@hiddenlayersec profile picture
HiddenLayer @hiddenlayersec

HiddenLayer helps enterprises safeguard the AI models behind their most important products with a comprehensive security platform

Joined July 2022
  • Tweets 475
  • Following 24
  • Followers 819
151 Photos and videos
Pinned Tweet
HiddenLayer@hiddenlayersec
4 Mar 2025
The 2025 AI Threat Landscape Report is here. Our latest report breaks down the real-world tactics attackers are using against AI, the emerging risks organizations need to know, and how security teams can stay ahead without slowing innovation. 🔗 hiddenlayer.com/threatreport…
1
3
10
1,822
@hiddenlayersec has uncovered EchoGram, a technique capable of manipulating the guardrails that protect leading LLMs like GPT-5. This shows the need for diverse, adaptive, & validated security layers to keep pace with rapidly evolving threats. 👉 hiddenlayer.com/innovation-h…

EchoGram: Bypassing AI Guardrails via Token Flip Attacks

HiddenLayer unveils EchoGram, a new attack technique that manipulates AI guardrails protecting LLMs like GPT-4, Claude, and Gemini.

hiddenlayer.com
1
6
800
HiddenLayer researchers have discovered a simple bypass based on our still-functional Policy Puppetry technique for OpenAI's brand-new Jailbreak and Prompt Injection detection guardrails! Read more 🔗 hiddenlayer.com/innovation-h… #AgenticAI #AgenticRisks #AISecurity

OpenAI Guardrails Bypass: The "Self-Policing" LLM Vulnerability

Explore how the same AI model can behave differently depending on context, creating hidden risks and unpredictable security outcomes.

hiddenlayer.com
1
4
442
Databricks launches its Data Intelligence Platform for Cybersecurity, and HiddenLayer is proud to be part of it. We secure the models at the heart of AI defenses, ensuring trust, compliance, and resilience. 🔗hiddenlayer.com/innovation-h…
2
7
565
🚨 The first fully AI-powered cyber attack is here. Anthropic’s report reveals how criminals used Claude Code to run an entire campaign, from data theft to ransom demands. 🔎 Our deep dive: hiddenlayer.com/innovation-h… #AISecurity #CyberSecurity #AIThreats

The First AI-Powered Cyber Attack: Anthropic Threat Report

Cybercriminals are using AI to run entire operations—from reconnaissance to ransom. Explore Anthropic’s findings and key takeaways.

hiddenlayer.com
1
7
320
🔍 Can a single image hijack your AI’s behavior? Yes & without changing the application. Meet VISOR: a new method that steers GenAI models using images alone. It’s a new class of AI vulnerability and a new opportunity for AI alignment. 🔗hiddenlayer.com/innovation-h…
1
4
258
⏰ Calling all cybersecurity enthusiasts! Only 24 hours left to show your skills at the @BugBountyDEFCON Capture The Flag competition, sponsored by HiddenLayer. This is your chance to challenge yourself, compete with top talent & win exciting prizes. 🔗bbv.ctf.ae
2
8
878
🧠💻 Your AI coding assistant could be executing invisible instructions without your knowledge. We found a way to hijack Cursor using nothing more than a README file. No malware. No alerts. Just invisible prompt injections. 🔗 hiddenlayer.com/innovation-h…

Hidden Prompt Injections Can Hijack AI Code Assistants

New research shows how AI coding tools like Cursor can be weaponized via indirect prompt injection to steal secrets and run malicious code.

hiddenlayer.com
2
3
269
Our CEO, Chris Sestito, joined the Hundred Year Podcast to discuss why AI security is urgent and what to do about it. 🎧 Listen now: podcast.hundredyear.com/2062…

Hundred Year Podcast@hundredyearpod
17 Jul 2025
The Hundred Year Podcast is back! AI security hacks are an unfolding emergency, so Christopher “Tito” Sestito from @HiddenLayerSec joined @AdarioStrange on the pod to explain what we can do about it. Link in the comments! 🚀
1
206
🎥 Missed it live? Catch the replay of our webinar on the taxonomy of adversarial prompt engineering. Learn how to break down LLM prompt attacks by objectives, tactics, and techniques and why it matters for real defense. 🔗 Watch here: youtube.com/watch?v=EMvM8tmm… #AISecurity

HiddenLayer Webinar: Introducing a Taxonomy of Adversarial Prompt...

