Introducing Julius: Open Source LLM Service Fingerprinting
The Growing Shadow AI Problem Over 14,000 Ollama server instances are publicly accessible on the internet right now. A recent Cisco analysis found that 20% of these actively host models susceptible to unauthorized access. Separately, BankInfoSecurity reported discovering more than 10,000 Ollama servers with no authentication layer—the result of hurried AI deployments by developers under […]
Stealing AI Models Through the API: A Practical Model Extraction Attack
Organizations invest significant resources training proprietary machine learning (ML) models that provide competitive advantages, whether for medical imaging, fraud detection, or recommendation systems. These models represent months of R&D, specialized datasets, and hard-won domain expertise. But what if an attacker could duplicate an expensive machine learning model at a fraction of the cost? Model extraction […]
As Strong As Your Weakest Parameter: An AI Authorization Bypass
In this AI gold rush, LLMs are becoming increasingly popular with many companies rolling out AI-assisted applications. When evaluating the security posture of these applications, it’s essential to pause and ask ourselves: what are we securing? Automated security tools that test models in isolation play an important role in identifying known vulnerabilities and establishing security […]
Exploiting LLM Write Primitives: System Prompt Extraction When Chat Output Is Locked Down
Prompt injection allows attackers to manipulate LLMs into ignoring their original instructions. As organizations integrate AI assistants into their applications, many are adopting architectural constraints to mitigate this risk. One increasingly common pattern: locking chatbots into templated responses so they can’t return free-form text. This seems secure. If an LLM can’t speak freely, it can’t […]
Where AI Systems Leak Data: A Lifecycle Review of Real Exposure Paths
AI data exposure rarely looks like a breach. No alerts are triggered, no obvious failure occurs, and most of the time nothing appears to be wrong at all. Instead, sensitive information moves through retrieval, reasoning, and storage layers that were never designed to enforce trust boundaries. Most organizations evaluate AI systems by reviewing individual components […]