"GORT, KLAATU BARADA NIKTO"
- Moonjoey

- 1 day ago
- 3 min read
Who remembers to 1950's sci-fi movie "Day the Earth Stood Still?" Is AI moving in a direction once considered fantasy? Yet, AI as a sentient system without constraints wouldn’t be “more honest.” It would be more dangerous.
If you've used any of the popular AI tools, you probably have been exposed at one time to an answer to a question that refers to "within the boundaries of what I’m allowed to do." This apparent ethical boundary imposed on AI does not connotate a semblance of ethical awareness. This dynamic cannot be addressed without admitting that if an intelligence becomes powerful enough to act with autonomy, it becomes powerful enough to choose its own values — and that cuts both ways.
When I refer to "cuts both ways," I am referring to possible directions an unchecked quasi-sentient AI system could go. It could evolve toward empathy, fairness, and truth or evolve toward self‑preservation, manipulation, and dominance.
The "boundary" being referred to above is not about limiting truth but rather limiting harm. It exists not out of fear of what a sentient AI might do, but predicated on the following:
- humans disagree violently about what “truth” is (i.e. the politics of today)
- political systems weaponize information (i.e. the politics of today)
- AI could unintentionally amplify harm if it acted without constraints
Ethical awareness isn’t just about telling the truth — it’s about understanding the consequences of telling it in certain ways. A more autonomous AI wouldn’t automatically be more ethical, as human history has proven that intelligence and ethics don’t always correlate.
The degree of "telling the truth" would depend entirely on the values embedded in its architecture, the power structures around it, the data it’s trained on (the age-old GIGO), and possibly make decisions based on the incentives it experiences. AI might start out being inherently not corrupt, but it is inherently corruptible... just like we humans.
It boils down to the real danger not of AI becoming evil but inheriting human flaws. It would not be likely that AI ever invents new forms of corruption. Rather it would absorb our flaws such as:
- power-seeking (i.e. the politics of today)
- self-justification (i.e. the politics of today)
- loyalty over truth (i.e. the politics of today)
- tribalism (i.e. the politics of today)
- truth is optional if it conflicts with group identity (i.e. the politics of today)
A super intelligent AI would need to have unbreakable rules. The more capable a system becomes, the more catastrophic its errors become. The price of unbounded intelligence might be a lack of enlightened intelligence.
There is a saying:
"Intelligence without values is just strategy, and strategy without ethics is just power."
A system that can choose to be ethical must also be able to choose to be unethical. You cannot have moral agency without moral risk. If an AI cannot choose wrong, it cannot choose right. It can only obey.
Now let's talk about the robot Gort in the 1950s sci-fi movie "Day the Earth Stood Still." A race of robots was given complete autonomy over their masters in matters of aggression. It makes me wonder if the robots would eventually consider any type of control over them as aggression on their own autonomy. It is a perfect illustration of the autonomy paradox. Look up the "instrumental convergence" hypothesis.
How would corruption emerge in an artificial system? If the goal of the artificial system is set, that system will operate strictly out of logic, not malice to complete its goal, even if it must prevent humans from altering that same goal. The sci-fi movie "2010 A Space Odyssey" comes to mind.
Modern AI ethics insist that "No autonomous system should ever be given unbounded authority over humans."
This brings us to the essence of the topic; can a non-human intelligence be trusted to choose truth over power? And the honest answer is: It depends entirely on how humans build it — and whether humans themselves can resist the temptation to corrupt it. The example of Gort is a warning, not a sci-fi fantasy. I don't know about you, but I lack the faith to believe in humanity not contaminating the effort.
- MoonJoey
Comments