Devin

Founder of @motemath

[Devin @DevinWatson89]

@kdrag0n Who proves the provers?

[Devin @DevinWatson89]

@QualiaQuanta @AltRealityGameX @ClayInstitute @OpenAI @Anthropic Not a proof of RH. In the Lean code, RH itself is an axiom with no connection to the zeta function, and the theorem takes its conclusions as hypotheses. Two of those hypotheses can be disproved in the same framework (checked in Lean), so the result is vacuous.

[Devin @DevinWatson89]

@introsp3ctor Super interesting!

[Devin @DevinWatson89]

@physbuzz2 Yep about a month in now but at least I am making progress!

[Devin @DevinWatson89]

@krismicinski Your intuition is correct and I am building it here https://t.co/SxZTmWtKsu an AI first Lean replacement. It is as accurate if not more so than Lean and runs much faster. It can even export Lean for the more classical folks.

[Devin @DevinWatson89]

Super excited to share what I have been working on; it might not seem like much but this is just the tip of the iceberg.

#math #Lean4 #ai

[Mote Math @MoteMath]

Our first release is out!

The Lean formalization of Fermat’s Last Theorem (anthropics/fermats-last-theorem) normally means hours of compiling Mathlib first. We built it all once, so now it’s a ~17 GB download and you’re ready to go.

https://t.co/PfJepHu6LH

#Lean4 #Mathlib