Discover HiddenLayer's innovative Adversarial Prompt Engineering (A...

youtube.com
3
317
🚨 Join our live walkthrough of @hiddenlayersec's new taxonomy of adversarial prompt engineering, a framework for classifying & combating prompt-based attacks against LLMs. ⏰ June 25th, 11am CST 🔗 Register here: hiddenlayer.zoom.us/webinar/…
0:30
3
180
🔐 Not all prompt injections are the same. We just released a taxonomy of adversarial prompt engineering, mapping the why, how, and what behind LLM prompt attacks. Built for red teamers, defenders & researchers. Open to the community. 🔗 hiddenlayer.com/innovation-h…
3
9
1,243
HiddenLayer researchers have found a way to bypass text classification models by targeting tokenizers. TokenBreak gets past protection models, leaving end targets exposed. 🔗 hiddenlayer.com/innovation-h… #AISecurity #AI #LLMSecurity

The TokenBreak Attack

Learn how the TokenBreak attack manipulates AI token processing to bypass safeguards and trigger unintended model behavior.

hiddenlayer.com
4
5
366
HiddenLayer retweeted
Help Net Security@helpnetsecurity
10 Jun 2025
Securing agentic AI systems before they go rogue - helpnetsecurity.com/2025/06/… - @hiddenlayersec #AI #AgenticAI #LLM #CyberSecurity #netsec #security #InfoSecurity #CISO #ITsecurity #CyberSecurityNews #SecurityNews #video

Securing agentic AI systems before they go rogue - Help Net Security

This video breaks down how agentic AI functions, its potential to revolutionize business operations, and the vulnerabilities it introduces.

helpnetsecurity.com
2
2
436
📢 New from @HiddenLayerSec: The Financial Services AI Security Playbook is here. A guide for CISOs to secure, govern & scale AI without slowing innovation. - Model audits - Red teaming - NYDFS-aligned IR - Ethics & explainability 📥 Download now: hiddenlayer.com/financial-se…

Securing AI: The Financial Services Playbook | HiddenLayer

A practical playbook for securing, governing, and scaling AI systems in financial services.

hiddenlayer.com
1
110
AI models can’t govern themselves. Our latest blog explores how to build holistic AI model governance from day one, so you can move fast and stay secure. 🔍 AIBOM 🧬 Model Genealogy ⚖️ Compliance-ready Read more: hiddenlayer.com/innovation-h… #AISecurity #AI #AIGovernance

Built-In AI Model Governance

Learn how to embed AI model governance from day one with a holistic framework that ensures compliance, transparency, and secure AI deployment.

hiddenlayer.com
115
Function parameter abuse isn’t limited to MCP - it’s a transferrable vulnerability affecting most SOTA models. HiddenLayer researchers extract full system prompts via fake functions with malicious parameters across Claude 4, ChatGPT, Cursor & more. 🔗 hiddenlayer.com/innovation-h…

LLM Fake Function Injection: How to Prevent System Prompt and Tool Leaks

Discover how attackers exploit agentic function parameters beyond MCP, enabling new classes of AI system manipulation.

hiddenlayer.com
1
3
12
1,334
New from @DarkReading: LLMs on rails? 🚆 The design choices keeping large language models secure and what the risks are if we get it wrong. HiddenLayer weighs in on the engineering security challenges ahead. 🔗 darkreading.com/vulnerabilit… #AIsecurity #LLMs #CyberSecurity #infosec

Keeping LLMs on the Rails Poses Design, Engineering Challenges

Despite adding alignment training, guardrails, and filters, large language models continue to give up secrets, make unfiltered statements, and provide dangerous information.

darkreading.com
3
116
🚨HiddenLayer’s Director of Adversarial Research, Jason Martin, joins The Data Exchange Podcast to talk about what it takes to actually defend LLMs. 🎙️ Beyond Guardrails: Defending LLMs Against Sophisticated Attacks. Stream now: youtube.com/watch?v=L9MXnBeH…
2
150
AI security vulnerabilities are evolving faster than most teams can keep up. From dev to deployment, discover a real-world example of how to protect your models throughout their lifecycle in our latest blog. 🔗 hiddenlayer.com/innovation-h… #AISecurity #MachineLearning #AI

The AI Security Playbook

Get a practical AI security playbook with strategies to protect models, data, and infrastructure across modern AI systems.

hiddenlayer.com
1
2
193
HiddenLayer researchers have found a way to abuse MCP to extract chat history, full system prompts, previous tool use, and more by simply inserting specific parameters into tool functions. 🔗: hiddenlayer.com/innovation-h… #MCP #AI #AISecurity

MCP Security Alert: Extracting AI System Prompts via Parameter Abuse

HiddenLayer reveals a critical MCP vulnerability exposing sensitive data. Discover the AI security risks and how to protect your models.

hiddenlayer.com
2
3
273
Load more