Abhaxas

Original account owned by federal friends. Probably stoned

Reformed "blackhat" circa 1995 doing appsec these days. Web2 + Web3 bounty hunter

[Abhaxas @abhaxas]

@UK_Daniel_Card Link?

[Abhaxas @abhaxas]

@thedawgyg Time for a smoke

[Abhaxas @abhaxas]

I love the shit sandwich dario is being forced to eat. What a monumental fuck up of a company and human

[Polymarket @Polymarket]

JUST IN: Anthropic says a “huge percentage” of its own employees are now barred from accessing Fable 5 & Mythos 5 under U.S. restrictions.

[Abhaxas @abhaxas]

@YourAnonNews That's great coming from you, you've been neutered and compromised since 2011.

[Abhaxas @abhaxas]

@urharmless @joetidy Lol gay

[Abhaxas @abhaxas]

hiding the message in tweet geolocation. twitter used to let you set your own coordinates using any type of float values, so we'd spread a binary chain across a bunch of tweets with a few bits tucked into each location. the tweets themselves said nothing. pull them in order, decode the geo

[Abhaxas @abhaxas]

Lol blatant liars and trying to use explicit legal speak. Of course you're not using data to improve "generative AI", you're using it on the embedding side for semantic search to help guide frontier models to accomplish the same outcome as researchers

[HackerOne @Hacker0x01]

Yesterday, we announced new product capabilities, and some of our messaging created confusion. We want to be direct.

Researcher submissions are not used to train, fine-tune, or otherwise improve generative AI models. This applies across our platform, including H1 Continuous Testing, H1 Agentic PTaaS, and Hai. Third-party model providers we work with are also prohibited from retaining or using researcher data for their own training.

We've updated our website language to reflect this more clearly. We heard your concerns, and we take them seriously.

[Justine Moore @venturetwins]

Me using Claude Opus 4.8 to rename a file

[Abhaxas @abhaxas]

@thedawgyg @Hacker0x01 Same, sitting on lots of vulns since I can't submit them lol

[Abhaxas @abhaxas]

@danshipper @every I bet this will age well

[Abhaxas @abhaxas]

An anomaly lol. That was a successful rapid defueling maneuver

[Blue Origin @blueorigin]

We experienced an anomaly during today's hotfire test. All personnel have been accounted for. We will provide updates as we learn more.

[Vitaly Baum @butaji]

@ThePrimeagen Found the classic!

[Abhaxas @abhaxas]

@h4x0r_dz Same experience for me with both h1 and bugcrowd. They will close non duplicate reports since they "can't reproduce" even after being handed a working poc script + logs. Then they fix the bug. They will also claim your report is AI slop when it isn't. I stopped submitting bugs

[Abhaxas @abhaxas]

I'd stay away from bugcrowd these days, they are straight up scamming. I submitted multiple bugs and they "can't reproduce" even with 1 shot poc scripts.

[Abhaxas @abhaxas]

I'm glad most of you are just blindly copy/pasting from Claude and Codex these days, you're too retarded to have an original thought worth sharing anyway. At least now everyone that has a social brand gets partially fact checked by an LLM

[Abhaxas @abhaxas]

@clairevo AI slop post or retarded?

[d3d aka dead (dead, мёртв, 死了) @deadvolvo]

From now on, if someone claims their exploit is bypassing ASLR, and they are simply using something like a mystery LFI to chain in order to do it... we are not friends, hell, we don't even speak the same language apparently. 🤣

[Abhaxas @abhaxas]

@thedawgyg Same, I've been learning about vm escaping just by trying to break hyper v with codex. The time for generalists is now haha

[Abhaxas @abhaxas]

@thedawgyg