Read our latest
Blog & Articles

The risk associated with unchecked AI usage is substantial. Gartner's 2023 AI in the Enterprise Survey reveals that nearly 30% of enterprises deploying AI have experienced an AI security breach. Of these breaches, 62% reported that internal parties compromised data.

Preamble's innovative AI Trust platform offers comprehensive solutions to protect AI systems from adversarial attacks, ensure the integrity of machine learning models, and safeguard data usage across organizations.

PITTSBURGH, PA, October 15, 2024- Preamble, a Pittsburgh-based Al safety and security operations platform startup, announced today that it has been granted a comprehensive patent for its groundbreaking system and methods to mitigate prompt injection in Al models.

Prompt Injection has been in the news lately as a major vulnerability with the use of instruction-following NLP models for general purpose tasks. In the interest of establishing an accurate historical record of the vulnerability and promoting AI security research, we are sharing our experience of a previously private responsible disclosure which Preamble made on May 3rd, 2022 to OpenAI.

Fintech companies are harnessing AI to manage numerous use cases for improved productivity and cost savings. However, one misstep out of regulatory compliance can cancel all these advancements through fines, penalties, or litigation costs, not to mention stock prices and reputation risks.

Even if you banned AI or haven’t started deploying AI tools across your company, your employees are probably using it. They're most likely to be the cause of a data leak!

The 2025 OWASP Top 10 list for Large Language Model (LLM) applications underscores the critical importance of robust security measures in AI systems. As LLMs become increasingly integrated into enterprise workflows, customer interactions, and autonomous systems, the risks they pose have grown exponentially. At Preamble, we’re proud to address these challenges head-on with solutions that secure AI applications and inspire confidence in their deployment.

At Preamble, we recognize this challenge. That’s why we’ve deliberately named our products, aligning as closely as possible with established cybersecurity definitions while accounting for the emerging AI safety and security needs.

Established by the Department of Commerce's National Institute of Standards and Technology (NIST), the U.S. AI Safety Institute Consortium (AISIC) will bring together AI creators and users, academics, government and industry researchers, and civil society organizations to meet this mission.

Much of the media we see online — whether from social media, news aggregators, or trending topics — is algorithmically selected and personalized. Content moderation addresses what should not appear on these platforms, such as misinformation and hate speech. But what should we see, out of the thousands or millions of items available? Content selection algorithms are at the core of our modern media infrastructure, so it is essential that we make principled choices about their goals.

We, humans, have a notoriously difficult time specifying what we actually want, and the AI systems we build suffer from it. With AI, this warning actually becomes “Be careful what you reward!”. When we design and deploy an AI agent for some application, we need to specify what we want it to do, and this typically takes the form of a reward function: a function that tells the agent which state and action combinations are good. A car reaching its destination is good, and a car crashing into another car is not so good.

As artificial intelligence (AI) becomes increasingly ubiquitous in our daily lives, it's more important than ever to address the potential vulnerabilities that come with these technologies. In order to mitigate the risks associated with AI, it's crucial for researchers and developers to engage in responsible disclosure practices. We'll explore what responsible disclosure is, why it's important, and how to go about it when it comes to AI vulnerabilities.

I am very much a fan of what my Stanford University colleague Francis Fukuyama has called “middleware”: content-curation services that could give users more control over the material they see on internet platforms such as Facebook or Twitter. Building on platforms’ stores of user-generated content, competing middleware services could offer feeds curated according to alternate ranking, labeling, or content-moderation rules

It's important to note that research thus far does not cover the gamut of what is essential to the businesses that may use AI in the future or even individuals. Organizations like Anthropic, OpenAI, Google, Microsoft, Meta, and Amazon may have policies about what is acceptable or not. Still, in the end, we are beholden to their consensus-based approach to acceptable or unacceptable behavior.