AI Taking away the enjoyment of finding things out

Richard Feynman:

Source: AI is proving things no one understands

The goal of Mathematicians is as well to build theory around it. A great example he uses is the Alan Turing example that he created a small little lemma to basically disprove Hilberts problem. If he did not create the Turing machine then we would not have had the computers today.

Nobody should think that proving a theory as the first one is going to be how we are going to do mathematics. Or basically, any future science for that matter. The goal should be to understand the world around us. This could also mean using a proof or any theory and then let an LLM help you understand the proof.

However, what it will boil down to is that we will all be individuals talking only to our LLM's and not have peer discussions anymore. But that does not need to be lost. We can create educational hubs, is what is the project tinybrains is all about. At least up till now LLM's cannot explain yet that good on why they did certain things or how do things work with new lemmas and tools that can be used for follow up questions.

They say that okey for now you can have someone else write a great paper that helps us better understand the problem, not that we solved the problem. This is the reason said the days of heroes has ended.

But in the end you

We will want to understand the problems AI solved! That is the next frontier of mathematics. Until the moment comes that AI can also explain to mathematicians how they came up with the proof, new methods they used, to make it human understandable. But even then, you need lots of back ground knowledge to understand these higher forms of mathematics. You basically need to study the mathematics from the basics and then work your way upwords i.e calculus - > lineara algebra -> anlysis -> group theory -> topology etcetera.

We will find ourselves at a spot where we have incredible tool that proofs amazing things or solves amazng problems but that us humans cannot use since it does not explain how things work. Making a space ship go to the moon purely on AI, does not teach us the laws of physics.

We need to do things for intrinsic reasons! They see how other people see you! but it should be for yourself ! you do it because you enjoy it yourself.

Mathematicians should more work in teams of understanding a problem instead of the heroes of the golden era of mathematics.

AI should be used to boost human understanding and not make us understand less of it. We could use AI to teach us, provide new problems for us to solve, then solve them and then explain them to us. If you then don't understand some terminology or definitions then you go one step down in the explanation until you are at the base ground from which you understand.

Again note the big problem that we are going to have if we put AI systems in control of systems that are critical but that we don't understand as humans.

And on you're analogy I think its like having someone annoying watching a movie and then telling you oh this guy dies at the end. You don't even want to continue cos the life has been sucked out of the experience

The Paradox of Efficiency

Disease treatments without understanding people would be happy with it.

We cheat the

There's so much more being lost. I'm a software developer so I think often my criticisms fall flat because people think I'm just worried about losing my job, but what I'm most worried about is losing the societal glue of having to rely on each other's skillsets since way back in our evolution, both for productivity and for passing that knowledge on. I don't want to live in a world where machines make all the breakthroughs - I want to benefit from the skills of humans who can do something I can't. I want to be able to repay that with my own knowledge. I want to be able to have art on the wall that was drawn by someone with a talent and a passion I lack.

2 years ago I felt the same about computer science. I love the journey to code, find creative solutions and combine modular parts of a system, into a whole product which is working. The last 3 months I just stopped mentally working as computer scientist anymore. I lost my whole fun and enjoyment in it and writing this makes me sad, because I have no clue how to feed my family.

There is a genuine feeling, of self-disappointment and resentment for humanity as a whole, when you realize you dont even need to think any more because we chose to make a machine that takes away all the learning and knowledge steps we had to do ourselves, to evolve as a species. All the little things that actually make you grow in a field. All the misteps and the joy of eventually overcoming them to earn our places

Doubt all you want, that doesn't mean your doubt is founded in reality. A train still runs even if no one understands how it works. Math is still correct even if humans don't understand it. Cures to diseases that AI may find will still work if we don't understand them fully just like anesthesia. Results are the only thing that matter, the rest like human understanding is egotistical.

it is not just maths but all forms of human endeavour where the journey, the process, the reflection, the struggle and creativity are what give meaning - that is where all the value in both art and science lie and maths is a far more creative pursuit than people may realise. Its like seeing som AI art that looks half-decent -raer as they are, but even then, there is no human experience, no process, no connection with the wonder that a human made it - art is just reduced to a background, to knick-knacks and nothing has meaning.

But I think it’s important to focus on the big picture. If OpenAI wanted to destroy the mathematical community, this would be a great way to go about it. Open problems are a resource that the mathematical community developed over decades or even centuries. The value they have is the value we have given them. They provide structure, a yardstick for progress and long-term goals. The advent of AI was always going to be a seismic shock, but this huge dump of papers is a tsunami that we have no time to prepare for. It’s clear that OpenAI will happily wash away our community’s structures to further their financial interests.

There is a long list of mathematical structures and norms that OpenAI seems to be intent on damaging. By refusing to name the authors of their papers, they imply that mathematics is no longer a human endeavour. They don’t submit their results to journals, which both implies that the norms of the mathematical community are defunct and makes it impossible for the community to digest their results. OpenAI haven’t shared fundamental information about how the problems are selected or the failure rates of their attempts. This information should be a necessary requirement for any serious scientific discussion of their tools. And of course there is the basic financial injustice that their hugely valuable models learned to reason by copying the very reasoning that our community made available for free through the Open Access movement.

I hope the mathematical community can adjust our practices and norms quickly enough to survive the shock of these developments, but I am fearful.