Russel Van Tuyl
@russelvantuyl.bsky.social
Tech enthusiasts, offensive cybersecurity professional, AI student
Reposted by Russel Van Tuyl
🚨 New blog post alert!
@xpnsec.com drops knowledge on LLM security w/ his latest post showing how attackers can by pass LLM WAFs by confusing the tokenization process to smuggle tokens to back-end LLMs.
Read more: ghst.ly/4koUJiz
@xpnsec.com drops knowledge on LLM security w/ his latest post showing how attackers can by pass LLM WAFs by confusing the tokenization process to smuggle tokens to back-end LLMs.
Read more: ghst.ly/4koUJiz
Tokenization Confusion - SpecterOps
Meta's Prompt Guard 2 aims to prevent prompt injection. This post looks at how much knowledge of ML we need to be effective at testing these LLM WAFs.
ghst.ly
June 3, 2025 at 5:44 PM
🚨 New blog post alert!
@xpnsec.com drops knowledge on LLM security w/ his latest post showing how attackers can by pass LLM WAFs by confusing the tokenization process to smuggle tokens to back-end LLMs.
Read more: ghst.ly/4koUJiz
@xpnsec.com drops knowledge on LLM security w/ his latest post showing how attackers can by pass LLM WAFs by confusing the tokenization process to smuggle tokens to back-end LLMs.
Read more: ghst.ly/4koUJiz
Reposted by Russel Van Tuyl
BIG NEWS: SpecterOps raises $75M Series B to strengthen identity security! Led by Insight Partners with Ansa Capital, M12, Ballistic Ventures, Decibel, and Cisco Investments. ghst.ly/seriesb
#IdentitySecurity #CyberSecurity
(1/6)
#IdentitySecurity #CyberSecurity
(1/6)
March 5, 2025 at 5:33 PM
BIG NEWS: SpecterOps raises $75M Series B to strengthen identity security! Led by Insight Partners with Ansa Capital, M12, Ballistic Ventures, Decibel, and Cisco Investments. ghst.ly/seriesb
#IdentitySecurity #CyberSecurity
(1/6)
#IdentitySecurity #CyberSecurity
(1/6)
Come join us, there isn’t a better place to work and show your technical excellence surrounded by the industry’s best if you ask me!
Our Consulting Services team is growing! 🙌
We are now hiring Consultants and Senior Consultants to join the team as operators, trainers, and program developers.
Learn more & apply today! ghst.ly/3PBmGFZ
We are now hiring Consultants and Senior Consultants to join the team as operators, trainers, and program developers.
Learn more & apply today! ghst.ly/3PBmGFZ
January 16, 2025 at 2:08 PM
Come join us, there isn’t a better place to work and show your technical excellence surrounded by the industry’s best if you ask me!
Interesting choice of words in the title. The model itself wasn’t hijacked but the code in the repository was through pull requests with code injection via branch names 🤯 www.bleepingcomputer.com/news/securit...
Ultralytics AI model hijacked to infect thousands with cryptominer
The popular Ultralytics YOLO11 AI model was compromised in a supply chain attack to deploy cryptominers on devices running versions 8.3.41 and 8.3.42 from the Python Package Index (PyPI)
www.bleepingcomputer.com
December 20, 2024 at 9:49 PM
Interesting choice of words in the title. The model itself wasn’t hijacked but the code in the repository was through pull requests with code injection via branch names 🤯 www.bleepingcomputer.com/news/securit...
The OWASP LLM and Generative AI Security Project has launched the Agentic Security Initiative to address security challenges posed by autonomous AI agents. This effort focuses on developing best practices to secure agentic LLM and Generative AI applications. genai.owasp.org/2024/12/15/a...
