I read the article and tested the demo. It’s interesting, but I’m not seeing how it materially differs from existing bot-detection approaches like Akamai’s behavioral tracking of clicks, mouse movement, timing, etc.
The obvious bots are already obvious. The hard problem is distinguishing sophisticated automation from real users without high false-positive rates.
I’m also curious how this handles accessibility: people who primarily use keyboard navigation, assistive devices, mouse keys, switch controls, or external tooling may look “non-human” by these signals.
My broader concern is that bots often operate with higher-level control than the browser UI itself, so behavioral signals seem easy to mimic or route around. What makes this approach robust against that?
@saumil@mayankagrawal Respectfully, you do not.
There is a HackerOne presence, but it has no disclosure policy, no scope, no payout guidance, and no rules researchers can rely on.
Please point me to the published program terms if I’m missing them.
Can somebody give a run-down of why I don't see online blackjack bots? Even with poor deck penetration and 8 decks, you will still surely see lots of EV > 0 games across hundreds of live tables.
I'm sure there will be too many bugs to count, but:
A completely unnecessary, over-the-top, pure JavaScript TLS state machine for TLS mimicry (fuck a 403)
https://t.co/9sFU8RcD2N
This would have never been released if it weren't for Claude
@crtyx_@MarkNFT@peachypings@unreleased they actually checked in with him to see if it was okay to risk de-valuing the brand like that. he gave the all clear and signed off on the collab