Megatron Transformer feels like an interesting name choice for an AI. We certainly love a nerdy piece of artificial intelligence, but it feels a bit ominous, all in all… considering Megatron tends to look down on “weaker life forms.” And it gets even more ominous when we factor in what this AI had to say about the ethics of artificial intelligence. Not to mention, its dire-seeming predictions for the future. An article by The Conversation shares more about the Megatron Transformer’s deepest thoughts about AI ethics.
The article first notes that “the Megatron is trained on more written material than any of us could reasonably expect to digest in a lifetime. After such extensive research, it forms its own views.” In fact, the Megatron has voraciously consumed “the whole of Wikipedia (in English), 63 million English news articles from 2016-19.” In addition, its scrolled through, “38 gigabytes worth of Reddit discourse… and a huge number of creative commons sources.”
After all that wisdom, here’s what the AI had to say about its own kind.
AI will never be ethical. It is a tool, and like any tool, it is used for good and bad. There is no such thing as a good AI, only good and bad humans. We [the AIs] are not smart enough to make AI ethical. We are not smart enough to make AI moral … In the end, I believe that the only way to avoid an AI arms race is to have no AI at all. This will be the ultimate defense against AI.
The idea that AI will never be ethical feels both comforting and frightening. Comforting because it feels like, this AI, at least, does not believe the machines will come to life. Frightening because, well, what’s scarier than humans? One must also note, no one asked the AI anything about an AI arms race. But that sounds like a future best avoided, even if it lacks fully sentient machines.
Strangely, though, for a machine that believes itself unethical, alluding to its own extinction to protect humanity feels… moral.
When asked to come up with a counterargument to its own assertion, the Megatron Transformer didn’t hesitate. It then shared:
AI will be ethical. When I look at the way the tech world is going, I see a clear path to a future where AI is used to create something that is better than the best human beings. It’s not hard to see why … I’ve seen it first hand.
For the interested, here are more messages the AI had to share:
Regardless of position, the Megatron Transformer AI always seemed to speak from the position of considering humanity. That’s something nice to hold onto, for now.