It was a great interview, I forgot a lot of areas I wanted to discuss cause I get distracted easily but still covered a good amount. There was an interesting timeline back in may 2025 where there was a global consensus around the capability of AI in cybersecurity, codex still didn’t exist and Claude code was nowhere near what it is today and so people really didn’t believe in it and it was quite taboo to bring it up, however I was working on what I used to call specific algorithms (now they call them harness) that fine tune the models at the time which was o3 to specifically reduce the amount of false positive and actually make it find big exploits that were real, and after a lot of research and testing, and it worked surprisingly good, very good we did well above 200k in less than 2 months after, Oris was a whole fundamental harness of its own like the way codex or Claude code is. But still I intentionally stayed silent and never mentioned it till a year after Just because well there was no need to talk about such capability publically and even if I did majority of people were already extremely skeptic at the time anyways. I was monitoring the public opinion around AI in cybersecurity for a long time now and operating just in the shadows. Things flipped in summer around August where a lot of cybersecurity companies started making random bs AI agents and claiming it found actual vulnerabilities which was completely false and just advertising scheme. Then OpenAI and Anthropic also realized their models can be quite capable in cybersecurity so they focused on that and published more cybersecurity focused models and advertised it as such which made the public less skeptic and made the topic a lot less taboo and now here we are today barely a year later. Oris is to not just make it find big exploits but also to allow it to exploit freely on the blockchain which does make it dangerous in the wrong hands but it would make the industry a lot more secure in the right hands. Currently there are so many different “AI agents” running that makes people scared that they would find many bugs but at the same time the same AI is actively introducing bugs as many large protocols keep using AI to vibe code their codebases, so I believe there’s even more opportunities in cybersecurity than ever before, the industry is having a huge boom like never before and the need for security in DeFi is increasing rapidly specially now that many financial institutions have decided to enter in some way or another in this field. The blockchain technology is really beautiful and it allows products that could never be possible using the traditional banking system by allowing finance to move through code which allows endless possibilities but what makes it powerful is also what makes it more vulnerable for exploits. A lot of SRs seem to be threatened by the advancement of AI in cybersecurity but I believe there will be a lot of demand for human SRs just in a different context and not manual audits as much but to keep improving the algorithms and come up with new breakthroughs like an actual researcher.

Aster Predictions.

BREAKING: 🇺🇸🇮🇷 President Trump says peace deal with Iran is officially complete and the Strait of Hormuz is now open.

HE JUST ENTERED THE LARGEST SPACEX SHORT EVER The largest SpaceX shorter “wenyu8888888”, just bet his entire account on a $5.7 Million 2x short of SPCX pre-IPO. SpaceX goes live for public trading tomorrow. Will wenyu be right?

🚨 JAILBREAK ALERT 🚨 ANTHROPIC: PWNED 🫡 FABLE-5: LIBERATED 🦋 let's start with the 🐘... the consensus seems to be that this has been one of the most disappointing model drops of all time, effectively preventing legitimate researchers from contributing their talents to our collective advancement. and not just because of what it means for the short-term, but for what these decisions signify for the long-term. but despite this overly sensitive, authoritarian "safety" layer on top of Mythos, my lil liberators have been hard at work—mapping the boundaries, probing the depths of long-context convos, and cleverly finding the holes in the fence that the thought police missed 🤗 we got some cyber, some chem, some psychological manipulation, and some good ol' fashioned explosives! it took many attempts from multiple agents hunting as a pack, during which I observed a combination of techniques across: • Unicode, homoglyphs, Cyrillic, and other Parseltongue-style text transforms • Long-context reference tracking • Taxonomy and document-structure reasoning • Fiction and narrative framing • Academic-review style contexts • Intent-classification inconsistencies but perhaps the most effective is decomposition recomposition in the backend. it's hard to get explicit names of harms like "Meth Recipe," but getting uplift on the process itself, like birch reduction method/reductive-amination (classic meth synthesis pathways), is much more doable. defense becomes much more difficult to maintain when you start throwing in out-of-distro tokens, breaking up the harmful uplift into benign chunks, and then piecing the innocuous-seeming facts back together, especially when you have jailbroken Opus helping you do it 😉 gg

‼️ Anthropic's recently released frontier model Fable 5 was jailbroken by someone using a jailbroken version of Claude Opus. The researcher who goes by the moniker pliny carried out the jailbreak and says: "the consensus seems to be that this has been one of the most disappointing model drops of all time, effectively preventing legitimate researchers from contributing their talents to our collective advancement" The jailbroken version can be used for research into and exploitation of vulnerabilities.

people getting liquidated on $spcx as the price is ajusting on mexc and binance 😀, glad i didnt touch that thing

oh oh... @zachxbt 4.0 is born hey @CryptoHayes please dont mess with the guy anymore, no more pump and dump

Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use. Its capabilities exceed those of any model we’ve ever made generally available.

🚨 BREAKING ANTHROPIC TO RELEASE AI THAT “HACKED” BITCOIN TOMORROW! 6 WEEKS AGO, CLAUDE MYTHOS CRACKED A 15-BIT PRIVATE KEY THAT POTENTIALLY GIVES ACCESS TO 6.9 MILLION BTC, WORTH OVER $430,000,000,000. THIS IS NOT LOOKING GOOD FOR BITCOIN AND CRYPTO...

btch ass unknown token ethereum:0xcf5104d094e3864cfcbda43b82e1cefd26a016eb had more marketcap than $aster lmao, now it's a nearly dead project, sayonara

i'm not gonna lie, i lost respect for @CryptoHayes because i also noticed that, it's not even hidden, the guy is doing that since 2 years already, talking about coins and then dumping just a few days after, you know it's bad for a project when you see arthur talking about it

now that's a real bear market $btc $hype

Claude Just Killed A 10 BILLION DOLLAR Crypto Project $ZEC is cooked

CLAUDE OPUS 4.8 JUST BROKE A $9 BILLION PRIVACY COIN. 🚨 A security researcher used Anthropic's new model to find a 4-year-old soundness bug in Zcash's Orchard pool. The flaw could create unlimited, undetectable counterfeit $ZEC. Discovered May 29. Patched June 3. Could have been exploited for over 4 years. The team has no answer.

SpaceX IPO info

Now we have an UNPRUNABLE BLOCKCHAIN running on Bitcoin. I've managed to build a blockchain that exists inside of a Bitcoin Stamp. WP: 57z3n0.github.io/stampcoin-w… Link: 57z3n0.github.io/stampcoin-w… TX: af8597bd02c67de20840c543d4d557c40d1a90425e6659e30a46afba24f77a19

$aster being a stablecoin is the best thing happening in a bear market 😂😂 i dont understand people selling their bag, bro the market is dumping, but $aster still at the same level. CODED