary tale, prompting reflections on the broader societal implications of our work. It raises pertinent questions about the future of employment, the value placed on different types of work, and the potential for AI to exacerbate existing inequalities. The novel invites us to consider not just the technical challenges of creating intelligent systems, but also the ethical responsibility to guide our development in ways that enhance rather than diminish human society. As such, Vonnegut;s work resonates deeply with ongoing debates in the AI community about the need for ethical frameworks, equitable access to technology, and the preservation of human dignity in an increasingly automated world.
The SuperDataScience podcast is available on all major podcasting platforms, YouTube, and at SuperDataScience.com.