Announcing the OWASP LLM and Gen AI Security Project Initiative for Securing Agentic Applications - OWASP Top 10 for LLM & Generative AI Security
The OWASP LLM and Generative AI Security Project is thrilled to announce the launch of the Agentic Security Initiative designed to tackle the unique security challenges posed by Autonomous AI agents. ...
genai.owasp.org
December 20, 2024 at 2:26 PM
The OWASP LLM and Generative AI Security Project has launched the Agentic Security Initiative to address security challenges posed by autonomous AI agents. This effort focuses on developing best practices to secure agentic LLM and Generative AI applications. genai.owasp.org/2024/12/15/a...
Reposted by Russel Van Tuyl
First dataset for the new @huggingface.bsky.social @bsky.app community organisation: one-million-bluesky-posts 🦋
📊 1M public posts from Bluesky's firehose API
🔍 Includes text, metadata, and language predictions
🔬 Perfect to experiment with using ML for Bluesky 🤗
huggingface.co/datasets/blu...
📊 1M public posts from Bluesky's firehose API
🔍 Includes text, metadata, and language predictions
🔬 Perfect to experiment with using ML for Bluesky 🤗
huggingface.co/datasets/blu...
bluesky-community/one-million-bluesky-posts · Datasets at Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
November 26, 2024 at 1:50 PM
First dataset for the new @huggingface.bsky.social @bsky.app community organisation: one-million-bluesky-posts 🦋
📊 1M public posts from Bluesky's firehose API
🔍 Includes text, metadata, and language predictions
🔬 Perfect to experiment with using ML for Bluesky 🤗
huggingface.co/datasets/blu...
📊 1M public posts from Bluesky's firehose API
🔍 Includes text, metadata, and language predictions
🔬 Perfect to experiment with using ML for Bluesky 🤗
huggingface.co/datasets/blu...
Reposted by Russel Van Tuyl
If you find yourself with too much free time over the (long) weekend / holidays, I have ~3h Building an LLM from the Ground Up workshop on YouTube that may come in handy: m.youtube.com/watch?v=quh7...
Building LLMs from the Ground Up: A 3-hour Coding Workshop
YouTube video by Sebastian Raschka
m.youtube.com
November 27, 2024 at 4:39 AM
If you find yourself with too much free time over the (long) weekend / holidays, I have ~3h Building an LLM from the Ground Up workshop on YouTube that may come in handy: m.youtube.com/watch?v=quh7...
I really enjoyed reading this paper from OpenAI. If you perform AI assessments, you should read it.
I thought they laid out a pragmatic approach to evaluating AI models that should be a component of any organization's assessment methodology.
cdn.openai.com/papers/opena...
I thought they laid out a pragmatic approach to evaluating AI models that should be a component of any organization's assessment methodology.
cdn.openai.com/papers/opena...
cdn.openai.com
November 26, 2024 at 6:21 PM
I really enjoyed reading this paper from OpenAI. If you perform AI assessments, you should read it.
I thought they laid out a pragmatic approach to evaluating AI models that should be a component of any organization's assessment methodology.
cdn.openai.com/papers/opena...
I thought they laid out a pragmatic approach to evaluating AI models that should be a component of any organization's assessment methodology.
cdn.openai.com/papers/opena...
Reposted by Russel Van Tuyl
I couldn't find any PowerShell examples of encrypting/decrypting data w/ Azure Key Vault keys, so I made some:
Protect-StringWithAzureKeyVaultKey
Unprotect-StringWithAzureKeyVaultKey
github.com/BloodHoundAD...
Explanatory blog post coming soon.
Protect-StringWithAzureKeyVaultKey
Unprotect-StringWithAzureKeyVaultKey
github.com/BloodHoundAD...
Explanatory blog post coming soon.
Add key vault cryptographic op funcs · BloodHoundAD/BARK@e1c82a1
github.com
November 19, 2024 at 12:24 AM
I couldn't find any PowerShell examples of encrypting/decrypting data w/ Azure Key Vault keys, so I made some:
Protect-StringWithAzureKeyVaultKey
Unprotect-StringWithAzureKeyVaultKey
github.com/BloodHoundAD...
