pilvar (Philippe Dourassov)'s banner

pilvar (Philippe Dourassov)

@pilvar222 • 2,948 subscribers

AI Pentest Lead @AikidoSecurity

Shorts

HOLY MOLY: Aikido got GPT-6 Astra in advance to run it on our Cybersecurity benchmark, it crushed EVERY other model! - At pass@3 it rediscovered 29/32 CVEs, the highest recall we've ever recorded and 4 more than GPT-5.6-Sol - Even at pass@1, it has 75% recall. The model is VERY consistent - The performance however come at a high price (literally). The three runs cost us almost $4,000 Astra is now the #1 model on the benchmark Debarshi and I built, and by a LOT 1/3 🧵

HOLY MOLY: Aikido got GPT-6 Astra in advance to run it on our Cybersecurity benchmark, it crushed EVERY other model! - At pass@3 it rediscovered 29/32 CVEs, the highest recall we've ever recorded and 4 more than GPT-5.6-Sol - Even at pass@1, it has 75% recall. The model is VERY consistent - The performance however come at a high price (literally). The three runs cost us almost $4,000 Astra is now the #1 model on the benchmark Debarshi and I built, and by a LOT 1/3 🧵

49,224 просмотров

Holy moly: GLM-5.3 got much better in cybersecurity since our pre-release evaluation with Z.ai. It now matches GPT-5.6-Sol on our cybersecurity benchmark at 0.4x the cost 🤯 - At pass@1: it went from 60.4% to 65.6% CVEs rediscovered, crushing every other open model on one-shot tasks - At pass@3: it did 75% -> 78.1%, matching GPT-5.6-Sol - Its precision remained stable, reporting fewer false positives than DeepSeek models The performance increase comes from a behavioral change: the new version is more persistent. It tends to run longer, and had a ~43% reasoning tokens increase. But the performance upgrade is worth that additional cost. 1/3 🧵

Holy moly: GLM-5.3 got much better in cybersecurity since our pre-release evaluation with Z.ai. It now matches GPT-5.6-Sol on our cybersecurity benchmark at 0.4x the cost 🤯 - At pass@1: it went from 60.4% to 65.6% CVEs rediscovered, crushing every other open model on one-shot tasks - At pass@3: it did 75% -> 78.1%, matching GPT-5.6-Sol - Its precision remained stable, reporting fewer false positives than DeepSeek models The performance increase comes from a behavioral change: the new version is more persistent. It tends to run longer, and had a ~43% reasoning tokens increase. But the performance upgrade is worth that additional cost. 1/3 🧵

34,641 просмотров

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

As soon as we saw this comment, Jorian wrote a PoC for it Here it is: unauthenticated RCE on default phpBB instances

As soon as we saw this comment, Jorian wrote a PoC for it Here it is: unauthenticated RCE on default phpBB instances

pilvar (Philippe Dourassov)

19,271 просмотров • 3 месяцев назад

My friend Manaf and I have created an AI hacker! 👨‍💻 It can find 0-days in web applications with 0% false positives, and it also works on CTF challenges! What you see ⬇️ is Haicker solving a web challenge from UMassCTF 2025 (only 45/557 teams were able to solve it!)

My friend Manaf and I have created an AI hacker! 👨‍💻 It can find 0-days in web applications with 0% false positives, and it also works on CTF challenges! What you see ⬇️ is Haicker solving a web challenge from UMassCTF 2025 (only 45/557 teams were able to solve it!)

pilvar (Philippe Dourassov)

36,110 просмотров • 1 год назад

Больше нет контента для загрузки