More than 200 years ago, Mary Shelley wrote Frankenstein—one of the most enduring cautionary tales in literature. Today, as artificial intelligence becomes increasingly powerful, her story seems more relevant than ever.
We imagine intelligent machines escaping human control, developing minds of their own, and turning against their creators. It’s a frightening scenario that has fed Hollywood for decades. But there’s a fundamental flaw with this comparison: We keep worrying about Frankenstein’s monster, when we should be worrying about Victor Frankenstein himself.
The Creature Wasn’t the Original Villain
In Shelley’s novel, Victor Frankenstein is a brilliant scientist obsessed with unlocking the secret of life. After years of relentless labor, he succeeds. But the instant his creation opens its eyes, Victor is horrified. Rather than taking responsibility, he abandons it.
Left alone, the creature tries to navigate a world it doesn’t understand. It learns to speak, read, and appreciate human kindness. It longs for friendship. Yet everywhere it turns, it meets fear, violence, and rejection.
Eventually, the creature turns destructive. While its actions cannot be excused, Shelley forces us to confront an uncomfortable question: Would things have turned out differently if Frankenstein had owned his creation from day one?
That question leads us directly into the age of artificial intelligence.
The Scientists Are Still in Charge
AI is already transforming how we work, learn, communicate, and conduct research. It analyzes massive datasets, writes complex code, assists in medical diagnoses, and executes tasks that seemed impossible just years ago.
Its potential benefits are extraordinary—but so are its risks.
Much of the public debate focuses on existential existential threats: Will AI outsmart us? Will it develop autonomous goals? Could it view humanity as an obstacle?
While those questions deserve attention as autonomous systems evolve, they can easily distract us from something far more immediate:
AI does not decide how it enters society. Human beings do.
-
Data & Design: Humans select what data AI models train on and what tasks they perform.
-
Release & Safeguards: Corporate executives choose whether to ship products before understanding their limits.
-
Policy & Oversight: Governments choose whether automated systems influence hiring, healthcare, policing, or defense.
-
Incentives: Investors dictate whether the priority is responsible engineering or winning the race to market.
These are not autonomous machine choices. They are human decisions—and the people making them must be held accountable.
When Technology Becomes an Excuse
We’ve already developed a convenient habit of blaming technology for choices made by people:
-
When an automated tool discriminates, we blame the algorithm.
-
When a chatbot generates false information, we say it hallucinated.
-
When social media spreads hate, we cite the recommendation engine.
Algorithms don’t set corporate priorities, trim safety budgets, or rush unvetted software to market. People do.
Describing these outcomes as “technological failures” masks the responsibility of the executives, developers, and investors who built and deployed them.
Look at social media: platforms originally framed as tools for global connection quickly turned into engines for outrage, misinformation, and screen addiction. That didn’t happen because software developed a malicious mind of its own—it happened because of business models, design choices, and human profit incentives.
AI has the potential to magnify those errors exponentially. A bad media algorithm alters what you read; a flawed AI tied to healthcare, financial systems, or defense infrastructure carries catastrophic real-world consequences.
The Temptation to Create First, Worry Later
Victor Frankenstein wasn’t trying to destroy the world. He was driven by curiosity, ambition, and a desire to push boundaries.
These motivations aren’t inherently bad—much of human progress stems from challenging limits. Today’s AI researchers are pursuing breakthroughs that could revolutionize medicine, climate science, and engineering. We should welcome those ambitions.
But scientific brilliance is not wisdom, and technical achievement is not sound judgment.
Frankenstein was so consumed by whether he could create life that he never paused to ask what responsibilities would follow. Today, we face the exact same trap:
-
Can we build a larger model?
-
Can we make it more autonomous?
-
Can we launch before our competitors?
These are questions of capability and speed. The far more critical questions are about responsibility:
-
Should this system be trusted with this decision?
-
Do we understand its failure modes?
-
Who is legally and ethically accountable when it fails?
The Real Lesson of Shelley’s Masterpiece
Of course, the metaphor has its limits. Today’s AI models are not conscious beings with feelings, desires, or the capacity for loneliness.
Yet Shelley’s core warning remains unmatched: The primary danger of technological advancement lies in the failure of creators to anticipate and accept the consequences of their creations.
Victor’s tragedy didn’t begin when his creature opened its eyes—it began when he turned his back on it.
AI is not a force of nature. It isn’t a hurricane or an earthquake. It is a tool deliberately designed, financed, trained, and released by human choices.
We still have time to guide how these tools are built, tested, regulated, and integrated.
Mary Shelley understood a fundamental truth about human nature two centuries ago: our capacity to invent advances far faster than our willingness to take responsibility.
The greatest threat we face isn’t that AI will become Frankenstein’s monster. It’s that the people building it will act like Victor Frankenstein—brilliant enough to bring something extraordinary into the world, but unwilling to answer for what happens next.
