Prompt Security

Securing GenAI for Riskified: ?? Riskified recognized the need for visibility into and monitoring capabilities of employees’ GenAI use, given how widespread it has become throughout its ranks. Such visibility and monitoring are vital to prevent sensitive data exposure, compliance violations, and other repercussions of?shadow AI. By integrating Prompt Security with Okta, Riskified can easily enforce AI-related policies across different user groups, facilitating a more scalable approach to AI security management. In short: Employees can use GenAI tools securely without disrupting their productivity. ?? Read the case study: https://lnkd.in/dEtqWEi9

...and that's a wrap for Cybertech 2025 ?? Thank you to everyone that in the past two and a half days stopped by our booth (was hard to miss it ??) It's been busy, productive, fun and tiring, and we can't wait to do it all over again next year. Cygnostic | Itamar Golan | Lior Drihem | Benjamin (Benji) Preminger | Yael Macias | Neta Ostrovsky | Or Tsimbler | Ori Tabac | Naama Peleg | Amit Guelmann | Ofir Tam | Sagie Karta | Shahaf Gibly | Omry Malka | Itay Savion

We couldn't be more excited to welcome David Moyal as our Head of Data Science ?? Looking forward to seeing you do great things ?? Read more here >> https://lnkd.in/d4YFZT37

?? AI is changing the way we write code—and Claude Code from Anthropic is one of the most powerful tools we’ve seen yet. At Prompt Security, we’ve been diving deep into it—and yes, we already support protecting it. ?? So what is Claude Code? It’s an autonomous coding agent built for complex dev workflows. ? Plans and executes multi-step dev tasks ? Writes and edits code across files ? Uses tools like terminal + editor ? Runs and debugs code independently ? Operates with a massive system prompt (~5KB!) ? Ships with 12 built-in tools (agents)—and you can plug in your own via MCP server ??We’re especially impressed by its agentic architecture—it doesn’t just autocomplete, it thinks through problems and makes decisions like a real engineer. ?? Watch how our agent steps in to stop Claude Code from leaking sensitive info—live in action in the demo below. ?? Curious about its system prompt and built-in tools? Drop a comment and I’ll send it your way.

Securing AI in 2025: A Practical Guide To Securing & Deploying AI ?? A few days ago, Francis Odum and his team at Software Analyst Cyber Research published a thorough and practical guide to AI Security. In it you can find: - State of Enterprise Adoption of AI In 2025 - Enterprise Risks In Developing AI In 2025 - Existing Cybersecurity Controls & Limitations - Market Solutions: How Organizations Should Secure AI - The Landscape for Security for AI - Detailed Breakdown For Vendor Market Offerings - Vendor Spotlight Get your copy of the guide here >> https://lnkd.in/dq6Gvkcp

Join us in welcoming Alan Merriam to the Prompt Security team ?? Alan joins us to lead our channels & alliances in North America. Best of luck, Alan! We're so happy to have you on board of this rocketship ??

Our CEO & Co-founder Itamar Golan has been selected #5 of the Top Cybersecurity Creators on the ranking by Favikon!?? Not bad, considering #1 is Asaf Rappaport on the week of Wiz's acquisition ?? You can see the entire list here >> https://lnkd.in/ddhPZfv6

Repeat after me: Network inspection isn't enough to enable secure GenAI adoption. GenAI tools introduce security blind spots that network inspection alone can’t cover. The only scalable, future-proof approach to securing GenAI is where AI usage happens: the browser. ? Protocol-agnostic: Covers thousands of AI tools effortlessly. ? Immune to API changes: No constant reconfigurations needed. ? Real-time data redaction & Shadow AI discovery: See the risks you know, and the ones you don’t. ? User coaching in the moment: Help employees make smarter security decisions, instead of just blocking access. Read more on the latest blog post by our CTO & Co-founder Lior Drihem >>https://lnkd.in/deFzgVQt

It’s almost time! ? The Premier International Cyber Event is back next week. Join Prompt Security at CyberTech Israel, March 24–26! Stop by our booth to see live demos of our GenAI Security platform and grab some exclusive swag. ??? Register for free: https://lnkd.in/dsFrhvy See you there! ?? Or Tsimbler | Benjamin (Benji) Preminger | Lior Drihem | Itamar Golan | Cygnostic | Yael Macias | Neta Ostrovsky

?? Keeping AI (and employees) within boundaries: Unveiling our comprehensive GenAI Authorization features ?? Organizations have spent years building robust, permission-based access systems, and here comes AI and introduces a brand new challenge. Employees can now simply ask AI to reveal sensitive information, like salary details or performance reviews, and LLMs may inadvertently comply. Our new Authorization features close this critical gap, enabling granular, context-aware access control at runtime, ultimately ensuring AI applications respect existing security boundaries. Key features include: - Contextual runtime authorization analyzing user identity and request context - Granular, department-specific policies tailored to roles and data privacy needs - Seamless integration with identity providers like Okta and Microsoft Entra - Real-time monitoring with flexible redaction options, from data masking to content blocking - Comprehensive audit logging for compliance and integration with SIEM solutions This isn’t just about restricting access, it’s about empowering enterprises to embrace AI with confidence, knowing their sensitive data remains protected. With our Authorization capabilities, we're delivering the missing piece that enables organizations to maintain their security posture while fully harnessing AI’s transformative potential. Read the full announcement >> https://lnkd.in/drXiHWfC