Explanatory blog post coming soon.
Protect-StringWithAzureKeyVaultKey
Unprotect-StringWithAzureKeyVaultKey
github.com/BloodHoundAD...
Explanatory blog post coming soon.
This State of Generative AI report from Menlo Ventures provided some good insights on where cybersecurity professionals might look for risk in terms of assessments and research.
menlovc.com/2024-the-sta...
menlovc.com/2024-the-sta...
2024: The State of Generative AI in the Enterprise - Menlo Ventures
The enterprise AI landscape is being rewritten in real time. We surveyed 600 U.S. enterprise IT decision-makers to reveal the emerging winners and losers.
menlovc.com
November 22, 2024 at 11:07 PM
This State of Generative AI report from Menlo Ventures provided some good insights on where cybersecurity professionals might look for risk in terms of assessments and research.
menlovc.com/2024-the-sta...
menlovc.com/2024-the-sta...
This looks like a fun challenge to evade prompt injection defenses microsoft.github.io/llmail-inject/
LLMail-Inject: Adaptive Prompt Injection Challenge
Code for the platform architecture and LLM application used during the Hack My Email SaTML 2024 Competition
microsoft.github.io
November 22, 2024 at 12:54 PM
This looks like a fun challenge to evade prompt injection defenses microsoft.github.io/llmail-inject/
Love this, hoping to do something similar with our assessments.
It’s always awesome when we (@CISAGov) gets to release a red team report that we worked on, and today is another one of those days!
Go check out our latest report and hopefully you can apply some of the same lessons to your environment!
www.cisa.gov/news-events/...
Go check out our latest report and hopefully you can apply some of the same lessons to your environment!
www.cisa.gov/news-events/...
November 21, 2024 at 3:57 PM
Love this, hoping to do something similar with our assessments.
“Cybersecurity professionals and ethical hackers need to understand the darker side of hacking to better prepare for potential threats. Unfiltered AI models can provide insights into hacking methodologies and scenarios typically censored, aiding in the development of robust cybersecurity measures.”
November 20, 2024 at 3:49 PM
“Cybersecurity professionals and ethical hackers need to understand the darker side of hacking to better prepare for potential threats. Unfiltered AI models can provide insights into hacking methodologies and scenarios typically censored, aiding in the development of robust cybersecurity measures.”
Reposted by Russel Van Tuyl
Great read on how "China Hawks are Manufacturing an AI Arms Race", a concerning trend for anyone advocating for regulation and safety of AI. An arms-race narrative would ensure an unfettered and unregulated development of AI in almost all contexts.
garrisonlovely.substack.com/p/china-hawk...
garrisonlovely.substack.com/p/china-hawk...
China Hawks are Manufacturing an AI Arms Race
An influential congressional commission is calling for a militarized race to build superintelligent AI based on threadbare evidence
garrisonlovely.substack.com
November 20, 2024 at 12:46 PM
Great read on how "China Hawks are Manufacturing an AI Arms Race", a concerning trend for anyone advocating for regulation and safety of AI. An arms-race narrative would ensure an unfettered and unregulated development of AI in almost all contexts.
garrisonlovely.substack.com/p/china-hawk...
garrisonlovely.substack.com/p/china-hawk...
Conflicted about this post on prompt injection for multi modal models. Turns out they read instructions and follow them 🤯. All data from input should be untrusted from the system and user prompts and not processed as one. www.lakera.ai/blog/visual-...
The Beginner's Guide to Visual Prompt Injections: Invisibility Cloaks, Cannibalistic Adverts, and Robot Women | Lakera – Protecting AI teams that disrupt the world.
Learn about visual prompt injections, their appearance, and top defense strategies against these attacks.
www.lakera.ai
November 18, 2024 at 7:55 PM
Conflicted about this post on prompt injection for multi modal models. Turns out they read instructions and follow them 🤯. All data from input should be untrusted from the system and user prompts and not processed as one. www.lakera.ai/blog/visual-...
