Is AI discovering new sequences?

[Dr George Whale]

It seems to me that AI's capacity for working across domains should make it a valuable aid for discovering new sequences in unlikely places.

I'd be interested to know, as a new SeqFan member: have others have found AI useful in this context?

[White Board]

IMO, No! A few months ago I was building things with AI, which later made me realize that if I wanted to build something meaningful using AI, it wouldn't be meaningful, nor would it be able to protect my personal reputation. Yes, AI can be useful in creating code or finding bugs. But in general, we humans are going to conquer this unknown world only by the power of human intelligence.

--Rad

On Wed, Aug 5, 2026 at 8:00 PM Dr George Whale <propa...@gmail.com> wrote:

It seems to me that AI's capacity for working across domains should make it a valuable aid for discovering new sequences in unlikely places.

I'd be interested to know, as a new SeqFan member: have others have found AI useful in this context?

--
You received this message because you are subscribed to the Google Groups "SeqFan" group.
To unsubscribe from this group and stop receiving emails from it, send an email to seqfan+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/seqfan/6c8dce51-78bf-4a8a-aac2-010a5d541b82n%40googlegroups.com.

[DONG HAOXUAN]

I actually think AI has become surprisingly useful for this. Its capabilities have improved enormously over the past couple of years, especially on mathematical problems, such as gpt5.6sol. I don’t think it can replace human judgment or creativity, but in my experience it’s often very good at suggesting new proof ideas, alternative formulations, or search directions that I might not have considered otherwise. Used that way, it can be a genuinely valuable research tool. That’s been my own experience, at least.

To view this discussion visit https://groups.google.com/d/msgid/seqfan/CA%2Bx4cj%3DoP6zZDoFMVPMpANxS2ezw1WroghK0%3DWPk_bgMZ%3D2DfQ%40mail.gmail.com.