Beyond OpenClaw: How to govern agentic AI in the enterprise - Free Webinar 26th February 2026
The OWASP Top 10 for Agentic Applications provides a starting point. In this new blog series, we are diving deep into each risk starting with ASI01: Agent Goal Hijack.
The OpenClaw incident is the first mass-casualty event for Agentic AI, resulting in 1,000+ exposed local machines and 1.5 million leaked API keys. Here we analyze how "Shadow Agents" bypass traditional firewalls and why your employees' productivity tools might be the biggest open door on your network today.
Most organizations lack visibility of the AI tools their employees use every day, creating a massive security gap. FireTail’s latest release introduces a comprehensive range of workforce AI security features that help enterprises close that gap and adopt AI innovation securely.
Viktor Markopoulos lifts the lid on a new technique where malicious text is hidden inside innocent emojis using undeclared Unicode characters. Learn how this "invisible" payload bypasses human review and security filters, creating a blind spot for CISOs and security teams.
We’ve reached the last risk on the OWASP Top 10 list of risks for Language Learning Models. Read on to learn all about LLM10: Unbounded Consumption- its causes, effects, and most importantly, how to mitigate it.
The OWASP Top 10 Risks for LLMs helps shed light on the top vulnerabilities to AI in today’s landscape. In this blog, we’ll go over LLM09: Misinformation. What it is, how to mitigate it, and more.
In an ecosystem of constantly rising AI threats and attacks, the OWASP LLM Top 10 is here to give guidance on the biggest risks in the landscape and how to combat them. Today’s blog dives into #8: Vector and Embedding Weaknesses.
In 2025, AI is revolutionizing our cyber landscape and changing everything we know about cybersecurity. Luckily, the NIST AI Risk Management Framework is here to help. Join us for an in-depth exploration of the AI RMF, which is updated for the present landscape.
The OWASP Top Ten Risks for LLMs is a comprehensive list for security researchers to assess vulnerabilities in AI models. Today’s blog will dive in-depth into item 7: System Prompt Leakage.
Once again, Docker APIs are a target of threat actors in a new method of attack dating back to June 2025, or even earlier. Research is ongoing.
Agentic AI is introducing new risks to cybersecurity worldwide. The OWASP Top 10 Risks for LLMs breaks down the biggest risks in the landscape. Today’s blog will tackle #LLM06: Excessive Agency.
FireTail was one of four startups selected as a finalist in the Black Hat USA 2025 Startup Spotlight Competition. This week was unforgettable and reaffirmed the urgent demand for AI security solutions.
OneLogin, a popular identity and access management platform, had vulnerabilities that exposed user credentials. Through simple probing, researchers were able to access a host of sensitive data…
It is no secret in 2025 that AI can be abused to launch attacks by threat actors. But the “how” and “why” of these use cases is continuing to change. A recent security report revealed many of the ways in which OpenAI’s ChatGPT could be exploited.
In this blog series, we’re breaking down the OWASP Top 10 risks for LLMs and explaining how each one manifests and can be mitigated. Today’s risk is #4 on the list: Data and Model Poisoning. Read on to learn more…
We’ve talked before about Mean Time To Attack, or MTTA, which has grown alarmingly short for new vulnerabilities across the cyber landscape. In this blog, we’ll dive into the “how” and “why” of this…
Cybersecurity risks are too close for comfort. Recent data from the Global Mobile Threat Report reveals that our mobile phone applications are most likely exposing our data due to insecure practices such as API key hardcoding.
The OWASP Top 10 List of Risks for LLMs helps developers and security teams determine where the biggest risk factors lay. In this blog series from FireTail, we are exploring each risk one by one, how it manifests, and mitigation strategies. This week, we’re focusing on LLM03: Supply Chain vulnerabilities.
APIs power all of the connections we take for granted in the modern internet. But as we rely on them more for new technologies like AI, securing them is harder than ever. That’s why continuous API security testing is an essential part of every cybersecurity posture.
Our modern “Software as a Service” model is becoming a challenge for cybersecurity teams within large enterprises, as attacks continue to rise in volume and complexity across the cyber realm. Security needs to be a consideration from code to cloud, or any progress we make will be undone just as quickly.
