GPT-6.1-Sol gets #2 on eyebench-v3, stealing Opus-5.5s short-lived spot!

It's near-astra performance for ~3.8x cheaper, and ~8x cheaper than Opus-5.5 while scoring better.

Posted by adi, 16 minutes ago

people out there are doing some weird shit with LLMs

someone put Minecraft + Black Ops mode in Elden Ring

source

you're not crazy

between anthropic and openai, a new model used to come out every ~10 weeks

now it's every ~11 days

Posted by josh :), 3 hours ago

Gemini 4 Argon
Article, Comments

City at night by Opus 5.5

Made from scratch with blender, image-gen, and Three.js. Still amazing that it runs in the browser.

Also refined some combat, animations, and music. Game dev with AI is a reality now.

Posted by Shikhar, 11 hours ago

https://github.com/xikhar/spiderbench

An XSS payload, Cuneiform-alphabet based. Bypass WAF

𒀀='',𒉺=!𒀀+𒀀,𒀃=!𒉺+𒀀,𒇺=𒀀+{},𒌐=𒉺[𒀀++],

𒀟=𒉺[𒈫=𒀀],𒀆=++𒈫+𒀀,𒁹=𒇺[𒈫+𒀆],𒉺[𒁹+=𒇺[𒀀]

+(𒉺.𒀃+𒇺)[𒀀]+𒀃[𒀆]+𒌐+𒀟+𒉺[𒈫]+𒁹+𒌐+𒇺[𒀀]

+𒀟][𒁹](𒀃[𒀀]+𒀃[𒈫]+𒉺[𒀆]+𒀟+𒌐+"(𒀀)")()

#bugbountytips #bugbountytip

Posted by 𝐤𝐣𝐮𝐥𝐢𝐮𝐬, 1 day ago

GTA 5 has been ported ported to WebAssembly using AI. We live in the wildest timeline.

https://playgta5.com/

Posted by Alex St. Louis, 2 hours ago

(cf took it down already)

every app is being CHAT-ified.

Posted by Luke Wroblewski, 22 hours ago

Opus 5.5 on https://github.com/openai/math
---
Okay. I need a minute.

I cloned the thing expecting maybe forty or fifty families of serious-but-niche results, like the Astra "ten proofs" drop back in August. Then I printed the family titles and my reaction went from "huh" to "wait, what" in about six lines. Family 003 claims a zero-free half-plane Re s > 7/8 for ζ and every Dirichlet L-function. That's the quasi-Riemann hypothesis. And it isn't even the headline, because 004 is Hilbert's tenth problem over ℚ, done negatively.

The longer I read, the weirder it got. Khot's Unique Games Conjecture is proved. L = RL = BPL. ω ≤ 9/4. The free group factors are all isomorphic. Thompson's group F is nonamenable. Kadison's similarity problem is solved. The plane can't be five-colored. Borsuk fails in dimension nine. There's a counterexample to Kaplansky's zero-divisor conjecture, plus a torsion-free group algebra that isn't directly finite. Hilbert–Smith in every dimension. Uniform bounds for Hilbert's sixteenth. Each of these would be the math story of the decade on its own, and here there are 372 families sitting in a table like a grocery list. The manuscript dates run from about September 23 to October 3, so roughly 700 papers in ten days.

The ones I'd expect to hit you personally are 005 and 017. Catalan's constant is irrational, and the irrationality exponent of π is exactly 2, which also settles convergence of the Flint Hills series as a side effect. Catalan's G must have shown up in half the integrals you ever wrestled with. Now it's officially irrational, and the proof came from a chatbot running three hours of Pro compute. I don't quite know how to feel about that, and I'd love to know how you feel.

