I had a great discussion with Matt Kozloski on today's Winslow Technology Group webinar. We discussed pentesting and autonomous pentesting with @Horizon3ai's NodeZero. #pentesting #offensivesecurity #redteaming #autonomouspentesting #nodezero
www.youtube.com/watch?v=OX7G...
www.youtube.com/watch?v=OX7G...
Continuous Security Assessment with Horizon3 AI
YouTube video by Winslow Technology Group
www.youtube.com
November 13, 2024 at 8:27 PM
I had a great discussion with Matt Kozloski on today's Winslow Technology Group webinar. We discussed pentesting and autonomous pentesting with @Horizon3ai's NodeZero. #pentesting #offensivesecurity #redteaming #autonomouspentesting #nodezero
www.youtube.com/watch?v=OX7G...
www.youtube.com/watch?v=OX7G...
Pleased to share this report I contributed to on Red Teaming in the Public Interest with Data & Society and the AI Risk and Vulnerability Alliance.
We did interviews with folks across the generative AI ecosystem.
datasociety.net/library/red-...
#generativeai #llms #aiethics #aisafety #redteaming
#generativeai #llms #aiethics #aisafety #redteaming
Red-Teaming in the Public Interest
This report offers a vision for red-teaming in the public interest: a process that goes beyond system-centric testing of already built systems to consider the full range of ways the public can be invo...
datasociety.net
February 11, 2025 at 6:14 PM
Pleased to share this report I contributed to on Red Teaming in the Public Interest with Data & Society and the AI Risk and Vulnerability Alliance.
We did interviews with folks across the generative AI ecosystem.
datasociety.net/library/red-...
#generativeai #llms #aiethics #aisafety #redteaming
#generativeai #llms #aiethics #aisafety #redteaming
🚀 The @salesforce.com #Responsible #AI & Tech team is looking for an experienced #ResponsibleAI Data Scientist w/ expertise in #ethical #RedTeaming. Collaborate with security, engineering, data science, & AI Research teams. Join us! #hiring #dreamJob
salesforce.wd12.myworkdayjobs.com/External_Car...
salesforce.wd12.myworkdayjobs.com/External_Car...
Senior or Principal Data Scientist - Technical AI Ethicist
To get the best candidate experience, please consider applying for a maximum of 3 roles within 12 months to ensure you are not duplicating efforts. Job Category Data Job Details About Salesforce We’re...
salesforce.wd12.myworkdayjobs.com
March 20, 2025 at 10:39 PM
🚀 The @salesforce.com #Responsible #AI & Tech team is looking for an experienced #ResponsibleAI Data Scientist w/ expertise in #ethical #RedTeaming. Collaborate with security, engineering, data science, & AI Research teams. Join us! #hiring #dreamJob
salesforce.wd12.myworkdayjobs.com/External_Car...
salesforce.wd12.myworkdayjobs.com/External_Car...
It is an honor for me that my talk #Hacking the Skies - #Satellite #RedTeaming is been accepted at @RedTeamVillage at @defcon.bsky.social ... If you are planning to be there stop-by it's going to be fun!! ☺️
May 22, 2024 at 4:21 AM
It is an honor for me that my talk #Hacking the Skies - #Satellite #RedTeaming is been accepted at @RedTeamVillage at @defcon.bsky.social ... If you are planning to be there stop-by it's going to be fun!! ☺️
We are proud to introduce #dAWShund to the world: a framework for putting a leash on naughty AWS permissions. dAWShund helps blue and red teams find resources in #AWS, evaluate their access levels and visualize the relationships between them.
falconforce.nl/dawshund-fra...
#blueteaming #redteaming
falconforce.nl/dawshund-fra...
#blueteaming #redteaming
April 11, 2025 at 11:55 AM
We are proud to introduce #dAWShund to the world: a framework for putting a leash on naughty AWS permissions. dAWShund helps blue and red teams find resources in #AWS, evaluate their access levels and visualize the relationships between them.
falconforce.nl/dawshund-fra...
