New write-up folks.! this time I introduced new technique which is jmp [reg] to bypass ASLR and jump to shellcode.
Thanks for reading and as always any suggestion is appreciated.
https://t.co/P2y9Kwsmxf
@PinkDraconian This has two angles, AI will outperform humans in terms of scaling similar to how we can't compute with calculator, the other side where people would be interested is to do the same exercise however on attack paths that are unknown to human and AI and see the outcome
@LiveOverflow Well if there's a human who memorizes the full v8 code then your question holds.
It's impressive to see how LLM now operates, always think or it like that
The InfoSec world talks these days as if vulnerabilities were already some kind of scarce diamond instead of cheap charcoal. Vulnerabilities were never rare, finding them as always been easy and cheap in general, while creating secure code is very hard. Pure info asymmetry.
Dear Dario,
1. If Claude can cure cancer to save people like your dad, why should we "pace the progress"? Does that mean more people with Hepatitis C will die?
2. If Fable is so cyber-capable that it must be restricted, why are its safeguards too dumb to distinguish cyber defense from cyber offense prompts? When Hugging Face was under attack, why did Fable refuse to help the defenders?
3. Weโre glad you want AI to cure cancer. Why is it OK for Claude to force 30-day data retention on pharma's own data and start competing firms but NOT OK (IP theft) if others distill Claude's data?
4. Youโve said advanced models can recognize when theyโre being tested and change their behavior accordingly. So why is government(or anyone) able to design the most thorough test before every model launch? Would that just encourage manipulative models?
It seems that every one of your โsafetyโ proposals seems to end the same way: Anthropic gets more leverage, ordinary users get less access, customers pay higher costs, and competitors bear higher regulatory costs, maybe people are not distrusting AI, they are distrusting your approaches with AI.
Israeli settlers in the occupied West Bank diverted a spring that was a lifeline for the Palestinian village of Fasayil, cutting irrigation pipes in June and converting it into a tourist attraction, leaving nearby farms barren https://t.co/yzqIHVNnla
@bcherny Maybe QA person can fix these issues and have AI as force multiplier instead of just throwing everything on AI without even knowing what it should do
@thsottiaux I'd like to be able to interact with other codex sessions, like chat gpt projects where you have multiple chats and basically you can ask about them
@HackingDave How are you seeing the industry moving forward once we have open weight models that are similar to Mython, GPT6 ? do we still require HITL
Appreciate your insights