IBM named a Leader in The Forrester Wave™: AI Governance Solutions, Q3 2025

We are proud of this recognition; not just for the ranking, but because it reflects the work we’ve done with our customers to make AI safer, more transparent and more accountable.

Over the past decade, we’ve worked with organizations across many industries — especially highly regulated ones — to understand what responsible AI looks like in practice. That experience has shaped watsonx.governance into a solution that supports not only AI risk management and compliance, but also evaluation, observability, guardrails and lifecycle governance for AI systems.

As organizations face growing AI risks and regulations, along with pressure to unlock value from AI, our solution helps align technology, business, assurance, and security teams around consistent governance frameworks that turn principles into practice.

Forrester recognized IBM for our ability to manage the complexity of AI governance across many roles and responsibilities. Our strategy reflects years of collaboration with customers in complex and highly regulated industries — and that experience is built into watsonx.governance.

Watsonx.governance supports key governance needs like policy management, auditability,  and observability. It also includes new capabilities for agentic AI — helping teams govern an inventory of agents and tools, monitor agent behaviors, evaluate decision-making and detect risks like hallucinations.

Forrester also cited our solid vision, roadmap, and flexible packaging. We believe this recognition reflects our focus on practical, scalable governance for fast-evolving AI technologies.

AI governance isn’t a challenge for just one department; it requires collaboration across engineering, legal, compliance, risk, security, data and business teams. Our goal with watsonx.governance is to strengthen that collaboration by offering tools that support policy management, auditability and transparency throughout the AI lifecycle.

As IBM and other organizations invest more in AI governance, the perception of these solutions is changing. They’re no longer seen just as compliance tools or business costs, but as essential enablers for scaling AI and unlocking value from data. In fact:

• 27% of Fortune 500 companies cite AI regulation as a risk in their annual reports (WSJ/2024).

• 68% of CEOs we surveyed say governance for generative AI must be built into the design phase (IBV/2025). 

• 65% of data leaders at a recent Gartner conference named data governance as their top focus for 2024 (IBV/2025).

We’ve also focused on preparing for and managing emerging risks. As AI systems become more autonomous — with agents making decisions, interacting with tools and taking actions independently — governance must evolve to ensure these agents follow business goals, policy standards and safety expectations

We’ve added 3 new capabilities to help teams govern these agents more effectively:

A centralized inventory of the tools and agents your teams use, enriched with governance metadata like ownership, descriptions and usage history. Users can compare assets side-by-side using evaluation metrics — such as quality, cost, latency and performance — to make better decisions about reuse and adoption, reducing the risk of unvetted or duplicate capabilities

Evaluators help measure the performance of your agentic application during development. Our SDK offers out-of-the-box metrics like faithfulness, context relevance, and answer relevance. These can be used in-the-loop (during the flow of the agent to control its logic or as decision points) or offline (after the agent flow to evaluate overall agent quality). This is available using a simple decorator-based approach using a single line to invoke the evaluator can be added to each node of the agent, making evaluation easy to implement. This helps developers build higher-quality applications that are more resilient, trustworthy and ready for validation and production.

Experimentation Tracking provides full visibility into how agents and tools perform over time — from development to deployment. Teams can log runs, compare outcomes and rank configurations using custom business, safety and quality metrics, making it easier to choose the best-performing approach based on metrics data.

Customers consistently say that watsonx.governance helps them scale AI with confidence. For example:

• The United States Tennis Association (USTA) used watsonx to build generative AI features like Match Reports and AI Commentary for the US Open. They emphasized how governance tools helped ensure their models were performant, compliant, and aligned with editorial goals.

• Deloitte highlighted how watsonx.governance supports responsible AI adoption, helping clients embed trust and transparency into their AI workflows.

Forrester’s report specifically notes that IBM customers appreciate “the ability to align technology, business, and risk and legal teams on policy and approvals” — a capability that’s becoming more critical as AI expands across organizations. 

As these customers scale their AI efforts, they’re asking for deeper model risk assessments; feedback that’s actively shaping our product roadmap and future capabilities. 

As organizations look for practical ways to bring consistency, control, and visibility to agentic development, demand for our Governed Agentic Catalog continues to grow. It helps teams manage tool usage, lineage, and reuse across projects — making it easier to scale responsibly while maintaining oversight. Production monitoring and Guardrails for agentic AI are also in demand and we have new features on our roadmap to address these areas.  

Governance isn’t just about compliance — it’s about the efficient, safe and trusted use of technology like artificial intelligence. As Forrester recognized in naming IBM a Leader, our experience working with clients in complex and regulated industries has shaped a solution that meets today’s AI governance challenges and prepares for tomorrow’s.

As AI continues to evolve, we’re committed to helping our customers stay ahead of risks while unlocking the value of these technologies. watsonx.governance was built with real-world complexity in mind, with automation to scale and a roadmap shaped by the practical needs of organizations implementing responsible AI.

To learn more about how watsonx.governance can help your organization, connect with our team for a live demo.

Get the report here