Study: OpenAI's GPT-4 Can Autonomously Exploit 87% of One-Day Vulnerabilities | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		Study: OpenAI's GPT-4 Can Autonomously Exploit 87% of One-Day Vulnerabilities (techrepublic.com)
		4 points by billybuckwheat on April 26, 2024 \| hide \| past \| favorite \| 2 comments

skilled on April 26, 2024 [–]

Recent,

GPT-4 can exploit vulnerabilities by reading CVEs (https://news.ycombinator.com/item?id=40101846) - Apr 2024 (29 comments)

Source of the story,

https://arxiv.org/abs/2404.08144

gnabgib on April 26, 2024 | [–]

And LLM Agents Can Not Autonomously Exploit One-Day Vulnerabilities https://news.ycombinator.com/item?id=40125884

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact