Casper and IBM Partner to Apply Blockchain for AI Governance

The solution intends to help companies better govern training data and models across internal teams and external partners as AI proliferates.

Casper and IBM Partner to Apply Blockchain for AI Governance
Image Credit: Casper

Enterprise blockchain provider Casper Labs has announced a collaboration with IBM Consulting to develop a solution leveraging blockchain technology to enhance oversight capabilities for organizations deploying generative AI systems.

The solution intends to help companies better govern training data and models across internal teams and external partners as AI proliferates.

"As different groups augment data sets or modify algorithms, model outputs evolve in ways current systems struggle to track," said Casper Labs CEO Mrinal Manohar. "With IBM’s help, we’re committed to delivering a better way to not only understand why AI systems behave the way that they do but also a clearer path to remediate behavior if issues emerge."

Integrating the Casper Blockchain with IBM's AI governance platform watson.x, the solution aims to log granular input and output data for training generative AI. It then synthesizes this data into dashboards and reports providing cross-organizational visibility into model lineage, performance, compliance, and more.

The system's permissioned access intends to help protect proprietary data while enabling collective oversight of interconnected models. And its version control may let teams efficiently revert AI systems to previous states if needed.

The solution aligns with growing pressure for enterprises to control generative AI risks amid high-profile controversies involving biased outputs. The plan is to offer a suite of tools to enhance AI governance:

  1. Compliance Dashboard: A one-stop platform for monitoring AI systems across organizations, aligning with ethical guidelines.
  2. Quality Control Toolkit: Tools to monitor AI system performance and enhance transparency in AI outputs.
  3. Version Control: Facilitating rollbacks to address performance issues or biases in AI systems.
  4. Audit and Reporting System: A comprehensive system for AI process auditing and detailed reporting based on metadata from Casper Labs’ ledger.

The beta version is expected to be available in the first quarter of 2024, with a broader rollout planned subsequently through IBM Cloud Marketplace and other channels.

Shyam Nagarajan of IBM Consulting highlights the necessity of such solutions, emphasizing the need for trust, explainability, and risk mitigation in deploying AI at scale. This collaboration aims to furnish enterprises across various sectors, including finance, healthcare, and retail, with the tools to deploy AI responsibly and effectively.

Let’s stay in touch. Get the latest AI news from Maginative in your inbox.

Subscribe
Mastodon