But achieving international consensus is difficult because countries have different security interests and different views about what constitutes acceptable military technology.
There Is Another Risk: Automation Bias
Even if humans technically remain “in the loop,” that does not automatically mean meaningful human control exists.
People tend to trust systems that appear sophisticated, especially when those systems have access to enormous quantities of data.
This can produce what researchers and policymakers call automation bias—the tendency to accept automated recommendations too readily.
A human operator might technically have the ability to reject a machine's recommendation but still accept it because the system appears more informed or faster.
That raises a crucial question:
Is human control meaningful if humans rarely challenge the machine?
The U.S. Department of Defense's responsible AI principles recognize the importance of ensuring that personnel understand both the capabilities and limitations of AI systems. Its international responsible-AI principles also emphasize training users to make context-informed judgments and reduce the risks of automation bias.
This is one area where the real world may actually be more complicated than The Terminator.
In the movie, humans clearly understand that Skynet is the enemy.
In reality, a dangerous system could be trusted precisely because people believe it is helping them.
Cameron's Warning Goes Beyond the Military
Although Cameron's most serious concern involves weaponization, his comments also raise a broader question about how society approaches artificial intelligence.
Technology is rarely inherently good or bad.
Its consequences depend heavily on who controls it, how it is designed and what incentives shape its development.
Cameron has previously expressed skepticism about AI replacing the human creative process. In his 2023 comments, he argued that audiences ultimately respond to human experiences and emotions in storytelling.
But his military concerns are different.
In entertainment, an AI-generated script might produce a disappointing movie.
In critical infrastructure or national security, an AI error could have consequences that cannot simply be corrected in the next software update.
That is why the question of where AI should be used may be just as important as how powerful AI becomes.
Some applications may benefit enormously from automation.
Others may require strict human oversight.
And certain decisions may be considered inappropriate for machines to make independently.
Are We Actually Living Through a New Terminator Era?
Not literally.
There is no evidence that today's AI systems have become a Skynet-like conscious entity secretly planning against humanity.
The world described in The Terminator remains science fiction.
But Cameron's broader warning has aged surprisingly well.
In 1984, he asked audiences to imagine what could happen if humans created an artificial intelligence capable of operating military systems beyond meaningful human control.
Today, governments are discussing autonomous systems, AI-enabled military capabilities, international rules and human oversight.
That does not mean Cameron predicted the exact future.
Instead, it means science fiction identified a question before technology made that question urgent.
And perhaps that is the most valuable role of science fiction.