Pangea Security

Pangea Security AI

Pangea Security AI

Pangea Security AI

Pangea Security AI

Pangea Security AI

Pangea Security AI

Pangea Security AI

Pangea Security AI

Pangea Security AI

Pangea Security AI

While most companies discover AI security gaps after deployment, @pangeacyber has pioneered solutions to identify these blindspots from the start Thank you @IndieHackers and @JamesOfTheDrum for the o

While most companies discover AI security gaps after deployment, @pangeacyber has pioneered solutions to identify these blindspots from the start Thank you @IndieHackers and @JamesOfTheDrum for the opportunity to share our Founder and CEO Oliver Friedrichs's journey toward building a more secure AI future! Read: https://t.co/CLG4ZVpaF5 #AISecurity #CyberSecurity #AIDR

Concerned about securing employee AI use and delivering safe AI apps quickly? New research shows vulnerabilities in LLMs that open the door to data loss, manipulation and unauthorized access. Join u

Concerned about securing employee AI use and delivering safe AI apps quickly? New research shows vulnerabilities in LLMs that open the door to data loss, manipulation and unauthorized access. Join us next week, Sept 16 for a webcast where we’ll: -Walk through exploits like Scamlexity, GitHub Copilot RCE -CVE-2025-53773, LegalPwn and more -Show live demos -Share strategies to stay ahead Register here → https://t.co/tDyXBqwbo1 with @MrJoeyMelo, AI Red Teaming Specialist at @pangeacyber #AIDR #AISecurity #PII #PromptInjection #LegalPwn

From jailbreaks to prompt injection, attackers are exploiting LLMs in ways traditional security misses. This quick, 30 minute monthly webinar series from @pangeacyber Labs and @MrJoeyMelo covers real-

From jailbreaks to prompt injection, attackers are exploiting LLMs in ways traditional security misses. This quick, 30 minute monthly webinar series from @pangeacyber Labs and @MrJoeyMelo covers real-world exploits like Scamlexity and GitHub Copilot RCE (CVE-2025-53773), demos of attacks such as LegalPwn and PROMISQROUTE, practical steps to protect your AI systems and much more. Sept 16 @ 9am PT / 12pm ET Register: https://t.co/tDyXBqwbo1 #AISecurity #AIDR #CyberSecurity #RedTeam

What if we told you that AI models trust legal documents so much, they'll ignore their own safety protocols? Our latest "LegalPwn" research shows exactly how. Thanks @ghalfacree at @TheRegister for

What if we told you that AI models trust legal documents so much, they'll ignore their own safety protocols? Our latest "LegalPwn" research shows exactly how. Thanks @ghalfacree at @TheRegister for the excellent writeup! 🙏 https://t.co/xpQsmj63YH #AISecurity #CyberSecurity #LegalPwn @pangeacyber

Researchers at cybersecurity firm @ESET have uncovered what they believe is the first AI-powered ransomware seen in the wild. The malware, dubbed PromptLock, leverages a hard-coded prompt injection at

Researchers at cybersecurity firm @ESET have uncovered what they believe is the first AI-powered ransomware seen in the wild. The malware, dubbed PromptLock, leverages a hard-coded prompt injection attack against an LLM, allowing it to inspect local filesystems, exfiltrate files and encrypt data. This marks a major turning point in how adversaries may operationalize LLMs for malicious purposes. Learn more: https://t.co/i1eOZiQqFI via Derek B Johnson of @CyberScoopNews #Malware #AIDR #AISecurity #PromptLock

Attending Gen AI Summit Los Angeles? Swing on by the @pangeacyber booth to learn how AI Detection & Response (AIDR) equips security teams with the detection and response needed for AI. Pangea's AI S