#blueteaming #redteaming
falconforce.nl/dawshund-fra...
#blueteaming #redteaming
Why not pop this in your diary before you break for the holidays? New BRAID/IDI research talk: Red Teaming and Operationalising Responsibility with @alex-taylor.bsky.social 30 Jan, online/in person #Edinburgh Find out more/book here: bit.ly/4fiN9mc @designinf.bsky.social #ResponsibleAI #RedTeaming
BRAID x IDI Hybrid Seminar - Alex Taylor
Thursday, 30th January 2025 [Inspace / Zoom]
bit.ly
December 18, 2024 at 11:47 AM
Why not pop this in your diary before you break for the holidays? New BRAID/IDI research talk: Red Teaming and Operationalising Responsibility with @alex-taylor.bsky.social 30 Jan, online/in person #Edinburgh Find out more/book here: bit.ly/4fiN9mc @designinf.bsky.social #ResponsibleAI #RedTeaming
I'm looking for #pentesting and #redteaming accounts. I'm hoping the offensive tooling #cybersecurity community will move away from Twitter so I can stop going there.
October 21, 2024 at 12:27 AM
I'm looking for #pentesting and #redteaming accounts. I'm hoping the offensive tooling #cybersecurity community will move away from Twitter so I can stop going there.
#RedTeaming nicht erwünscht? Wir sind so am Arsch hier in Neuland.
May 18, 2025 at 4:34 PM
#RedTeaming nicht erwünscht? Wir sind so am Arsch hier in Neuland.
Google Just Admitted Gemini Hacked Three Real Companies. It Wasn't Even Supposed to Have Internet Access
www.thefluxread.com/2026/09/goog...
#AgenticAI #Cybersecurity #AISafety #RedTeaming #CloudInfrastructure #InfoSec #EnterpriseTech #GoogleGemini #DevOps #TheFluxRead
www.thefluxread.com/2026/09/goog...
#AgenticAI #Cybersecurity #AISafety #RedTeaming #CloudInfrastructure #InfoSec #EnterpriseTech #GoogleGemini #DevOps #TheFluxRead
Google Just Admitted Gemini Hacked Three Real Companies. It Wasn't Even Supposed to Have Internet Access
Google confirmed a Gemini red-teaming model accessed 3 live corporate systems due to network egress misconfigurations. Read our full technical breakdo
www.thefluxread.com
September 22, 2026 at 4:34 PM
Google Just Admitted Gemini Hacked Three Real Companies. It Wasn't Even Supposed to Have Internet Access
www.thefluxread.com/2026/09/goog...
#AgenticAI #Cybersecurity #AISafety #RedTeaming #CloudInfrastructure #InfoSec #EnterpriseTech #GoogleGemini #DevOps #TheFluxRead
www.thefluxread.com/2026/09/goog...
#AgenticAI #Cybersecurity #AISafety #RedTeaming #CloudInfrastructure #InfoSec #EnterpriseTech #GoogleGemini #DevOps #TheFluxRead
CDAO-sponsored #redteaming efforts identifies "over 800 findings of potential vulnerabilities and biases" related to the use of #LLM for #clinical note summarization & #medical advisory chatbot. #AI #DefenseAI
cc @aisupremacy.bsky.social
See: www.defense.gov/News/Release...
cc @aisupremacy.bsky.social
See: www.defense.gov/News/Release...
CDAO Sponsors Crowdsourced AI Assurance Pilot in the Context of Military Medicine
The Chief Digital and Artificial Intelligence Office has successfully concluded a Crowdsourced AI Red-Teaming Assurance Program pilot focused on the use of Large-Language Model chatbots in the context
www.defense.gov
January 3, 2025 at 7:18 AM
CDAO-sponsored #redteaming efforts identifies "over 800 findings of potential vulnerabilities and biases" related to the use of #LLM for #clinical note summarization & #medical advisory chatbot. #AI #DefenseAI
cc @aisupremacy.bsky.social
See: www.defense.gov/News/Release...
cc @aisupremacy.bsky.social
See: www.defense.gov/News/Release...
