Principles of
Responsible AI

Artificial intelligence can empower legal teams to make quicker decisions, achieve better outcomes, and deliver greater business value to their organizations.

The opportunities are endless, from streamlining eDiscovery case assessment and review to identifying and protecting PXI or privileged data to analyzing and providing insights that support legal portfolio management and decision-making

But the use of AI also presents challenges. To minimize the risks, ProSearch is committed to building new applications with responsibility and usefulness top of mind. To that end, our work aligns with these principles:

We focus on creating solutions that provide real-world value. Every ProSearch AI solution is designed to help our clients solve a specific legal or compliance challenge.

ProSearchers understand and explain the costs of AI solutions in the context of their development and supporting workflow, as well as the value of the insights and benefits provided.

We work to reduce unfair biases in AI models by thoughtfully designing AI solutions, carefully curating training data, and thoroughly testing models. Fairness is prioritized by proactively mitigating biases through inclusive data practices and rigorous testing.

Our approach ensures AI systems perform consistently in different scenarios through robust training, monitoring, and testing. This confirms consistent performance in different situations.

We are committed to clearly communicating what our solutions can and cannot do and how clients’ data is stored and processed by ProSearch. Advocating transparency about actual capabilities, limitations, and data handling practices is crucial. Close collaboration with IT enhances privacy and security protections.

We prioritize the design of AI solutions that protect privacy and are secure from intrusions. We collaborate closely with clients’ IT and compliance teams to ensure we align with ISO, cybersecurity frameworks, and data privacy regulations. We commit to build and communicate processes that protect the handling of data by stakeholders that may interact with the system directly or indirectly.

ProSearch is dedicated to ensuring proper functioning of AI solutions. Most importantly, incorporating meaningful human oversight throughout the entire life cycle of an AI system, from development to deployment, maintains accountability. We commit to assess the impact of incorrect predictions and, whenever possible, design systems with human-in-the-loop review processes.

These principles guide our development of technologies and workflows and underscore our commitment to ProSearch clients and partners. As AI continues to evolve, we expect to evolve these principles over time, but always with the goal of driving positive change in the legal technology community.