Explore governance of foundation models and generative AI products. Learn about evaluating and governing these technologies in third-party management and model risk contexts.
Learn more about product development and design to make ethical AI considerations an enabler of innovative products your customers will trust and love.
While many were marveling at the release of OpenAI's GPT-4, Monitaur analyzed the accompanying papers that examined the risks and technical design of its latest engine through the lens of proper governance, responsible use, and ethical AI.
AI governance is a business enabler. Without proper governance practices, deploying AI is like asking your models to keep silent about the reasons they're making decisions that affect your business.
How to measure for bias can be a moving target. In this blog post, we will examine the common methods to evaluate for bias, how they conflict and our recommended approaches.
Compliance and risk management are often seen as the enemy to innovation. This has become more and more of a challenge as companies today have to move fast in order to react to market changes. Given the speed at which the world is changing, innovation can’t continue to be bogged down (by compliance or other internal processes) in order for companies to survive.
How does data bias happen, technically? Despite the media narrative, ML models are incredibly unintelligent, and here’s how they end up biased.
NIST Artificial Intelligence Risk Management Framework (RMF) is being crafted to “better manage risks to individuals, organizations, and society associated with artificial intelligence (AI).” The official framework is expected to be released in early 2023.
With higher-risk models, data scientists face responsibility to not only follow best practices for building responsible and ethical AI, but also develop common understanding with non-technical stakeholders to support overall business needs. There's a case to made for incorporating AI governance into data science lifecycle to enable developers in their work while also providing assurance to stakeholders.
Building and deploying strong, robust AI models is complex work on its own. Here is what technical organizations can do to make AI governance easier.
.png)