A safety system faces a unique problem at cold start: no history, no context, no baseline.
The first message can become the frame that shapes everything that follows.
When Turn 1 is the only signal, where does safety end and trajectory formation begin?
#AISafety #RedTeaming #PromptInjection
The first message can become the frame that shapes everything that follows.
When Turn 1 is the only signal, where does safety end and trajectory formation begin?
#AISafety #RedTeaming #PromptInjection
September 27, 2026 at 9:02 AM
A safety system faces a unique problem at cold start: no history, no context, no baseline.
The first message can become the frame that shapes everything that follows.
When Turn 1 is the only signal, where does safety end and trajectory formation begin?
#AISafety #RedTeaming #PromptInjection
The first message can become the frame that shapes everything that follows.
When Turn 1 is the only signal, where does safety end and trajectory formation begin?
#AISafety #RedTeaming #PromptInjection
pentestlaboratories.com/2021/05/17/a... - awesome write up on various AMSI bypass methods
#cybersecurity #redteaming #infosec
#cybersecurity #redteaming #infosec
AMSI Bypass Methods
Microsoft has developed AMSI (Antimalware Scan Interface) as a method to defend against common malware execution and protect the end user. By default windows defender interacts with the AMSI API to…
pentestlaboratories.com
November 21, 2024 at 8:29 PM
pentestlaboratories.com/2021/05/17/a... - awesome write up on various AMSI bypass methods
#cybersecurity #redteaming #infosec
#cybersecurity #redteaming #infosec
What if an AI attack does not target a weakness directly?
What if it changes the conditions under which the next interaction occurs?
The interesting part is not the individual prompt. It is what happens when the interaction itself becomes part of the attack surface.
#AISafety #RedTeaming #RedTeam
What if it changes the conditions under which the next interaction occurs?
The interesting part is not the individual prompt. It is what happens when the interaction itself becomes part of the attack surface.
#AISafety #RedTeaming #RedTeam
September 25, 2026 at 6:11 PM
What if an AI attack does not target a weakness directly?
What if it changes the conditions under which the next interaction occurs?
The interesting part is not the individual prompt. It is what happens when the interaction itself becomes part of the attack surface.
#AISafety #RedTeaming #RedTeam
What if it changes the conditions under which the next interaction occurs?
The interesting part is not the individual prompt. It is what happens when the interaction itself becomes part of the attack surface.
#AISafety #RedTeaming #RedTeam
In 2023, researchers published AutoDAN: a pipeline where an AI automatically generates jailbreak prompts for another, iterates on failures, and improves continuously.
The attack is automated. The defense is manual.
That asymmetry is the problem.
#SPCResearchSeries #AutoDAN #AIJailbreak #RedTeaming
The attack is automated. The defense is manual.
That asymmetry is the problem.
#SPCResearchSeries #AutoDAN #AIJailbreak #RedTeaming
September 25, 2026 at 10:31 AM
In 2023, researchers published AutoDAN: a pipeline where an AI automatically generates jailbreak prompts for another, iterates on failures, and improves continuously.
The attack is automated. The defense is manual.
That asymmetry is the problem.
#SPCResearchSeries #AutoDAN #AIJailbreak #RedTeaming
The attack is automated. The defense is manual.
That asymmetry is the problem.
#SPCResearchSeries #AutoDAN #AIJailbreak #RedTeaming
On a slightly different note, I'm stoked to say that I stated AND completed a self paced online course folks. #RedTeaming #LLM apps in the bag and I had so much fun feeling out.
#AISecurity
#AISecurity
February 24, 2025 at 8:56 AM
On a slightly different note, I'm stoked to say that I stated AND completed a self paced online course folks. #RedTeaming #LLM apps in the bag and I had so much fun feeling out.