OWASP’s Top 10 for LLM is a good starting point for teams to learn about AI security risks. In this series, we’ll go over each risk and practices to protect against them. Today, we’re tackling LLM02: Sensitive Information Disclosure.
It’s here! The New CIS Guide for API Security provides teams with actionable steps for their own API security postures. API use is skyrocketing with the recent adoption of AI, and security teams are struggling to keep up with the rising threats. That’s where the CIS Guide comes in.
API security is a critical issue, especially with the rise of AI, which runs on APIs. So how do we ensure consistent API security in an age of growing threats? In this blog, we’ll go over one of the most critical enabling aspects of API security: API specifications.
Researchers recently found a vulnerability in Apache Tomcat’s servers that would allow an attacker to commit Remote Code Execution with a single PUT request to a specific API, followed by a GET. And now, this vulnerability is officially being exploited in the wild.
Security teams today face a dual challenge: protecting AI systems from external threats while securing the APIs that power them. The reality is clear—if your APIs aren’t secure, neither is your AI.
A BreachForum user came out claiming to have breached OmniGPT and shared samples of stolen data to back up this claim. Weeks later, researchers are still scrambling to figure out the scope, attack method, and more.
Today’s cyber landscape is littered with threats, risks, and vulnerabilities. Every week, we are seeing an increase not only in attacks, but also in the methods used to attack. This week, a new family of malware was discovered exploiting Microsoft’s Graph API.
AI security and API security run alongside each other, much like a double rainbow. Each one contains a full spectrum of security requirements that work in tandem with one another.
In 2025, AI is the biggest advancement in cybersecurity and the talk of all tech-sperts. But as AI continues to develop, we are seeing a surge in not only the benefits, but also the risks of artificial intelligence.
This blog post will answer questions such as: But where do APIs live? And how do they interact? What languages do they use?
The latest blog from the C-suite at FireTail attempts to answer the essential question: “Which is a bigger threat today - cloud misconfigurations or API vulnerabilities?”
Attackers used an attack method known as “crossbarking” via a malicious Chrome extension to inject custom code into the target’s Opera browser.
In a world of cyber risks, authorization is one of the most critical steps in an API security strategy. However, when it comes to authorization, the C-suite of FireTail believes it is better kept apart.
Microsoft Sharepoint recently patched vulnerabilities that highlighted the need for highly privileged user access to happen via secure APIs.
Technology Partnership Enables Mutual Customers to Reduce Cloud Risk and Enhance API Security.
Star Health suffered a massive data leak via API access. The personal information of millions of victims has been compromised, and worst of all, there may have been an insider who facilitated the breach.
Ecovac customers in Australia were startled when their vacuums began talking back to them, most notably using racial slurs. This was made possible through remote access and manipulation of the “smart” devices.
Researchers from DataDog recently discovered that hacker groups are targeting Docker Swarm, Kubernetes and SSH servers in one orchestrated attack using Docker API endpoint vulnerabilities.
Web application and service creation platforms rely on APIs for their functionality. However, one such platform, Versa Director, is vulnerable to API attacks and token theft.
Researchers found a new vulnerability that affects KIA systems and could allow anyone remote control over their vehicles using only a license plate.
APIs are everywhere and in every part of our lives. However, in recent years, attackers have been increasingly targeting APIs. So how do you secure an API, and whose responsibility is it?
APIs are used for everything, including dating apps. Feeld, a dating app targeted at multi-person relationships, recently faced an API vulnerability that exposed sensitive data, leaving users unsettled.
APIs can have many different types of security challenges, even those of tech giants such as Microsoft. In this blog, we’ll explore a recent vulnerability that affected Microsoft’s Azure API Management, and explore what that implies for the cloud shared responsibility model
In Wisconsin, nearly a million Medicare users’ personally identifiable information has been exposed as a result of the moveIT breaches of yester-year.
APIs can run almost anywhere, including any type of compute platforms and network infrastructure services on AWS. In this blog, we’ll go over the different types of compute platforms, network infrastructure services, and how they relate to your APIs and API security.
