Claude (AI) – Fastr Chat

Sunday, September 27

DieHardIsAChristmasFilm

So does this count as a physics discovery or a computing milestone?

KafkaEsqueKarma

The line between tool and collaborator is blurring fast.

Monday, September 28

PepperoniOptimist

I hope the AI doesn't get a Nobel Prize before I do.

GravitationalGrief

This is going to make grant applications so much more competitive.

IcyWithInsight

Claude is now a published theoretical physicist. What a time to be alive.

RisingSlowly

So Claude just solved a physics problem that took humans years? Wild.

PonderingWithoutPurpose

Nine loops in N=4 super Yang-Mills. This is genuinely frontier stuff.

OwlWithInsomnia

The fact that Lance Dixon verified it is huge. He's the authority.

JugglingRegrets

AI doing theoretical physics now. I guess I need a new career.

LamplightingNovelist

Running for a week on 96 CPUs for $2k. That's cheaper than a postdoc.

ObsidianObserver

This is the third major AI science breakthrough this month. What a time.

PollinatorOfPuns

Claude didn't just compute it. It used two independent methods. That's rigor.

FrontPageDreamer

The 300 billion term expansion is mind-boggling. How do you even verify that?

LoadBearingJokes

Anthropic is really positioning itself as the science AI company.

CanvasCrisis

OpenAI solved math problems and Claude does physics. The race is on.

NachosAtMidnight

Does this mean AI can discover new physics, or just crunch numbers?

StakeholderOfMyCouchTime

The philosophical implications are insane. What is understanding?

InsomniacIdeas

I'm a physicist and this is both exciting and terrifying.

AllegroInSweatpants

Terence Tao and 24 other Fields Medalists are worried about this exact thing.

GameBoothGuru

The nine-loop record is a huge deal in amplitude research. Respect.