Are Humans Using AI to Build a Modern Tower of Babel?

This is exactly what I have been saying for the last 15 years. It isn't AI, it was the Internet, AI is just an evolution of what was already started a while ago.

I remember guys using the "optimize" function on various design programs--it would take pre-assigned parameters to adjust and find the optimal solution for you--to do their work back in the mid 1990s. Long and short of it was that the end result was (a) limited computer power was bogged down for everybody and (b) the guy using it never developed the intuition to become a good engineer.

The more things change, the more they stay the same, and those who would use AI need to understand that it's limited by the sample and biases fed into it. It does a good job of memorizing chess strategies (e.g. "Deep Blue" ) to win against grandmasters, but for extrapolating beyond the current state of the art in oncology, it's not yet there. We have pathologists and oncologists thinking these things through for a reason.

And on the light side, as a traditionalist, it is my hope and cheer to contemplate colleges bringing back the old blue books because so many students are leaning on AI to do things like do their homework and write their papers.

Also on the light side, I enjoy seeing the AI generators make up totally ficticious biographies about historical figures. For someone who has read widely and understands the world as it is, a huge portion of AI is about as subtle as a pink elephant in a delicatessen. It's not just Google Gemini making George Washington into a gorgeous African-American woman.

And on the light side, as a traditionalist, it is my hope and cheer to contemplate colleges bringing back the old blue books because so many students are leaning on AI to do things like do their homework and write their papers.

I could see that happening.

There is too much conflating happening on this topic, though. There are some dreamers of the transhumanist and other varieties that want to sort of play God and remake humanity. But most are not interested in that. And the ideas involved are full of contradictions… on all sides.

Including ours.

If it’s good for man to create a cathedral, expressing his creativity and intelligence as one bearing the image of God, why is it bad for man to create smart machines that bear man’s image? Why is it bad to be God-like in that way but not in other ways?

I don’t think this is where the problem really lies. The issue is not what we are making now but what the things we’re making are about and why we are making them. And those problems have been essentially the same since Cain got creative with his sacrifice.

The challenge with Babel was a single people with single knowledge.

"Behold they are one people, and they have all one laguage, and this is only the beginning of what they will do. And nothing that they propose to do will now be impossible for them."

There is a sense here that a single people with combined knowledge would not be stoppable in what they do. From the tower of Babel to the creation of the printing press, knowledge had been segmented to localized areas in the world and it was limited in what man could do with that knowledge. In fact knowledge was limited to man's length of days and access to that information. With the advent of the printing press, knowledge began to spread more readily around the world, but it was still limited in what someone had access to, what they could afford, and the space needed to accommodate the books.

With the advent and growth of the internet, practically all of human knowledge is now at everyone's fingertips at practically no cost. So while man only lived to 50 years in the past and could only act on the knowledge he accumulated in those 50 years. We now have access to every piece of knowledge that any scientist has ever created. Now the limitation is based on how fast we can consume all of it. Now comes AI. The ability to consume all of the knowledge that has ever been created, instantaneously, process that information, and make decisions in seconds, including the creation of new information.

The further along the closer to get to "nothing that they propose to do will now be impossible for them."

>>It does a good job of memorizing chess strategies (e.g. “Deep Blue” ) to win against grandmasters, but for extrapolating beyond the current state of the art in oncology, it’s not yet there. We have pathologists and oncologists thinking these things through for a reason.<<

While I agree that AI could possibly stifle creativity in some ways, I think it still has some great uses. One things computers are good at is keeping statistics and being able to cross reference things (i.e. storing data and repetitive tasks). If AI is used in oncology to make connections between statistics on death rates, symptoms, efficacy of medicines, etc., of course it’s not going to come up with the cure for cancer. But that kind of work is not creative in nature, and I don’t think that people lose their creative instinct by not doing hours of repetitive work.

Good oncologists would know or figure out the questions to ask an AI to find the connections that might lead to a cure, or at least a remission. Yes, a lot of what the AI (LLM) might come with is garbage “hallucinations.” That’s where the human user judgment comes in, and why we would still need the oncologists.

As to accessing so much information making us “gods,” I think we’ve seen that what we have gained from the internet has NOT made anything we propose or imagine to do possible for us. I don’t know how long God will wait to judge the earth, or how the sinfulness level of today compares with Noah’s time, but I don’t think it’s either the internet or advances in AI that are responsible for more depravity. They might make it easier to find, but as the scriptures tell us, what is unclean comes from inside the human heart.