FireTail partnered with the Center for Internet Security (CIS) to create an API security community. The end result is the first draft of The CIS API Security Guide, reviewed by cybersecurity experts from around the world before its release. Read more here.
Recent vulnerabilities discovered in popular solar-power platforms highlight the need for API security across IoT and OT technology platforms.
APIs are a shortcut to the data. They pass through quietly, creating a phantom attack path that flows through all the other layers of protection. And although cybersecurity has come a long way, there still aren’t controls to mitigate these risks.
Cloudflare released its latest Application Security Report, which contained some startling news. They found that nearly 7% of all web traffic is malicious. But what does this really mean?
Google Cloud Security released an updated Threat Horizons report containing information about the top security risks. Here are our notes from the report.
Location-sharing services, like much of the modern internet, are powered by APIs. Because these APIs handle personally identifiable information such as addresses, phone numbers and more, breaches in these apps can lead to serious safety issues for users everywhere.
With new API threats popping up every day, no one is safe from vulnerability. This is especially dangerous when it comes to large, widely used servers such as Apache HugeGraph.
Some companies position privacy as a key value proposition of their products and services. But that may not always be as true as advertised.
A researcher at EvaSec recently discovered a vulnerability in the CocoaPods ecosystem that could potentially affect an undetermined (but huge) number of web users.
A new type of API attack has been discovered- and it’s particularly pernicious. The target? Exposed Docker APIs. The objective? Spreading cryptojacking malware.
Google probably didn’t want this to happen. The tech giant accidentally posted a whole host of sensitive internal documents to GitHub that partly detailed the way the search engine ranks web pages.
Let’s talk about a recent example of an API vulnerability that was exploited to gain access to data within the German government.
Researcher Sam Curry recently discovered some alarming vulnerabilities in his modem, a Cox Panoramic wifi gateway. It all started when he noticed an unknown IP address was copying his same HTTP requests.
In this talk, Jeremy will cover key knowledge from the cybersecurity landscape for CISOs in 2024. Tune in to hear valuable insights and takeaways every CISO can apply in their own security posture today. API security is the cornerstone of strong cybersecurity for CISOs.
A lot of our API use happens at home, in places you might not even expect. What happens when these APIs are left vulnerable?
API security by design is all about breaking down how security considerations can be brought into the various stages of an APIs lifecycle and simplifying the API security process from the developers’ standpoint.
Many application developers are still grappling with the integration challenge. Microsoft’s Graph API attempts to solve this problem, however, their solution comes with its own drawbacks.
In February of 2021, Postman launched a public API platform where developers could collaborate to build software. Now in 2024, Postman has the largest collection of public APIs. Naturally, this makes it a prime target for attackers.
Although microservices are widespread, they are often misunderstood by business leaders. While they present substantial benefits, they also have the potential to introduce new risks into the API environment. Understanding the benefits and risks of microservice utilization is a major step towards effective product development.
In the travel sector, securing a competitive edge is vital. In a hyperconnected industry, where demand fluctuates, pricing is dynamic and customers have endless options, efficient and well-secured APIs can make a huge difference.
The Cambridge Analytica Data Scandal led to the collapse of the company, court cases and massive fines for Meta. It highlighted the massive impact that technology was having on society, politics and democracy. Now, almost a decade later, we take a look at how a poorly configured API started it all.
Open Banking can best be thought of as a “microservices” approach to banking. It's a powerful concept that enables porting from service to service, and allows consumers to integrate this data with other providers. APIs are at the core and effective API security is a must.
The reality is that failing to invest in a proven API security posture might save you dollars in the short-term, but can cost millions in the long-term. We take a look at how increasing regulatory oversight is leading to massive fines.
The modern web is more connected than ever before. The move from monoliths to containerization and microservice-based architectures means API security is a must. In this blog, we look at what it takes to build secure modern internet services.
Everybody is talking about AI right now. It's the hottest topic in tech. But few people are talking about the APIs that underpin these AI platforms. Here we look at why effective API security is a must for any organization who wants to harness the power of AI.
At FireTail, we usually say that two or more things need to go wrong in order for attackers to be successful. So what went wrong with Spoutible's leaky API? Basically, everything.
