What comes to mind when we mention 007? Spies, beautiful women, cool gadgets, intense battles, and martinis that are shaken, not stirred. As a devoted fan of the entire 007 series, these are the elements that I have largely observed.

That is, until I recently played the latest 007 game, 007: First Light. When the villain was revealed, I realized the plot struck a chord within my own industry. Spoiler alert: the antagonist in this game is an AI large model company.

I write about AI companies daily, but I never thought it would take James Bond to take them down. Generally, if their funding dries up, they'll collapse on their own.

Yet, looking back at the dozens of 007 films over the decades, I realized that beneath the spy’s romanticized adventures, there has always been a clear theme: the evil of technology.

In fact, it could be said that every time a new technology emerges, it inevitably assumes the role of the villain in 007. I want to explore this apprehension, as well as the logical mechanisms and collective psychology behind it.

If we focus solely on 007's cool appearance and romantic stories, we can easily overlook that this is actually a standard tech-pessimistic narrative.

Ian Fleming, the original author of the 007 series, personally participated in intelligence work during World War II. He witnessed the terrifying power of the V-2 rocket and experienced the nuclear arms race between Germany and the United States. This instilled in him a deep-seated worry and fear of unknown technologies, a theme he infused into almost every 007 story, which was then adapted by the films and games.

007's first layer of imagination regarding the evil of technology is the atomic nightmare that enveloped the globe during the Cold War. In the original novel Moonraker, the antagonist is a British nuclear weapons expert with malicious intent, planning to launch nuclear missiles at London. In the film Goldfinger, the villain prepares to attack Fort Knox's gold reserves with nuclear weapons, thereby collapsing the global financial system. These fears of nuclear weapons are closely tied to the nuclear arms race during the Cold War in the 1950s and 1960s. At that time, major countries worldwide, including Britain, promoted content on how to face nuclear threats and respond to missile attacks. This tense atmosphere permeated social culture, fostering a deep-seated fear of nuclear technology.

As the shadows of the Cold War dissipated, the sense of nuclear fear gradually diminished. However, the imagination of technology's malevolence had already taken root deep within popular culture, as represented by 007. In the 1970s and 1980s, gene panic replaced nuclear fear. Topics like cloning, gene editing, and human enhancement emerged, reawakening horrific memories of Frankenstein. In A View to a Kill, the antagonist is a victim of Nazi gene experiments, granting him ultra-high intelligence and a deranged personality. He plans to destroy Silicon Valley by causing a massive earthquake, allowing his company to monopolize the chip market. These bizarre conspiracies seem to stem from uncontrolled genetic modifications. The ultimate conjecture about the evil of technology shifted from whether humans would blow up the Earth to whether they would cultivate monsters.

By the 1990s, things changed again. The rise of the internet and the wave of globalized economics swept in. The villains in 007 gradually shed their Nazi identities. After all, it had been decades since World War II, and it seemed unrealistic for Nazis to remain immortal. The new antagonists were mostly capitalists who wielded internet influence and immense wealth, always harboring malicious intentions toward human civilization. For example, in Tomorrow Never Dies, there's a famous line: "Words are the new weapons, satellites are the new cannons." The villain in this film is a media tycoon who uses satellites to tamper with GPS signals, create fake news, and provoke nations into war, all for higher ratings and media monopoly. It seems that even in media, everyone suffers from traffic anxiety.

Post-war fears centered on atomic energy and nuclear weapons, the 1970s on gene editing and human enhancement, and the 1990s on the internet and globalized media. Over decades, 007 repeatedly judged and questioned these fruits of human technology.

Has the situation improved in the 21st century? Not really, because AI has arrived.

How does 007's tech horror story unfold in the age of AI? 007: First Light offers an eye-opening answer.

In this work, we see technological marvels everywhere, connected by quantum computers, drones, humanoid robots, and massive servers. Beneath these cold, enormous technological images, the story's suspicious perspective on the evil of technology is even more astonishing.

What's going on? The globally renowned new energy tech giant, the Weber family, invests heavily in developing an AI large model called THEIA. I don't know if any real AI company also dabbles in new energy, but this one does. The Webers pitch their large model to MI6, claiming it can predict potential threats worldwide, such as terrorist attacks or illegal transactions. Agents only need to act according to the model's predictions, ensuring infallible success.

