The only counterargument I would like to present is that the concept of "right AI alignment" that everyone discusses is, in a way, a type of belief. It isn't grounded in logic or concrete data, but rather rests on convictions, even if those convictions are injected by humans.
Fair point. I think I’m trying to get to the existential difference. But if it’s programmed into the AI, is that (in a sense) existential for the machine?
It may not be essential for the machine to exist, although it is a very novel area, and I’m not sure an answer is possible. Perhaps the safeguard is not only for us but also for the machine itself (preventing self-destruction). A complex system can be predictable but it is nearly never deterministic 🤷.
This is quite persuasive. Though, judging by human behaviour, belief, disbelief, doubt, etc. are cognitive burdens that humans long to be rid of; that is, I’d say, part of what makes AI tools so seductive.
Yes, though seduction implies their use is negative. I wouldn’t accept that. I’d say it’s complicated. Where much of the discourse (pro and con) disappoint or frustrate me is in the binary nature of it.
Indeed, Joel. Nuance is always welcome. I’m merely trying to fight a human tendency to gloss over or ignore inconvenient facts. Take the automobile. It transforms the lives of individuals, widens the scope of their lives, fuels their imaginations — AND majorly contributes to global warming. Many commentators will focus on one and ignore the other. We should be able to walk AND talk.
That's a great essay and excellent perspective on the topic. One major obstacle to convincing people to be more skeptical of A.I. is the extent to which many people believe what they see on the internet regardless of evidence to the contrary or common sense.
Have you seen the recent news story about people showing their Costco ID as a valid form of identification at TSA checkpoints? Something they read on the internet convinced them that because it has a star on it, it is the equivalent of a Real ID secured driver's license. Good luck convincing those folks that something that can sound as intelligent as AI would be giving them bad info.
I have found local friends who are deep thinkers, wise, and bold. When I get a "new" idea I tell them, "Here is what I am thinking. Can you talk me out of it?"
Great post! Thank you for sharing.
The only counterargument I would like to present is that the concept of "right AI alignment" that everyone discusses is, in a way, a type of belief. It isn't grounded in logic or concrete data, but rather rests on convictions, even if those convictions are injected by humans.
Fair point. I think I’m trying to get to the existential difference. But if it’s programmed into the AI, is that (in a sense) existential for the machine?
It may not be essential for the machine to exist, although it is a very novel area, and I’m not sure an answer is possible. Perhaps the safeguard is not only for us but also for the machine itself (preventing self-destruction). A complex system can be predictable but it is nearly never deterministic 🤷.
This is quite persuasive. Though, judging by human behaviour, belief, disbelief, doubt, etc. are cognitive burdens that humans long to be rid of; that is, I’d say, part of what makes AI tools so seductive.
Yes, though seduction implies their use is negative. I wouldn’t accept that. I’d say it’s complicated. Where much of the discourse (pro and con) disappoint or frustrate me is in the binary nature of it.
Indeed, Joel. Nuance is always welcome. I’m merely trying to fight a human tendency to gloss over or ignore inconvenient facts. Take the automobile. It transforms the lives of individuals, widens the scope of their lives, fuels their imaginations — AND majorly contributes to global warming. Many commentators will focus on one and ignore the other. We should be able to walk AND talk.
I like this! Thanks
My pleasure!
Khalil - Would I be correct in saying the convictions it rests on are not its own?
I think the answer to that is yes.
Indeed.
This is an insightful challenge. You've given me a lot to cogitate on.
You’re welcome.
That's a great essay and excellent perspective on the topic. One major obstacle to convincing people to be more skeptical of A.I. is the extent to which many people believe what they see on the internet regardless of evidence to the contrary or common sense.
Have you seen the recent news story about people showing their Costco ID as a valid form of identification at TSA checkpoints? Something they read on the internet convinced them that because it has a star on it, it is the equivalent of a Real ID secured driver's license. Good luck convincing those folks that something that can sound as intelligent as AI would be giving them bad info.
LOL. I don’t know how much we can plan around stupid people.
I have found local friends who are deep thinkers, wise, and bold. When I get a "new" idea I tell them, "Here is what I am thinking. Can you talk me out of it?"
That’s great! And you should keep those relationships going as long as possible. What a treasure.
I think you're exactly right, Joel. And in a weird synchronicity (I've been listening to The Police lately :), Spencer Klavan wrote about the same thing in his Substack post today: https://open.substack.com/pub/thenewjerusalem/p/letter-256-this-is-not-a-post?utm_source=share&utm_medium=android&r=2mpv4
That’s cool. I think we’re all turning our wheels on some of the same ground these days. Thanks for sharing the link.
Is this a book excerpt?
No. Just something I’ve been noodling on.
This post makes me wonder if anyone has ever thought to ask AI if it should be put in charge of running the world, And why or why not.
I’m sure people have thought it. Terrible idea.