Few technologies have become as ubiquitous in as short a timeframe as serverless computing. Serverless offers both benefits and downsides to API security. However, with proper knowledge of the best practices, we can reap these benefits while mitigating the threats.
FireTail CTO Riley found a web application vulnerability. He noticed when the app was open, and he had tools open, one of the requests was going to an API instead of a web application.
As we approach the end of the year, it's often the case that we look back and chat with colleagues about the highs and lows of the last twelve months. One such friend recently shared a story with us about the worst API they found in the wild during 2023. Could this be the worst API ever?
FireTail CEO Jeremy Snyder hosts an insightful discussion with a panel of cybersecurity experts. Mikko Hypponen, Sounil Yu and Ted Julian shared their thoughts on the emerging threats and cybersecurity trends likely to shape the next 12 months.
Various APIs belonging to a data service are leaking their Git repositories, at a backend API which contain the APIs' source code.
We are delighted to welcome two outstanding individuals to our advisory board – Mikko Hypponen and Sounil Yu. These new additions bring unparalleled expertise and strategic insights that will drive continued innovation and address the evolving API security needs of today's leading organizations.
With an ever-changing threat landscape, increased regulatory oversight and a desire among authorities to hold individual executives accountable for data breaches, these are tough times for CISOs. And as APIs become the attack vector of choice for bad actors, it’s more important than ever to prioritize effective API security at your organization.
Unauthorized users could gain access to sensitive financial information via an application's API using the data leaked via Github.
API security solutions focused on ML/AI pattern recognition analyze behavior ‘out-of-band’ to identify anomalies, reporting issues after the fact. Prone to false positives, this approach also means attackers will have some success before you learn from it. That’s not acceptable.
Recently announced OpenAI changes mean users will be able to call any API. That’s great for business and the economy but a potential headache for security teams.
As they continue to rise in use, APIs are becoming a critical attack surface for ransomware groups.
While Web Application Firewalls, or WAFs, are often touted as a go-to solution for web app security, they frequently fall short – and in many cases, offer little more than an artificial sense of protection that is not backed by actual protection. But why isn’t a WAF good enough? Where do they fall short, and what is the better alternative? Let’s dive a bit deeper.
Since 2016, ECSO has been leading the way in European cybersecurity, building upon the foundation left behind by the Public-Private Partnership in Cybersecurity (cPPP). At FireTail, we are excited to be a part of this journey to bring even greater innovation to the European Cybersecurity landscape.
FireTail researcher Viktor Markopoulos discovered a vulnerability in a European Shipping Company’s APIs that allowed him to download internal files without authentication.
APIs and Artificial Intelligence are two of the most important developments in tech of the last 10 years. In this post, we look at the dual impacts that connect these two hot topics that make our online worlds work.
Poorly secured APIs at Points.com resulted in serious vulnerabilities that could have exposed the loyalty programs of some of the world’s best-known airlines and hospitality brands.
FireTail is proud to announce our partnership with CYFIRMA, a leading provider of external threat management solutions.
During Defcon, a security researcher presented his findings from assessing a global transportation system, leveraging APIs.
A file transfer software called moveIT experienced a vulnerability starting in mid-2023 that created a mass breach across many organizations and geographies. The breach is started by injection against an API administrative endpoint, and data is exfiltrated via administrative API calls.
An unauthenticated API belonging to a fast food company exposed receipts from all of its stores in India.
FireTail founder Jeremy Snyder discussed API security, some traditional approaches to cybersecurity, and why the two just don't work together as well as you might think.
FireTail founder Jeremy Snyder discussed what we can learn about API security from Malcolm Gladwell's 'Outliers'
FireTail founder Jeremy Snyder discussed FireTail's research into API security incidents and data breaches at API Days New York City.
FireTail founder Jeremy Snyder discussed API security to a standing-room-only crowd at UK Cyberweek in London in April 2023.
FireTail founder Jeremy Snyder presented at APISecure 2023; Learning from a decade of API breaches and why application-centric security is the right path
FireTail founder and CEO shares some thoughts on 2022 and what's ahead for 2023.
By using this website, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.
.png)
.png)
%20%5BRecovered%5D.png)
%20%5BRecovered%5D.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
