Writer AI patched WriteOut, a one-click session isolation flaw in its enterprise AI platform.
A victim who clicked a public preview link while logged in could have their session cookie forwarded into an attacker-controlled sandbox, where it could be recovered and replayed.
How the preview proxy crossed tenant lines: thehackernews.com/2026/07/wrโฆ
๐ซ We at Sand Security recently discovered WriteOut - a ๐ฐ๐ฟ๐ถ๐๐ถ๐ฐ๐ฎ๐น ๐๐๐น๐ป๐ฒ๐ฟ๐ฎ๐ฏ๐ถ๐น๐ถ๐๐ in Writer, the enterprise AI platform trusted by Fortune 500.
sandsecurity.ai/blog/writeouโฆ
๐ซ It runs in a sandbox.
We treat this like a final security answer.
Here is our overview of what a sandbox is, and what it isn't:
sandsecurity.ai/blog/what-a-โฆ
Our Google Cloud VRP researchers don't miss! ๐ฅ Check out @terminatorLM's latest Looker research uncovering 9 novel cross-tenant vulns in Looker.
See how it was done: ๐
๐ซฃLeakyLooker: 1 Cross-tenant vulnerability? How about 9? (1/10)๐งต
Iโm incredibly proud to share LeakyLooker. I discovered 9 novel cross-tenant vulnerabilities in Google Cloudโs Looker Studio that broke fundamental design assumptions.
Here is how I broke tenant isolation: ๐
๐ซฃLeakyLooker: 1 Cross-tenant vulnerability? How about 9? (1/10)๐งต
Iโm incredibly proud to share LeakyLooker. I discovered 9 novel cross-tenant vulnerabilities in Google Cloudโs Looker Studio that broke fundamental design assumptions.
Here is how I broke tenant isolation: ๐
Disclosure (9/10)๐งต
Huge thanks to the Google VRP team. They handled these reports professionally and moved quickly to remediate them all. All issues are now fully patched. No customer action is required.
๐ LookOut: Novel Remote Code Execution & Internal Database Access vulnerabilities that I discovered in Google Looker
tenable.com/blog/google-lookโฆ
๐๐๐
I went to a meetup last night about cloud attacks and watched a talk by @terminatorLM about GCP, and it was SO GOOD!!
I came home with so many ideas and so much motivation.
Turns out itโs also on YouTube! please watch it, no matter which cloud youโre into๐คญ piped.video/nZWpDeY9p6g?siโฆ
๐ค HackedGPT: Unpacking 7 Vulnerabilities we discovered in ChatGPT
Following up on our work: Yarden Curiel, Moshe Bernstein, and I are proud to share the technical details of our ChatGPT research
tenable.com/blog/hackedgpt-nโฆ
๐ต๏ธ#๐ญ ๐๐ฆ๐ฅ๐๐๐ & #๐ญ๐ด ๐ช๐ข๐ฅ๐๐๐ช๐๐๐: ๐๐ผ๐ผ๐ด๐น๐ฒ ๐ฉ๐๐น๐ป๐ฒ๐ฟ๐ฎ๐ฏ๐ถ๐น๐ถ๐๐ ๐ฅ๐ฒ๐๐ฒ๐ฎ๐ฟ๐ฐ๐ต๐ฒ๐ฟ๐ ๐๐ฒ๐ฎ๐ฑ๐ฒ๐ฟ๐ฏ๐ผ๐ฎ๐ฟ๐ฑ
Massive personal milestone! ๐ I'm ranked #1 in Israel and #18 worldwide on the Google VRP! Thrilled to be a part of it.