opt1

Founder | Builder | Hacker

[opt1 @xopt1x]

Used a weekly reset credit, codex is still going.

[opt1 @xopt1x]

Make codex run for days non-stop in an iterative loop with this one simple trick. Comment and like for the trick.

[opt1 @xopt1x]

@UK_Daniel_Card lol

[opt1 @xopt1x]

Make codex run for days non-stop in an iterative loop with this one simple trick. Comment and like for the trick.

[BridgeMind @bridgemindai]

FABLE 5 CAME BACK NERFED.

We re-ran the July 1st version of Claude Fable 5 on BridgeBench.

The results are brutal:

Debugging: 86.2 → 25.9
Refactoring: 73.6 → 38.4
Hallucination: 75.9 → 61.7

The new guardrails are kicking in on way too many tasks and falling back to Opus 4.8.

This is not the model that got banned.

Anthropic owes everyone an explanation.

[opt1 @xopt1x]

@Jhaddix Was looking to get this but waiting on a 50% off sale.

[Georgia Channing @cgeorgiaw]

Seems like no one's noticed the 80TB of astrophysics data from 30+ sources that just dropped on @huggingface.

...and you only need ~4GB of RAM to load it.

We're talking over 80TB of galaxy imagery taken across the spectrum, spectra of galaxies and stars, time series of variable stars, and a whole zoo of assorted measurements and physical data.

And all of it can now be wrangled on your laptop, thanks to Multimodal Universe's just released cross-matching. SDSS x Gaia means you can match 800k objects against 122M objects, and it never climbs above ~4GB of RAM.

Huge congrats to @smith42mike for leading this and making the world of astro accessible to probably 10,000x more people. Let's discover some shit

[NVIDIA AI @NVIDIAAI]

We took a 30B model and split it in two to write tokens in parallel instead of one at a time.

Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch.

We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

[opt1 @xopt1x]

Well those are each one layer. A lot of large corporations also implemented better WIFI practices like isolated clients. McDonalds, Starbucks, smart Airports. But a lot of lazy places didn't, like hotels, restaurants, etc. So it was still game for a while, also the discussion Danny focuses on is stealing web traffic, he doesn't touch on public file shares, weak remote access services, os or service vulns. Or even already backdoored systems connecting to a non-isolated network.

I have been hacking WIFI since 2000, and stopped around 2017. So I am a bit out of the current, but, there are a lot of ways we did hack devices over networks.

I found all types of services connecting to wifi from peoples devices. Unauthenticated rsync servers (lol), smtp, ftp, http, list goes on.

So should you trust public wifi? Well if client isolation is on, its less risky, but even like danny said, is it tapped? Lol

[opt1 @xopt1x]

@DenLoginoff @UK_Daniel_Card It used to be an issue years ago. Around the time I published this automated cookie farming tool, the public WIFI technologies began changing.

https://t.co/HIriapqRR4

[cr3ghost @cr3ghost]

Phrack 72 published the analysis of an actual North Korean APT workstation dump.

Real Kimsuky source code. A kernel-level remote backdoor. A private Cobalt Strike beacon. Android malware. Stolen South Korean government GPKI certificates. Access to the Defense Counterintelligence Command and Ministry of Foreign Affairs.

A South Korean security firm independently confirmed the findings. They matched the leaked rootkit source code to a rootkit found during a real 2022 incident at a South Korean financial institution. Same code, same encryption keys.

Opens with “Dear Kimsuky, you are no hacker.”

https://t.co/kaMJUzmaLr

https://t.co/YJ7glibYHQ

Authors: Saber + cyb0rg

Published in @phrack Issue 72, 40th Anniversary

Follow-up analysis by ENKI White

#ThreatIntel #Malware #InfoSec

[opt1 @xopt1x]

Was hired by a music company to build an automated A&R pipeline. Ingested over 170,000 songs to train the ML, along with its contextual metadata from their systems. We finally did round 2 of testing today and got wicked results. Crazy fun project - I want to build a bigger one next.

[opt1 @xopt1x]

"I don't give up until the wheels fall off - That's what I'm made of" - Rodney Mullen.

It's hard to have something internal that propels you forward, relentlessly striving to build and create. And I never (and still don't) understand my internal mechanism that keeps me going. It's like the locomotive sits on the tracks, and some *thing* keeps shoveling fuel in to it.

I don't know why I keep going, but I am not built to stop.

[Suraj Sharma @suraj_sharma14]

Bro I'm so sick of pretending this isn't weird.

The internet spent 20 years creating tutorials, open-source projects, blog posts & answers for free.

AI companies turned all of it into products worth billions.

And now the same people who created that knowledge are being told they're replaceable.

We built the library.

Someone else started charging admission.

[opt1 @xopt1x]

@HackingDave Your numbers and mine are an Anagram!
6'1, 244lb. Lolol

[Cursor @cursor_ai]

We're launching code storage and git hosting.

Origin gives teams and agents a place to host, review, and collaborate on code.

Available this fall. Join the waitlist.

https://t.co/uamaIarJXY

[opt1 @xopt1x]

@infosec_fox

[SpaceX @SpaceX]

SpaceX has exercised the option to acquire @cursor_ai in an all-stock transaction with the goal of building the world’s most useful AI models.

For the past few months, SpaceXAI has been jointly training a model with Cursor, which will be released in Cursor and Grok Build soon.

We look forward to working closely with the Cursor team to advance our frontier AI capabilities

[opt1 @xopt1x]

@UK_Daniel_Card Sure

[opt1 @xopt1x]

@HackingDave Over promoted its capabilities, scared uneducated politicians, and problems ensue.