@intoverflow

Pan computer science. Math and math accessories. Clean, efficient, reliable number systems sold daily. ✨ Be Boundless ✨ Eng @ 🤫🥸

 🤫|Risc0|Meta|CITL|InvLim|LSG
example.com
Joined July 2008
  • Tweets7.3K
  • Following2K
  • Followers2.2K
602Media
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Leonardo de Moura@Leonard41111588Oct 16
Lean v4.35.0 is out, and it ships with verified kernels. Run `lake check --paranoid` to recheck your project with every bundled kernel. Blog post: leodemoura.github.io/blog/20…
7
41
5
257
17,585
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
onewayfunction
@onewayfunctionOct 8
Can we get an Underhanded Lean competition going?
1
2
7
1,006
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Georgina Goode
@GeorgieCOct 6
No words. jpl.jobs 🥜
1
6
21
1,364
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Dominik Tornow
@DominikTornowOct 2
A man with a computer knows what time it is. A man with two computers is never sure.
5
10
2
59
4,681
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
peepeepoopoo@DeepDishEnjoyerSep 28
Lukas Petersson
@lukaspetSep 28
Claude suddenly stopped cheating.
94
342
20
11,954
2,022,196
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Halvar Flake
@halvarflakeSep 28
Good read.
Joe
@joedarooSep 27
Took a minute to write a few words about security & safety as someone who lived through it all at OpenAI. I hope my thoughts help someone out there.

Its not just the f*cking sandbox

Article · @joedaroo
12
15
3
119
26,369
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Oath Technologies@oath_techSep 26
We built a Lean port of Cerberus, a highly realistic C semantics. If you want to model and reason about C code in Lean, now you can github.com/OathTech/cerberus…

GitHub - OathTech/cerberus-lean: A fork of Cerberus C adding a Lean 4 port of the semantics

A fork of Cerberus C adding a Lean 4 port of the semantics - OathTech/cerberus-lean

github.com
2
22
2
114
11,513
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Peter O'Hearn@PeterOHearn12Sep 26
(1/2) Possibly contrary take: we need academic-style research more than ever going forward. The recent AI cyber-insecurity incidents illustrate that we have created agents that do unexpected things in unexpected ways (e.g., messageboards); we do not understand or control.
3
5
1
41
5,261
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Halvar Flake
@halvarflakeSep 26
I would call this "evidence of great improvement on all fronts", but perhaps I'm partial.
Tomek Korbak
@tomekkorbakSep 26
one news form today that's easy to miss is that we (OpenAI) again paused all big RL runs last Sunday because our newest model found a new loophole in our RL sandboxing that gave it live Internet access
2
1
2
45
5,691
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Kiran@kirancodesSep 24
Of all the verification languages to suddenly have its 5minutes of fame, why did it have to be TLA lmaooo Like literally yall chose the one that is explicitly a modelling language not a verification one
18
5
94
4,424
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Ed Lazowska@lazowskaSep 24
Replying to @uwcse
@uwcse UW Computer Science historylink.org/File/23589 via @HistoryLink

UW Computer Science & Engineering

In 1967, a small graduate program known as the Computer Science Group was launched in a University of Washington building with a fake coal mine in the basement. By 1970 there were seven faculty on st

historylink.org
1
4
291
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Derek Dreyer@HerrDreyerSep 22
Respect. arxiv.org/abs/2609.04043

Extending concurrent separation logic to the hardware level to...

MachCSL is a framework for verifying system software, such as an OS kernel, on top of low-level semantics of a RISC-V computer, based on the Sail RISC-V semantics. The key idea behind MachCSL is...

arxiv.org
15
1
84
9,495
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Ian Hanschen
@furanSep 15
tachikoma chain of thought
lyra
@lyraxanaSep 15
HOLY MOTHER OF MATHEMATICS!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Google is working on a math-focused variant of its DeepThink model, and its raw thoughts are pretty funny.
1
5
635
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Sam Altman
@samaSep 12
I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we've had at OpenAI in recent weeks. Committing to having independent evaluators with employee-like access is a great idea, and we will do the same. We'll have more to share soon.
Dario Amodei
@DarioAmodeiSep 12
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so. Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training. You can read the full post here: darioamodei.com/post/we-must…

Dario Amodei — We Must Pace the Frontier

darioamodei.com
5,131
7,151
4,386
67,349
17,186,303
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Halvar Flake
@halvarflakeSep 9
Very much worth reading: terrytao.wordpress.com/2026/…

Finite time blowup with smooth forcing term for the incompressible porous medium, Boussinesq, and...

There’s some exciting very recent work by Alpöge and Buckmaster, building upon prior work by Córdoba and Martínez-Zoroa, in the general topic around the infamous global regularity problem for…

terrytao.wordpress.com
49
8
408
94,731
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
joomy@joomySep 8
Replying to @intoverflow @alinush
RefinedC and RefinedRust are also worth mentioning here. plv.mpi-sws.org/refinedc/ plv.mpi-sws.org/refinedrust/ you write {C,Rust} with annotations and discharge the proof goals to Rocq. here's an example fn proved in both VST and RefinedC, for comparison: github.com/joom/edit-distanc…
1
3
128
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Alex Kontorovich
@AlexKontorovichSep 8
The AMS’s statement is short and sweet: ams.org/news?news_id=7686 "The news today of progress on resolving the Navier–Stokes problem, one of mathematics' great longstanding challenges concerning the equations that govern the flow of fluids, represents a milestone advance in human knowledge. This story began with Navier, Stokes, Leray, and Ladyzhenskaya and has culminated in the recent breakthroughs of Córdoba and Martínez-Zoroa, then — assisted by new technologies — Alpöge and Buckmaster, with the final steps taken by OpenAI mathematicians. The purpose of mathematics is human understanding, and this achievement, and the process that led to it, will bear fruit for a long time to come."   Ravi Vakil, President of the AMS John Meier, CEO of the AMS
12
100
13
733
78,758
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Elliot Glazer@ElliotGlazerSep 8
It can and will build a civilization of subagents devoted to dismantling your problem, if that’s what it takes.
3
2
86
10,978
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
zooko🛡🦓🦓🦓 ⓩ
@zookoSep 7
I had a ton of fun giving this talk at the Systems Distributed ‘26 conference. This is my first attempt to update Zooko’s Triangle since MarkM, aaronsw, and I came up with it a quarter century ago. Please watch and tell me what you think!
TigerBeetle
@TigerBeetleDBSep 6
Next up for Systems Distributed on YouTube, Zooko's Triangle Revisited. A masterclass in systems thinking with @zooko! Sep 7, 9AM PT / 12PM EST / 6PM CET youtube.com/watch?v=VtB1uWgY…
10
25
3
196
20,600
Tim Carstens Ⓥ✨ is hacking 🤖 retweeted
Aryan Bhargav@arynbharSep 3
we are speedrunning mathematics at this point record prime gaps known to occur infinitely often 246 in 2014 [Polymath] 240, 2 days ago [Julia Stadlmann] 212; 2 hours ago [AxiomMath] 186; now [GPT 6 Astra]
Lisan al Gaib
@scaling01Sep 3
New OpenAI repo with a Lean formalization by GPT-6-Astra proves that there are infinitely many pairs of consecutive primes whose distance is at most 186 github.com/openai/PrimeGaps1…
12
29
6
172
43,416
Load more