Jon Bottarini (@jon_bottarini) header
Jon Bottarini (@jon_bottarini) profile picture

Jon Bottarini

@jon_bottarini

Safeguards, Threat Intel, etc @AnthropicAi - Cybersecurity and a little bit of everything. This is a personal account. Formerly: @Google

 Austin, TX
jonbottarini.com/
Joined September 2012
  • Tweets1.5K
  • Following760
  • Followers12.7K
115Media
Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)Photo by Jon Bottarini (@jon_bottarini)
We're excited to join forces with Gray Swan on this competition to strengthen Claude's defenses against prompt injection. This competition starts today - and there's $300k in prizes to those who are successful! 💸
Gray Swan AI (@GraySwanAI) profile pictureGray Swan AI
@GraySwanAIOct 15
Is prompt injection solved? Here’s $300,000 to find out. In partnership with @AnthropicAI, @Meta, @GoogleDeepMind, @Reflection_ai and our other frontier lab sponsors, @GraySwanAI is excited to announce the largest and most realistic AI red teaming challenge to date. Indirect Prompt Injection: Black ⚫
Image from Gray Swan AI (@GraySwanAI)
6
877
Jon Bottarini retweeted
Anthropic (@AnthropicAI) profile picture
Anthropic
@AnthropicAIOct 6
We’re expanding our Cyber Verification Program to give security professionals broader access to our most capable models. Through this program, verified security professionals can access Claude Mythos 5.1, Opus 5.5, and Sonnet 5.5 with safeguards designed for defensive work. We’re also opening up new tiers to allow for authorized offensive work, like penetration testing and red-teaming. anthropic.com/news/cyber-ver…
Expanding the Cyber Verification Program

Expanding the Cyber Verification Program

We’re launching a new, expanded version of our Cyber Verification Program (CVP), which makes advanced cyber capabilities and reduced blocking classifiers available to qualifying security profession...

anthropic.com
504
622
374
5,807
1,348,038
A personal update - a few weeks ago I joined Anthropic, where I'm working on Safeguards. Spending the last ~6 years at Google has been a true privilege. I'm deeply grateful for the people I had the opportunity to work alongside and the challenges we tackled together.
12
1
160
11,166
Jon Bottarini retweeted
skull (@brutecat) profile picture
skull
@brutecatJun 11
Hacking Google with A.I. for $500,000 brutecat.com/r/hacking-googl…
Hacking Google with A.I. for $500,000

Hacking Google with A.I. for $500,000

What happens when you unleash an AI across all of Google's infrastructure? 1,500 APIs, 3,600 keys, and $500,000 in bounties later, here's what I found.

brutecat.com
66
534
64
2,576
370,590
Vercel's (@vercel) bug bounty program is now live on HackerOne—up to $50k for finding WAF bypasses 👀 hackerone.com/vercel_platfor…
Image from Patrik Grobshäuser (@ITSecurityguard)
Guillermo Rauch (@rauchg) profile pictureGuillermo Rauch
@rauchg5 Dec 2025
Replying to @infosec_au @assetnote
DM’d you. You have a working repro for bypassing Cloudflare but not Vercel. Would love to correct the record or see the evidence.
5
16
2
248
48,604
Jon Bottarini retweeted
Dirk-jan (@_dirkjan) profile picture
Dirk-jan
@_dirkjan17 Sep 2025
I've been researching the Microsoft cloud for almost 7 years now. A few months ago that research resulted in the most impactful vulnerability I will probably ever find: a token validation flaw allowing me to get Global Admin in any Entra ID tenant. Blog: dirkjanm.io/obtaining-global…

One Token to rule them all - obtaining Global Admin in every Entra ID tenant via Actor tokens

While preparing for my Black Hat and DEF CON talks in July of this year, I found the most impactful Entra ID vulnerability that I will probably ever find. One that could have allowed me to compromise...

dirkjanm.io
138
899
112
3,171
482,306
Jon Bottarini retweeted
Awais Nazeer (@awais0x1) profile picture
Awais Nazeer@awais0x13 Sep 2025
Facebook Messenger Remote code execution Worth a $111,750 Video Poc Here: youtube.com/watch?v=wvywPUdT… Report: vulnano.com/2025/09/remote-c… @Google @intigriti
Image from Awais Nazeer (@awais0x1)
2
78
3
480
28,374
Jon Bottarini retweeted
James Kettle (@albinowax) profile picture
James Kettle
@albinowax6 Aug 2025
The whitepaper is live! Learn how to win the HTTP desync endgame... and why HTTP/1.1 needs to die: http1mustdie.com/
HTTP/1.1 Must Die

HTTP/1.1 Must Die

Upstream HTTP/1.1 is inherently insecure, and routinely exposes millions of websites to hostile takeover. Join the mission to kill HTTP/1.1 now

http1mustdie.com
18
240
14
744
87,211
Jon Bottarini retweeted
xEHLE (@xEHLE_) profile picture
xEHLE
@xEHLE_7 Jul 2025
New writeup: Early last month, @samwcyo, @sshell_, and I found a Django ORM injection in an online shooter game that let us steal cryptocurrency from the game's wallet. Read the blog post here: blog.p1.gs/writeup/2025/07/0…
Exploiting an ORM Injection to Steal Cryptocurrency from an Online Shooter

Exploiting an ORM Injection to Steal Cryptocurrency from an Online Shooter

blog writeup

blog.p1.gs
36
72
1
264
20,095
Jon Bottarini retweeted
skull (@brutecat) profile picture
skull
@brutecat9 Jun 2025
Leaking the phone number of any Google user brutecat.com/articles/leakin…