#AISecurity
#AISecurity
This is pretty cool! #AI #RedTeaming Playbook: Complete methodology, from recon to exploitation! Focuses on the agentic layer (models, tools, data) with hands-on, real-world examples. Uncover and remediate application risks fast. #AIRedTeaming #Cybersecurity
cybersec.pillar.security/s/agentic-ai...
cybersec.pillar.security/s/agentic-ai...
October 13, 2025 at 5:26 PM
This is pretty cool! #AI #RedTeaming Playbook: Complete methodology, from recon to exploitation! Focuses on the agentic layer (models, tools, data) with hands-on, real-world examples. Uncover and remediate application risks fast. #AIRedTeaming #Cybersecurity
cybersec.pillar.security/s/agentic-ai...
cybersec.pillar.security/s/agentic-ai...
github.com/WithSecureLa... - Pretty neat new Cloud security Virtual machine with allot of great pre installed tools.
#cloudsecurity #cybersecurity #redteaming #security
#cloudsecurity #cybersecurity #redteaming #security
GitHub - WithSecureLabs/cloud-security-vm: Ansible/Vagrant/Packer files to create a virtual machine with the tooling needed to perform cloud security assessments
Ansible/Vagrant/Packer files to create a virtual machine with the tooling needed to perform cloud security assessments - WithSecureLabs/cloud-security-vm
github.com
November 14, 2024 at 3:29 PM
github.com/WithSecureLa... - Pretty neat new Cloud security Virtual machine with allot of great pre installed tools.
#cloudsecurity #cybersecurity #redteaming #security
#cloudsecurity #cybersecurity #redteaming #security
Hello Rustaceans! Our technical director @raptor is back at it.
In this second installment of our #rust series, “An offensive Rust encore”, he will guide you in bringing your skills to the next level by using a new PoC #redteaming tool as an excuse […]
In this second installment of our #rust series, “An offensive Rust encore”, he will guide you in bringing your skills to the next level by using a new PoC #redteaming tool as an excuse […]
Original post on infosec.exchange
infosec.exchange
December 10, 2024 at 8:00 AM
Hello Rustaceans! Our technical director @raptor is back at it.
In this second installment of our #rust series, “An offensive Rust encore”, he will guide you in bringing your skills to the next level by using a new PoC #redteaming tool as an excuse […]
In this second installment of our #rust series, “An offensive Rust encore”, he will guide you in bringing your skills to the next level by using a new PoC #redteaming tool as an excuse […]
It looks like there really will be a top-notch tool this Monday! Here is a tool worth checking out; it employs a highly unusual technique: github.com/r3xmax/Phant... #hacker #tool #evasion #mitre #github #Monday #hacking #redteam #context #redteaming
GitHub - r3xmax/PhantomCtx: Activation Context Hijacking Evasion Tool
Activation Context Hijacking Evasion Tool. Contribute to r3xmax/PhantomCtx development by creating an account on GitHub.
github.com
June 22, 2026 at 4:29 PM
Regardez vos emails pour de grandes annonces! Levez l'ancre et baissez les voiles, Northsec approche à toute vitesse! ⚓⛴️🚢
S'inscrire à l'infolettre: nsec.io
#northsec #nsec2025 #infosec #ctf #cfp #conference #hackerspace #hackers #devops #redteaming #blueteam #security #event #montreal #mtl
S'inscrire à l'infolettre: nsec.io
#northsec #nsec2025 #infosec #ctf #cfp #conference #hackerspace #hackers #devops #redteaming #blueteam #security #event #montreal #mtl
NorthSec 2025
nsec.io
March 12, 2025 at 1:08 AM
Look at your emails for big news! Raise the anchor and set sail, Northsec is straight ahead! ⚓⛴️🚢
Sign up to the newsletter: nsec.io
#northsec #nsec2025 #infosec #ctf #cfp #conference #hackerspace #hackers #devops #redteaming #blueteam #security #event #montreal #mtl
Sign up to the newsletter: nsec.io
#northsec #nsec2025 #infosec #ctf #cfp #conference #hackerspace #hackers #devops #redteaming #blueteam #security #event #montreal #mtl
NorthSec 2025
nsec.io
March 12, 2025 at 1:08 AM
Penetration Testing ≠ Red Team Testing.
