Prevalent, Inc., has announced the launch of Alfred™, claimed to be the world’s first virtual third-party risk advisor. Alfred is available in the Prevalent TPRM Platform and is the latest capability delivered as part of Prevalent’s broader strategy to securely and responsibly harness the power of AI to address today’s most complex real-world third party risk management challenges, while ensuring governance and security over AI usage.
Alfred is a conversational AI backed by a large language model (LLM) trained on billions of events and more than 20 years of Prevalent’s risk management experience. Alfred offers a guided user experience that provides contextual information to reduce the time typically required to analyze risk and make decisions, improving the quality of decision-making.
Alfred uses anonymized data, such as a third-party risk or event observation, to provide an informative, intuitive and guided user experience. It then leverages a curated LLM to deliver expert risk management insights in the context of industry guidelines including NIST, ISO, and SOC2 among others.
Alfred is seamlessly embedded into the Prevalent TPRM solution, empowering users to benefit from its guidance without leaving the platform or disrupting existing workflows.
Prevalent is committed to responsible AI innovation, prioritizing the safety and security of customer data. The company has employed a series of controls to ensure high-quality, secure and socially responsible results. Prevalent addresses AI’s inherent risks including:
- Bias: the LLM incorporated into the Prevalent Platform is trained on real, anonymized third-party risk data and events validated by over 20 years of industry experience.
- Data security: the model was trained on anonymized risk and event data, with no other context provided that could link the data back to a specific company.
- Governance and oversight: the Prevalent team continually audits and reviews outputs to validate the accuracy of findings.