But in reality, how accurate can AI be? The idiots at MI6 apparently don't know about something called "large model hallucinations." THEIA often makes inaccurate predictions, inventing many so-called airstrikes or murders. Generally, AI companies choose to retrain their models and release more accurate versions, but Weber Tech doesn't think that way. Perhaps because they sold the British government a locally deployed model and the government, trying to lower costs during procurement, refused subsequent model upgrade services.

In any case, Weber Tech chooses to manually "fix" AI hallucinations in the physical world. For example, if THEIA predicts a terrorist attack somewhere that never happens, old Weber sends his son to carry one out. If THEIA predicts someone will be murdered, but they're alive and well on the scheduled day, old Weber has his son kill them. It's a comedic display of father-son "filial piety" for the sake of AI accuracy.

The Weber family's ultimate ambition is to have their infallible large model accepted worldwide, deployed in global energy hubs, military bases, and civilian infrastructure. They believe humans are untrustworthy and that all power should eventually be handed over to AI. To achieve this ultimate ideal, they're willing to cause bloodshed for their unreliable AI.

It's hard to imagine how people with such twisted logic could establish themselves in the competitive tech industry. But this story indeed showcases a fundamental suspicion many have about AI today: What if the information AI tells me is wrong? What if the people behind AI deliberately feed me false answers? As AI's capabilities grow, such suspicions and conspiracy theories become heavier, turning into an unsolvable deadlock.

This points to 007's fundamental attitude toward technology: Technology may be innocent, but one can never be sure about the malice lurking in the hearts of those who wield it.

An AI company resorting to terrorism to maintain model accuracy seems like an absurd setup even for a sci-fi film. But it demonstrates a logical deduction commonly used in the 007 series: I can't see the people wielding technology, so I don't trust their goodness.

In fact, 007 doesn't oppose technology. The series loves to gain attention through Q Branch's various bizarre inventions. But what 007 trusts are specific, minor, and harmless cool techniques, not the technological concentration of power that can affect public safety and dominate collective discourse. In other words, 007 is a lover of low technology and a skeptic of high technology.

007's creed is that when technology's capabilities are strong enough, the specific individuals holding it are absolutely untrustworthy.

These individuals can be entrepreneurs, capitalists, government technical researchers, or bureaucrats with the power to use technology. Their grasp of technology often goes through three stages: In the first stage, they witness technology's terror, beginning to believe that technology equals power. In the second stage, they view technological power as part of their own, starting to believe they are superior and above others. In the third stage, moral rationality and humanism completely collapse, and they believe they can overturn order and reshape the world. They may even feel obligated to do so, leading to an extreme anti-human stance.

All humans who grasp the secrets of technology seem inevitably to enter this cycle.

This is just an extreme deduction from 007 and many similar works. But this prisoner's dilemma-like game repeatedly brings about tech panic in modern society, turning into box office success and endless online debates.

Should we be wary of technology and fear AI? To some extent, yes. AI is increasingly showing signs of concentration of power. Anthropic is as eerie as a cult, and Elon Musk's wild statements are suffocating. A decade ago, many called for banning AI in military applications, but now it's blatantly being used for warfare. Many voices warn against AI fraud, but it's increasing like wildfire.

We have reason to worry that AI could become the supreme power for some madmen, fearing a scenario where "everyone thinks it's fine, and then something goes wrong."

But on the other hand, 007's tech panic rhetoric has never succeeded. Many things drag down the world's average, but technology isn't one of them. From nuclear technology to gene technology, and from the internet to AI and new energy, the world keeps improving under technology's rapid advancement. Often, it's technology alone that fiercely combats many evil forces.

Another issue is that many times, being wary of technology's evil isn't about "it's just me," but rather "it's not just me." When our vigilance exceeds reasonable limits, it doesn't lead to effective regulation of technology but instead sends traffic and money to those who repeatedly remind us of technology's malevolence.

Indeed, we should be wary of technology's evil, but we should also guard against excessive anxiety. When 007 stories can only weave AI company villain narratives from such bizarre angles, it perhaps indicates that everything is still optimistic to some extent. At least the screenwriters lack real-life experiences of AI destroying the world.

Even if we provide endless worries, the tech tree will keep growing. We should be cautious about this tree strangling the house but might as well enjoy some shade under it.