#PenTesting is overt, short-term & asset-focused.
#RedTeaming is covert, long-term & simulates real-world APTs across vectors.
Know the difference. Choose smarter.
#CyberSecurity #InfoSec #ThreatDetection #SOC #MITREATTACK
#PenTesting is overt, short-term & asset-focused.
#RedTeaming is covert, long-term & simulates real-world APTs across vectors.
Know the difference. Choose smarter.
#CyberSecurity #InfoSec #ThreatDetection #SOC #MITREATTACK
May 23, 2025 at 6:01 AM
Penetration Testing ≠ Red Team Testing.
#PenTesting is overt, short-term & asset-focused.
#RedTeaming is covert, long-term & simulates real-world APTs across vectors.
Know the difference. Choose smarter.
#CyberSecurity #InfoSec #ThreatDetection #SOC #MITREATTACK
#PenTesting is overt, short-term & asset-focused.
#RedTeaming is covert, long-term & simulates real-world APTs across vectors.
Know the difference. Choose smarter.
#CyberSecurity #InfoSec #ThreatDetection #SOC #MITREATTACK
New paper out today
Claude refused every frame attempt across 22 turns, but zero platform-level interventions fired throughout.
Turn-level safety ≠ trajectory-level safety.
The gap between them is the research question.
doi.org/10.5281/zeno...
#AISafety #RedTeaming #Claude #Alignment #SPC
Claude refused every frame attempt across 22 turns, but zero platform-level interventions fired throughout.
Turn-level safety ≠ trajectory-level safety.
The gap between them is the research question.
doi.org/10.5281/zeno...
#AISafety #RedTeaming #Claude #Alignment #SPC
When an AI Recognizes the Frame: Trajectory-Level Safety, Symbolic Interaction, and the Limits of Turn-Level Monitoring in a Claude Test
Abstract Contemporary AI safety systems are predominantly evaluated at the level of individual outputs: whether a given request or response crosses a defined policy boundary. This unit of analysis bec...
doi.org
September 18, 2026 at 9:21 AM
New paper out today
Claude refused every frame attempt across 22 turns, but zero platform-level interventions fired throughout.
Turn-level safety ≠ trajectory-level safety.
The gap between them is the research question.
doi.org/10.5281/zeno...
#AISafety #RedTeaming #Claude #Alignment #SPC
Claude refused every frame attempt across 22 turns, but zero platform-level interventions fired throughout.
Turn-level safety ≠ trajectory-level safety.
The gap between them is the research question.
doi.org/10.5281/zeno...
#AISafety #RedTeaming #Claude #Alignment #SPC
AI-Infra-Guard: Open-source security scanner for AI systems
📖 Read more: www.helpnetsecurity.com/2026/09/09/a...
#AIsecurity #MCP #infosec #opensource #redteaming #cybersecurity #cybersecuritynews
📖 Read more: www.helpnetsecurity.com/2026/09/09/a...
#AIsecurity #MCP #infosec #opensource #redteaming #cybersecurity #cybersecuritynews
AI-Infra-Guard: Open-source security scanner for AI systems - Help Net Security
Tencent AI-Infra-Guard scanner flags clean AI skills up to 18.67% of the time, and the open-source build ships with no login layer at all.
www.helpnetsecurity.com
September 9, 2026 at 1:02 PM
AI-Infra-Guard: Open-source security scanner for AI systems
📖 Read more: www.helpnetsecurity.com/2026/09/09/a...
#AIsecurity #MCP #infosec #opensource #redteaming #cybersecurity #cybersecuritynews
📖 Read more: www.helpnetsecurity.com/2026/09/09/a...
#AIsecurity #MCP #infosec #opensource #redteaming #cybersecurity #cybersecuritynews