Rishi Mehta (@rishicomplex) | teapawt

Tweets
All Posts Highlights
Sort by
Most recent Popular
Tweets & Replies
Reposts
Media
Search

Rishi Mehta retweeted

@philipakleinNov 2

So I finally caught "You Can See Everything" last night and I never would have predicted that in a documentary about one of the worst corporate fraudsters in history and her psychopath husband that the person who would come out worst would randomly be Amanda Seyfried.

174

3,516

486,993

Rishi Mehta retweeted

@karinanguyenOct 22

Anthropic has been warning about AI-enabled cyberattacks at least since 2022, with @logangraham at the forefront of that work. Back then, no VC was taking it seriously. The paradox is that risks only feel real once they’re legible, but waiting for incidents is too late. Anthropic also built sycophancy evals (@EthanJPerez et al.) a couple years before real-world incidents with ChatGPT made the risks of models excessively agreeing with users legible at scale. The lesson is to take early evidence of AI risks seriously before harm at scale makes them impossible to ignore.

9

156

12,124

@rishicomplexSep 28

sonnet 5.5 is a huge upgrade for sonnet users

@claudeaiSep 28

Replying to @claudeai

Sonnet 5.5 improves on Sonnet 5 across benchmarks, in some cases dramatically. It’s a faster, lower-cost complement to Claude Opus 5.5, strongest at well-scoped everyday tasks, fixing bugs, and creating polished documents, slides, and spreadsheets.

Image from Claude (@claudeai)

8

495

@rishicomplexSep 26

Opus 5.5 tops RoastBench

Image from Rishi Mehta (@rishicomplex)

Image from Rishi Mehta (@rishicomplex)

@rishicomplexJun 1

Made a little benchmark called RoastBench - it compares frontier models on their roast jokes. The models roast 10 personalities from comedy central roasts I enjoyed, and I manually rank their jokes. I also mark the ones that made me laugh. LLMs are way worse than top humans.

Image from Rishi Mehta (@rishicomplex)

2

12

1,515

@rishicomplexSep 26

banger

Image from Rishi Mehta (@rishicomplex)

5

173

@rishicomplexSep 24

vibes

6

589

@rishicomplexSep 22

i love this model. it's cracked at pretty much everything and talks like a smart human

@claudeaiSep 22

Replying to @claudeai

Opus 5.5 is a major step up from Opus 5, leading on agentic coding, computer use, and knowledge work.

Image from Claude (@claudeai)

147

4,183

@rishicomplexSep 7

RoastBench results for Fable 5.1 and GPT-6. They're tied on the main set, but fable does a lot better on the second set. Neither felt like a major step up though.

Image from Rishi Mehta (@rishicomplex)

Image from Rishi Mehta (@rishicomplex)

@rishicomplexJun 1

Made a little benchmark called RoastBench - it compares frontier models on their roast jokes. The models roast 10 personalities from comedy central roasts I enjoyed, and I manually rank their jokes. I also mark the ones that made me laugh. LLMs are way worse than top humans.

Image from Rishi Mehta (@rishicomplex)

1

6

2,067

@rishicomplexSep 7

Couple bangers from them

Image from Rishi Mehta (@rishicomplex)

Image from Rishi Mehta (@rishicomplex)

1

1

325

@rishicomplexSep 7

All jokes here rishimehta.xyz/roast_bench/

2

1

153

@rishicomplexSep 5

i would've said "nw, i'm waiting with the table!" and then bailed

Blaine Anderson

@datingbyblaineSep 4

Single girlfriend in NYC's text exchange with a Raya match. This is tough... In her defense, 15 minutes isn't crazy late for NYC. The guy's texts also invite drama. You're already at the table, why not wait and see how she reacts in-person, then decide whether you're still interested? In his defense, she should've at least been in the car by 7:15 for a 7:15 date. An apology for being late feels 100% appropriate. IMO they both could've done better. It's also clearly not a match, so at least they figured that part out 🥲

Image from Blaine Anderson (@datingbyblaine)