Attending Gen AI Summit Los Angeles? Swing on by the @pangeacyber booth to learn how AI Detection & Response (AIDR) equips security teams with the detection and response needed for AI. Pangea's AI Security Platform helps security teams protect employee AI use and ship secure AI apps fast with the industry's most comprehensive set of guardrails. Secure AI from cloud to code! Explore and get a demo: https://t.co/e2LxUKua5X #CyberSecurity #AISecurity #LLMSecurity #AIDR @AI_AInstitute

Great news! @pangeacyber has been named the winner of the AI-Powered Threat Detection category in the 2025 TechForward Awards. It's an honor to be recognized in the inaugural year of this program. Th

Great news! @pangeacyber has been named the winner of the AI-Powered Threat Detection category in the 2025 TechForward Awards. It's an honor to be recognized in the inaugural year of this program. Thank you @SiliconANGLE. Details: https://t.co/Gc5QLwjlbw #CyberSecurity #AISecurity #AIDR

https://t.co/qiIqhBcrEM is facing a class-action lawsuit alleging it secretly recorded private conversations on Zoom, Google Meet and Microsoft Teams without participants’ consent. The suit claims tho

https://t.co/qiIqhBcrEM is facing a class-action lawsuit alleging it secretly recorded private conversations on Zoom, Google Meet and Microsoft Teams without participants’ consent. The suit claims those recordings were then used to train Otter’s #AI transcription models, violating state and federal privacy laws. The case highlights growing concerns over transparency and trust in workplace AI tools. Learn more: https://t.co/09qVpg86Fb via @BobbyAllyn of @NPR #CyberSecurity #AIDR #AISecurity #LLMSecurity

Great piece by our very own @MrJoeyMelo on LegalPwn, our latest research discovery. We found that AI models can be tricked into executing malicious code just by wrapping it in fake copyright notices

Great piece by our very own @MrJoeyMelo on LegalPwn, our latest research discovery. We found that AI models can be tricked into executing malicious code just by wrapping it in fake copyright notices and legal disclaimers. Learn more: https://t.co/oDCaeR1Dik #AISecurity #CyberSecurity #ThreatResearch #AIVulnerability #LegalPwn @pangeacyber

This recent on-demand webcast, led by @pangeacyber CTO Sourabh Satish and hosted by the @AI_AInstitute, offers a crash course in building and securing AI agents. Watch the session: https://t.co/v2axn

This recent on-demand webcast, led by @pangeacyber CTO Sourabh Satish and hosted by the @AI_AInstitute, offers a crash course in building and securing AI agents. Watch the session: https://t.co/v2axnskNvj It covers how agents work, the components involved and how Model Context Protocol (MCP) architectures can simplify agent design and enhance flexibility. It also dives into critical security considerations, including risks like tool poisoning, memory misuse and prompt injection, and compares guardrail options across open-source, DIY and commercial solutions. The session closes with actionable best practices for engineering resilient, secure and scalable agent-based AI applications. You'll learn how to: ▫️Identify emerging threats – See how attackers exploit AI agents via prompt injection, tool poisoning and memory leakage and how to defend against them. ▫️Architect with security in mind – Understand secure design patterns, including enforcing least privilege and logging AI interactions. ▫️Choose the right guardrails – Compare open-source, DIY and commercial guardrail solutions to balance flexibility, cost and security. Get the companion eBook: https://t.co/7NndavVqaA #AIDR #CyberSecurity #AISecurity #LLMSecurity #AgenticSecurity #PromptInjection #OpenSource

@pangeacyber Co-Founder & CTO, Sourabh Satish met with @BriannaMonsanto at @ITBrew to discuss the future of super agents. In her story, "@Walmart debuts super agents. Experts say more companies will

@pangeacyber Co-Founder & CTO, Sourabh Satish met with @BriannaMonsanto at @ITBrew to discuss the future of super agents. In her story, "@Walmart debuts super agents. Experts say more companies will follow," Sourabh explains: "As enterprises are launching these AI agents, there has been a realization that there's going to be a sprawl of AI agents, and it's going to be challenging for employees and users trying to use the right agent for the right purpose. So, this is very much in line with the expectation that there will be consolidation from a usability perspective." This shift introduces significant security complexities around authentication, authorization and data monitoring as information flows between multiple agents and enterprise systems, leaving enterprises vulnerable without proper security guardrails. Read more: https://t.co/8gajn5iViw #AIAgents #AISecurity #CyberSecurity #GenAI #ShadowAI #AIDR