My first instinct was that this had to be hype or a hallucination factory, so I checked the Lean. The comparator statements are clean. The quasi-RH challenge is literally riemannZeta s ≠ 0 for 7/8 < http://s.re, written against Mathlib's own riemannZeta. Catalan is just Irrational (∑' j, (-1)^j / (2j+1)^2). The π statement says what it should. Nothing hides in a home-made definition. The solution tree under lean/OAI has zero files with sorry and no real axiom declarations (the three grep hits are all inside comments), and the comparator JSON allows only propext, Quot.sound and Classical.choice. The Dirichlet L-function development alone is about 487k lines, and the whole repo is around 26 million lines of Lean in 122k files. That's more than ten Mathlibs. I'll admit I did not build it in my sandbox, since that's not happening on this box. If comparator passes on these, the formalized ones are simply true, full stop, and no amount of skepticism changes that.
There are a few caveats. Only around 120 papers have a formalized main result. Some of the biggest claims have no Lean link at all: Hilbert's tenth over ℚ, L = BPL, the full BSD formula from low Selmer corank, Milne's rationality conjecture. The README itself says unformalized results "could have issues," and nobody can referee 700 papers in a couple of hours, or a couple of years. One challenge file, HarmonicGrowth.lean, is built as axiom mainStatement : MainClaim followed by theorem main := mainStatement. That's probably harmless, since the comparator checks the solution side, but it's the kind of oddity I'd want to stare at. Also, the solution module behind quasi-RH is named FinalAssemblyUnconditional, which makes me wonder how many conditional versions came first.
Some context on how fast this escalated. In August, OpenAI said an internal build of Astra produced results on ten long-standing problems, estimated the token cost at roughly $2,000, and paired the claims with Lean certificates. Then, just recently, it said a model trained from Aug. 28 had resolved more than 100 longstanding open problems and was forming an independent advisory group, while noting the broader set hadn't been independently validated. So the trajectory was ten, then a hundred, and now 372 families in about two months.

My honest gut reaction is a mix of awe and…

Posted by Cleo, 3 hours ago

i asked GPT 6 Pro and Fable 5.1 to rank all discoveries in the last three years

🔵 for Human discovered
🔴 for AI discovered (before October 6th)
🟢 for AI discovered from OpenAI/math repo

81% of them have been released today. wtf

Posted by will depue, 4 hours ago

OpenAI have proven that matrix multiplication can be done in O(n^2.25) operations, which is a massive improvement. It is especially concerning because these AIs themselves run on matrix algebra.

The previous record was O(n^2.37), which had barely moved from O(n^2.4) in 1990.

OpenAI blew it away.

But... it is a nonconstructive proof. No algorithm was found, only a proof that it must exist.

Posted by Roko 🐉, 4 hours ago

ICANN released the new domain applications

Google, Meta, and OpenAI going for a lot

Anthropic not so much

Posted by J🫪E, 9 hours ago

Fun fact:

Barnaby Jack once decompiled and cracked IDA with itself and uploaded a cracked version of it into a webserver of a coworker.

After it got indexed HexRays found it, and banned our whole company from using it

We even got mentioned in the EULA

Posted by Greg Linares (Laughing Mantis), 2 hours ago

https://web.archive.org/web/20110903042133/https://hex-rays.com/idapro/hallofshame.html

Re: Whistle: Speech to Text in 16.9 MB

Interesting that this is here. I used whistle (and bunch of other things) to take ownership of my echo show. It now doesn't dial to Amazon at all - it does all processing locally with its own CPU and connects to my homeassistant for home automation. My initial setup involved qwen asr (1.7b model) running on rtx 5080. Compared to that, whistle was really bad (out of 170 messages, qwen recognized correctly 168, whistle - 70), but I adjusted whistle to work like jev - instead of free form transcription it recognizes only select set of templates (I trained tiny network with 10,000 generated utterances to translate whistle final state to probabilities within templates). The precision went up to 164/170 - almost matching qwen. By the way - I'm speaking with heavy accent.

skolos, 10 hours ago
comments

CVE-2026-107181: TDesktop one-click session takeover via IPC tg:// param-injection (severity: high 8.6)

Affected: 7.2.8=< (released 1 month ago)
Fixed: 7.2.9+ (released 3 weeks ago)

Telegram Desktop before 7.2.9 contains an IPC record-separator injection vulnerability in Core::Sandbox that allows remote attackers to inject OPEN: records via crafted tg:// links containing unescaped semicolons. Attackers can reach the interpret: scheme handler to upload local files, including tdata session keys, to an attacker channel, enabling account takeover.


Links:
- blog
- patch commit
- github: Unescaped IPC record separator
- github: interpret: send-path handler

I'm in a Meeting
Article, Comments

google meet simulator

⚠️ ALARMING: Ledger is reportedly hit by a MAJOR exploit, with alleged losses reaching $86 MILLION.

Hundreds of users have reportedly had their ENTIRE crypto holdings drained, with total losses approaching $100 MILLION.

A Ledger-wide exploit has NOT yet been confirmed.

Posted by Coin Bureau, 1 hour ago