As Generative AI (GenAI) and Large Language Models (LLMs) become integral to modern apps, we face a critical challenge of protecting sensitive user data from inadvertent exposure. In this article, we’ll explore the importance of content filtering in LLM-powered apps, and provide strategies for its implementation. Looking for step-by-step tutorials on prompt sanitization for OpenAI, Langchain, Anthropic, and more? Skip down to the “Tutorials & further learning” section below.

In the rapidly evolving cybersecurity landscape, the sheer volume of Common Vulnerability and Exposure (CVE) notices has become a daunting challenge for SecOps teams. In 2023 alone, the National Institute of Standards and Technology (NIST) issued 28,901 CVE notices, reflecting the growing complexity and intensity of potential threats.

Fireblocks offers businesses a broad suite of solutions to meet all of their tokenization needs. The Fireblocks Tokenization Engine — only one part of our ecosystem — is a platform powering the future of financial markets and digital asset ownership. Businesses can use the Tokenization Engine to launch and manage any tokenization solution and securely mint, custody, distribute, and manage tokenized assets.

This blog is part of the ongoing “I&O Perspectives” series, which features insights from industry experts about the impact of current threats, networking, and other cybersecurity trends. In the first part of this blog series, we observed the inception of an internet-based model where corporate networks have no borders anymore, home is the office, applications are in the cloud. This paradigm shift makes connectivity ubiquitous. But security risks have dramatically increased.

With the adoption of AI in almost every sphere of our lives and its unending advancement, cyberattacks are rapidly increasing. Threat actors with malicious intent use AI tools to create phishing emails and other AI-generated content to bypass traditional security measures. On the bright side, the security capabilities of AI are limitless. AI-enhanced attacks refer to cybersecurity events that use artificial intelligence to compromise individuals' and organizations' safety.

Today’s social platforms are “public by default”, from Instagram and LinkedIn to Venmo and Strava.

JFrog’s security research team continuously monitors open-source software registries, proactively identifying and addressing potential malware and vulnerability threats to foster a secure and reliable ecosystem for open-source software development and deployment. This blog details a PyPI supply chain attack technique the JFrog research team discovered had been recently exploited in the wild.

Keeper is cost-effective and easy to deploy – and the industry is taking notice. Keeper Security’s zero-trust Privileged Access Management (PAM) solution, KeeperPAM, has again been recognized as a Value Leader by the IT and data management analyst research and consulting firm Enterprise Management Associates (EMA) in its 2024 PAM Radar Report: Read the Full Report.

As American poet Nikki Giovanni wisely observed, "Mistakes are a fact of life. It is the response to error that counts." This rings particularly true in the world of cyber security. Even the most vigilant individuals can make mistakes—after all, we’re only human. What truly matters is how we respond. Imagine a platform that automatically detects risky security behaviours, alerting employees and nudging them to fix their mistakes before they escalate?