Leaking the phone number of any Google user

From rate limits to no limits: How IPv6's massive address space and a crafty botguard bypass left every Google user's phone number vulnerable

brutecat.com
123
175
13
962
82,246
Entire confession is absolutely wild. Talk about an insider threat risk...
Parker Conrad (@parkerconrad) profile pictureParker Conrad
@parkerconrad2 Apr 2025
Deel CEO and company founder @Bouazizalex personally orchestrated his company’s alleged spy scheme, the spy said in a full confession Alex allegedly recruited the spy, received the stolen info, and arranged payment via a person known only by their pseudonym: “The Watchman”
Image from Parker Conrad (@parkerconrad)
Image from Parker Conrad (@parkerconrad)
Image from Parker Conrad (@parkerconrad)
2
1,925
Jon Bottarini retweeted
skull (@brutecat) profile picture
skull
@brutecat13 Mar 2025
Disclosing YouTube Creator emails via Content ID for $20,000 brutecat.com/articles/youtub…
Disclosing YouTube Creator Emails for a $20k Bounty

Disclosing YouTube Creator Emails for a $20k Bounty

From creator privacy to phishing paradise: How a secret parameter could have exposed the private email addresses of monetized YouTube channels

brutecat.com
11
65
2
388
24,314
Jon Bottarini retweeted
skull (@brutecat) profile picture
skull
@brutecat12 Feb 2025
Leaking the email of any YouTube user for $10,000 brutecat.com/articles/leakin…
Leaking the email of any YouTube user for $10,000

Leaking the email of any YouTube user for $10,000

What could've been the largest data breach in the world - an attack chain on Google services to leak the email address of any YouTube channel

brutecat.com
20
140
9
866
64,107
Jon Bottarini retweeted
Sam Curry (@samwcyo) profile picture
Sam Curry@samwcyo23 Jan 2025
New blog post with @infosec_au: We found a vulnerability in Subaru where an attacker, with just a license plate, could retrieve the full location history, unlock, and start vehicles remotely. The issue was reported and patched. Full post here: samcurry.net/hacking-subaru
Hacking Subaru: Tracking and Controlling Cars via the STARLINK Admin Panel

Hacking Subaru: Tracking and Controlling Cars via the STARLINK Admin Panel

On November 20, 2024, Shubham Shah and I discovered a security vulnerability in Subaru’s STARLINK admin panel that gave us unrestricted access to all vehicles and customer accounts in the United...

samcurry.net
47
312
18
1,015
118,887
Jon Bottarini retweeted
daniel (@hackermondev) profile picture
daniel@hackermondev11 Oct 2024
1 Bug, $50K+ in bounties: how Zendesk left a backdoor in hundreds of companies #bugbountytips gist.github.com/hackermondev…
1 bug, $50,000+ in bounties, how Zendesk intentionally left a backdoor in hundreds of Fortune 500...

1 bug, $50,000+ in bounties, how Zendesk intentionally left a backdoor in hundreds of Fortune 500...

1 bug, $50,000+ in bounties, how Zendesk intentionally left a backdoor in hundreds of Fortune 500 companies - zendesk.md

gist.github.com
72
336
60
1,320
2,023,373
Jon Bottarini retweeted
Tanner (@itscachemoney) profile picture
Tanner@itscachemoney11 Oct 2024
Hey this was my bug! Thanks to Doppa for digging in and writing such a detailed post + PoC!
blueblue (@piedpiper1616) profile pictureblueblue@piedpiper161610 Oct 2024
Analyst CVE-2024-8698 on KeyCloak - huydoppa.hashnode.dev/analys…
2
16
4,558
Attacking UNIX Systems via CUPS, Part I evilsocket.net/2024/09/26/At…
Attacking UNIX Systems via CUPS, Part I | evilsocket

Attacking UNIX Systems via CUPS, Part I | evilsocket

Hello friends, this is the first of two, possibly three (if and when I have time to finish the Windows research) writeups. We will start with targeting GNU/Lin…

evilsocket.net
168
1,107
291
3,579
1,061,373
Jon Bottarini retweeted
cts🌸 (@gf_256) profile picture
cts🌸
@gf_25626 Sep 2024
Seems like it's in CUPS. GL;HF github.com/OpenPrinting/cups…
Review locking/multi-threading implementation · Issue #36 · OpenPrinting/cups-browsed

Review locking/multi-threading implementation · Issue #36 · OpenPrinting/cups-browsed

According to @evilsocket, cups-browsed can be held up for an extended period of time: The lock acquired here doesn't get unlocked until the IPP server has responded. A malicious IPP server can ...

github.com
cts🌸 (@gf_256) profile picturects🌸
@gf_25624 Sep 2024
Alright yall who's ready for Eternalblue: Linux Edition, dropping October 6?
10
32
20
439
263,627
Jon Bottarini retweeted
Sam Curry (@samwcyo) profile picture
Sam Curry@samwcyo26 Sep 2024
New writeup from @_specters_ and I: we're finally allowed to disclose a vulnerability reported to Kia which would've allowed an attacker to remotely control almost all vehicles made after 2013 using only the license plate. Full disclosure: samcurry.net/hacking-kia
84
956
132
3,504
345,139
Nation state behavior but you only have $20 - Taking over the mobi TLD WHOIS server: labs.watchtowr.com/we-spent-…
We Spent $20 To Achieve RCE And Accidentally Became The Admins Of .MOBI

We Spent $20 To Achieve RCE And Accidentally Became The Admins Of .MOBI

Welcome back to another watchTowr Labs blog. Brace yourselves, this is one of our most astounding discoveries. Summary What started out as a bit of fun between colleagues while avoiding the Vegas...

labs.watchtowr.com
9
52
4,197
Load more