Has anyone fired up this PentestGPT during an actual assessment? I did like their pentesting task tree (PTT) to track the status of tests. www.usenix.org/conference/u...
PentestGPT: Evaluating and Harnessing Large Language Models for Automated Penetration Testing | USENIXusenix_logo_notag_white
www.usenix.org
November 18, 2024 at 3:40 PM
Has anyone fired up this PentestGPT during an actual assessment? I did like their pentesting task tree (PTT) to track the status of tests. www.usenix.org/conference/u...
Can the sandbox reach the Internet 👀? Asking for a friend.
Are we going for security through obscurity by keeping system prompts private?
Are we going for security through obscurity by keeping system prompts private?
I thought we stopped doing this?
www.darkreading.com/cloud-securi...
www.darkreading.com/cloud-securi...
ChatGPT Exposes Its Instructions, Knowledge & OS Files
According to Mozilla, users have a lot more power to manipulate ChatGPT than they might realize. OpenAI hopes those manipulations remain within a clearly delineated sandbox.
www.darkreading.com
November 18, 2024 at 2:57 PM
Can the sandbox reach the Internet 👀? Asking for a friend.
Are we going for security through obscurity by keeping system prompts private?
Are we going for security through obscurity by keeping system prompts private?
Great paper, especially like the parts on data & model provenance and the Supply-chain Levels for Software Artifacts. These could really make offensive security operations challenging.
We are happy to publish a whitepaper on how we're thinking on securing the AI supply chain both internally and for OSS. This is a culmination of nearly a year of thinking about this space, from people working on AI or security, across multiple Google PAs.
research.google/pubs/securin...
research.google/pubs/securin...
Securing the AI Software Supply Chain
research.google
November 18, 2024 at 1:38 PM
Great paper, especially like the parts on data & model provenance and the Supply-chain Levels for Software Artifacts. These could really make offensive security operations challenging.
Reposted by Russel Van Tuyl
Microsoft’s Orca Agent-Instruct dataset has been released!
Permissively licensed 1M synthetic instruction pairs covering different capabilities, such as text editing, creative writing, coding, reading comprehension
Paper: arxiv.org/abs/2407.03502
Dataset: huggingface.co/datasets/mic...
Permissively licensed 1M synthetic instruction pairs covering different capabilities, such as text editing, creative writing, coding, reading comprehension
Paper: arxiv.org/abs/2407.03502
Dataset: huggingface.co/datasets/mic...
AgentInstruct: Toward Generative Teaching with Agentic Flows
Synthetic data is becoming increasingly important for accelerating the development of language models, both large and small. Despite several successful use cases, researchers also raised concerns arou...
arxiv.org
November 17, 2024 at 6:39 AM
Microsoft’s Orca Agent-Instruct dataset has been released!
Permissively licensed 1M synthetic instruction pairs covering different capabilities, such as text editing, creative writing, coding, reading comprehension
Paper: arxiv.org/abs/2407.03502
Dataset: huggingface.co/datasets/mic...
Permissively licensed 1M synthetic instruction pairs covering different capabilities, such as text editing, creative writing, coding, reading comprehension
Paper: arxiv.org/abs/2407.03502
Dataset: huggingface.co/datasets/mic...
Nice write up from Mandiant on some practical use cases for leveraging AI to help red team operations. What are some other use cases ya’ll are thinking of? cloud.google.com/blog/topics/...
AI Enhancing Your Adversarial Emulation | Google Cloud Blog
Learn how Mandiant Red Team is using Gemini and LLMs for adversarial emulation and defense.
cloud.google.com
November 17, 2024 at 1:16 AM
Nice write up from Mandiant on some practical use cases for leveraging AI to help red team operations. What are some other use cases ya’ll are thinking of? cloud.google.com/blog/topics/...