2

29

4,904

@rishicomplexSep 4

claude formalized flt

@AnthropicAISep 4

Checking that a major mathematical proof is correct can take years. Formalization—converting the mathematical reasoning into a form computer proof assistants like Lean can verify—can help. Last month, Claude completed the first formalized proof of Fermat’s Last Theorem, one of the most famous theorems of all time. This was a project experts thought would take many years. It is the largest Lean proof ever written. Fermat’s Last Theorem was first proven in 1995 by Sir Andrew Wiles, more than 350 years after it was conjectured. Our proof, which totals over 13 million lines of code, provides machine verification. More importantly, it proves over 29,000 other theorems that the proof requires, across many areas of math which had never before been formalized. We see this as a major step in the long process of firming up the core of mathematical knowledge, building on work from three centuries of mathematicians and hundreds of contributors to Lean and Mathlib. We are optimistic that AI-assisted verification of mathematical proofs will help reduce the burden of refereeing mathematics in an era where more proofs are being produced than ever before. You can read about the process on our Science Blog: anthropic.com/research/forma… And see the complete proof on GitHub: github.com/anthropics/fermat…

1

15

1,590

Rishi Mehta retweeted

@cursor_aiSep 1

Claude Fable 5.1 is now available in Cursor! It's the most capable model we’ve run on CursorBench 3.2, scoring 73.4% at max effort. We found it especially skilled at verifying its own work, allowing it to take on difficult coding tasks from start to finish.

278

7,652

570,135

@rishicomplexSep 1

if you've been holding out on fable due to price or zdr you really should try fable 5.1

Image from Rishi Mehta (@rishicomplex)

We’re introducing Claude Fable 5.1 and Claude Mythos 5.1. They're the world’s most advanced models for coding and knowledge work.

2

101

10,910

@rishicomplexSep 1

fable 5.1 is here! smarter & cheaper

Replying to @claudeai

Claude Fable 5.1 is available everywhere today. Claude Mythos 5.1, our model for cyberdefenders and life scientists, is available through trusted access programs. Read more: anthropic.com/claude-fable-a…

Introducing Claude Fable 5.1 and Claude Mythos 5.1

Introducing Claude Fable 5.1 and Claude Mythos 5.1

Our most advanced models for coding and knowledge work. Their research capabilities also offer an early glimpse of how AI models will contribute to scientific progress.

2

48

3,274

Rishi Mehta retweeted

Jesse Mu@jayelmnopAug 10

wake up babe, new neurosymbolic harness just dropped

Image from Jesse Mu (@jayelmnop)

49

4,068

193,226

@rishicomplexAug 10

curious to hear from mathematician friends re how impressive this is

@AnthropicAIAug 10

We asked an unreleased research version of Claude to take a stab at the Riemann hypothesis. It didn’t solve it, but it did make strides on a related problem: it increased the lower bound for the fraction of zeros of the Riemann zeta function that satisfy the hypothesis from 41.6% to 67.2%. anthropic.com/research/riema…

$Claude has improved on a longstanding lower bound for the fraction of zeros of the Riemann zeta...$

Claude has improved on a longstanding lower bound for the fraction of zeros of the Riemann zeta...

An unreleased Claude model improved the lower bound for the fraction of zeroes of the Riemann zeta function that satisfy the hypothesis, raising it from 41.6% to 67.2%.

4

739

@rishicomplexJul 28

Added Opus 5, GPT 5.6 Sol and Kimi K3 to RoastBench. Opus 5 narrowly edges out Fable at the top

Image from Rishi Mehta (@rishicomplex)

@rishicomplexJun 1

Made a little benchmark called RoastBench - it compares frontier models on their roast jokes. The models roast 10 personalities from comedy central roasts I enjoyed, and I manually rank their jokes. I also mark the ones that made me laugh. LLMs are way worse than top humans.

Image from Rishi Mehta (@rishicomplex)

1

2

864

@rishicomplexJul 28

this one made me laugh

Image from Rishi Mehta (@rishicomplex)

1

2

333

@rishicomplexJul 28

also lol'd at this from kimi

Image from Rishi Mehta (@rishicomplex)

3

252