⡷⡇⡱⢎⢸⠽⢸⢼

65 posts

⡷⡇⡱⢎⢸⠽⢸⢼ banner
⡷⡇⡱⢎⢸⠽⢸⢼

⡷⡇⡱⢎⢸⠽⢸⢼

@hexpwn

I'm back but I'm not happy about it (ノ｀Д´)ノ彡┻━┻ https://t.co/g7PBoJlLwS and https://t.co/CmCSsmB1oh

8.2 ± 0.4 kpc from Sgr.A* Katılım Aralık 2019
347 Takip Edilen156 Takipçiler
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
I'm sure we'll eventually see adversary Agent Automation attacks that have hidden "token mining" as their goal. LLM related tools that stealthly are syphoning your tokens for their own queries.
English
0
0
1
25
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
I can't help but verbally abuse Claude... these things are so far from intelligent. Useful? Yes. Intelligent? No. We are still in "this wastes a lot of time" territory for complex activities. I'm sure we'll get there, but right now it's like managing 50 monkeys with typewriters.
English
1
0
2
158
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
So now I can be 5x as frustrated wrangling monkeys with typewriters for the same price 🫠
English
0
0
0
10
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
anthropic threw a wrench on opencode usage recently (x.com/thdxr/status/2…); so I tried claude code again after a few months, now with more day to day experience with the LLM, I'm impressed at how much more efficient CC is. My token burn rate is maybe 20% of what it used to be.
dax@thdxr

opencode 1.3.0 will no longer autoload the claude max plugin we did our best to convince anthropic to support developer choice but they sent lawyers it's your right to access services however you wish but it is also their right to block whoever they want we can't maintain an official plugin so it's been removed from github and marked deprecated on npm appreciate our partners at openai, github and gitlab who are going the other direction and supporting developer freedom

English
1
0
0
179
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
LLMs lack two very important hacking tools that I don't see them developing and surpassing humans for a while: persistence and creativity.
English
0
0
0
23
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
I just heard about i.blackhat.com/BH-EU-25/eu-25… and it pisses me off. Not the research; but Amazon's posture on denying multiple requests to disclose my research. My only guess is that ND policy applies differently depending on who the researcher is (or who is behind them)
English
0
0
0
54
⡷⡇⡱⢎⢸⠽⢸⢼ retweetledi
Ahmad
Ahmad@TheAhmadOsman·
Dario and Anthropic are internet clowns now
English
66
245
2.7K
294.4K
⡷⡇⡱⢎⢸⠽⢸⢼ retweetledi
Rebane
Rebane@rebane2001·
i built an entire x86 CPU emulator in CSS (no javascript) you can write programs in C, compile them to x86 machine code with GCC, and run them inside CSS
English
331
1.1K
10.2K
1.1M
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
@thedawgyg Yep... it definitely is a very useful tool and can speed up a lot of the workflow, but it fails catastrophically in other ways and cannot drive itself, at least yet. Regardless I think people who have not yet used or incorporated these tools should do so; things are moving fast
English
0
0
1
64
dawgyg - WoH
dawgyg - WoH@thedawgyg·
Hot take: Everyone that is worried that ClaudeAI now doing 'code security' is going to end security jobs have obviously never used as SCA/SAST tool and it shows... if your worried they are gonna replace you, then you probably aren't very good to begin with lol
English
24
22
358
15.8K
LiveOverflow 🔴
LiveOverflow 🔴@LiveOverflow·
After a long pause, a new video coming today! Part 1 of small documentary about Pwn2Own…
LiveOverflow 🔴 tweet media
English
12
40
515
24.3K
⡷⡇⡱⢎⢸⠽⢸⢼
⡷⡇⡱⢎⢸⠽⢸⢼@hexpwn·
i love TOCTOU vulns. I love their name. I love how they work. But I especially love how I keep stumbling onto them 🥰
English
0
0
1
30

Keşfet

@thedawgyg @LiveOverflow @elonmusk @BarackObama @taylorswift13 @cristiano @BillGates @NASA