Secure GenAI Across Your Enterprise! Join us in Vegas this week to explore how to detect Shadow AI, enforce responsible GenAI use and secure your enterprise from prompt-based attacks. Private suite

Secure GenAI Across Your Enterprise! Join us in Vegas this week to explore how to detect Shadow AI, enforce responsible GenAI use and secure your enterprise from prompt-based attacks. Private suite meetings available with @pangeacyber Founder and CEO Oliver Friedrichs, Co-Founder & CTO Sourabh Satish and the rest of the executive team. Let's meet: https://t.co/AZ1WCrumH5 #CyberSecurity #AISecurity #PromptInjection #ShadowAI @defcon @BlackHatEvents #BHUSA

Excellent deep dive by @bendee983 at @bdtechtalks on our "LegalPwn" research: "Abusing Legal Disclaimers to Trigger Prompt Injections." He thoroughly explores how this attack exploits LLMs' programmi

Excellent deep dive by @bendee983 at @bdtechtalks on our "LegalPwn" research: "Abusing Legal Disclaimers to Trigger Prompt Injections." He thoroughly explores how this attack exploits LLMs' programming to respect legal disclaimers and copyright notices, hiding malicious instructions within seemingly official language. Key findings from @pangeacyber Labs: ◽️Real-world impact demonstrated on gemini-cli and GitHub Copilot ◽️Cross-model vulnerability affecting Google Gemini, Meta Llama, and xAI Grok ◽️Advanced payloads bypassed even models with strong system prompts ◽️Consistent resistance observed in Anthropic Claude models and Meta's Llama Guard 4 The data shows that Pangea's AI Guard consistently detected and blocked every #LegalPwn variation tested, highlighting the need for dedicated AI guardrails as LLMs integrate deeper into critical systems. Read the full analysis: https://t.co/4ztpgEsjWn #AISecurity #CyberSecurity #GenAI #PromptInjection #AIDR

Great piece by @fredericl at @thenewstack: "Why Generative AI Needs A New Kind of Security Solution.” Frederic goes beyond our AIDR launch to explore Pangea's broader mission and the fundamental shif

Great piece by @fredericl at @thenewstack: "Why Generative AI Needs A New Kind of Security Solution.” Frederic goes beyond our AIDR launch to explore Pangea's broader mission and the fundamental shifts happening in cybersecurity. Highlights: ◽️Prompt injection attacks that can trick models into providing unauthorized data ◽️#Data leakage from employees using third-party AI tools with confidential information is a major risk ◽️With Pangea #AIDR, we're providing end-to-end monitoring and security across the entire AI workflow, from shadow AI usage to internal agent frameworks Read more: https://t.co/NINI3yC0ns #AISecurity #CyberSecurity #CyberDefense #PromptInjection

🎉 Thank you @duncanriley at @SiliconANGLE for the excellent coverage of @pangeacyber AI Detection and Response (AIDR) launch. He captures the critical challenge we're addressing: ▫️Generative AI rep

🎉 Thank you @duncanriley at @SiliconANGLE for the excellent coverage of @pangeacyber AI Detection and Response (AIDR) launch. He captures the critical challenge we're addressing: ▫️Generative AI represents both the fastest software adoption curve in history AND the fastest-growing security blind spot. ▫️Traditional security controls simply can't keep pace with emerging AI threats like indirect prompt injection and excessive agency. ▫️AIDR shifts security teams from reactive governance to proactive AI oversight, providing full visibility into AI usage. Read the full article to better protect your org: https://t.co/ttXMZMwntQ #AIDR #AISecurity #CyberSecurity #GenAI #InfoSec