The AI Governance Gap: Why Responsible AI Drives Adoption

In conversations with customers across industries, trust is the barrier to AI adoption that comes up more often. Leaders and organizations need confidence that AI systems will perform reliably and remain accountable for suggested business actions, protecting sensitive data and treating people fairly. Ethical AI deployment closes that gap. When a system is technically robust, legally aligned, and ethically governed, customers, partners, and employees can scale it with confidence rather than caution.

SAP has updated the SAP Global AI Ethics policy to version 3.0, reflecting the fast-changing AI landscape and lessons drawn from deep-dive interviews with more than 50 international experts. The revision reflects concrete changes made to strengthen the ethical safeguards, accountability, and safety measures within enterprise AI adoption, addressing the substance of business leaders' concerns, not just the perception of them.

Our approach to AI rests on three words: relevant, reliable, and responsible. For SAP, ethical deployment isn't a regulatory checkbox or a marketing layer; it's the foundation that makes the Autonomous Enterprise possible, where intelligent agents coordinate and execute complex workflows while humans remain firmly in control.

Three disciplines for responsible AI

Ethics sit at the very center of SAP's broader social sustainability agenda, where responsible AI depends on three disciplines working together, not in isolation:

1. Ethics define our values. We set clear moral boundaries for how technology should serve humanity.
2. Security protects our systems and data. We safeguard core enterprise assets against emerging cyber threats and vulnerabilities.
3. Compliance ensures legal alignment. We are proactively preparing for global regulatory standards like the EU AI Act.

Together, they help teams assess whether an AI use case should proceed, what safeguards it requires, and who is accountable for its performance after deployment, ensuring that as business processes become automated, human dignity, fairness, and agency remain intact.

Ethics underpin the Autonomous Enterprise

As AI shifts from assisting people to acting on their behalf, the governance question becomes more urgent. The Autonomous Enterprise represents a repositioning where AI agents no longer just make recommendations, but execute business processes end-to-end, across finance, supply chain, procurement, HR, and customer experience. As CEO Christian Klein put it, "For the mission-critical processes of our customers, 'almost right' just isn't good enough."

That's precisely why ethics and governance can't be an afterthought bolted onto agentic AI. They have to be the foundation it runs on. Every action an SAP agent takes is logged and traceable, so organizations can always know what an agent did, why it did it, and what data it used. Autonomy without that foundation is just an experiment; autonomy built on it is an operating model organizations can trust at scale.

Translating principles into practice

The SAP Global AI Ethics policy, grounded in UNESCO's Recommendation on the Ethics of Artificial Intelligence, governs the development, deployment, use, and sale of every AI system across the company. SAP's updated AI Ethics Handbook translates these principles into day-to-day practice. It helps give engineering and product teams a one-stop reference for applying the policy's guiding principles, from fairness and non-discrimination to human oversight and sustainability.

To bridge the gap between high-level principles and day-to-day engineering, we embed strict operational checkpoints into our product life cycle:

• Mandatory AI ethics impact assessments: Every SAP AI use case must undergo a rigorous evaluation before reaching deployment, based on 10 AI ethics principles. Use cases receive a risk score and the SAP's AI Ethics Office helps product teams enforce mitigation measures and safeguards in place. Depending on the risks involved, some use cases go through several review steps and may be escalated to the SAP Global AI Ethics steering committee for a final decision.
• Human oversight by design: SAP systems are built to ensure humans remain in control of critical decisions.
• Internal accountability: Compliance is non-negotiable across SAP. All employees acknowledge the policy, and the rules apply universally, including to internal, self-developed AI applications built using tools like Hyperspace AI or Claude Code.

The process provides clear guardrails that are visible to SAP's stakeholders, including customers and employees.

Leading the industry in AI governance

The demand for transparent governance is urgent, and independent evaluations highlight significant gaps in corporate readiness worldwide. In benchmark research by the Thomson Reuters Foundation and UNESCO assessing 2,972 companies, SAP was among the only 12.4% found to maintain an explicit policy ensuring human oversight in AI systems.

SAP's leadership in ethical AI has earned external recognition: the company was named a leading positive example in the Collective Impact Coalition for Ethical AI Progress Report. SAP is also an active participant in the World Economic Forum's Global Coalition for AI and Social Innovation, convened by the Global Alliance for Social Entrepreneurship in partnership with the Schwab Foundation. Working alongside leading technology companies and social innovators, SAP is helping shape responsible AI deployment to accelerate social impact and extend the benefits of enterprise AI to organizations tackling the world's most complex challenges.

The broader landscape shows how much ground the industry still has to cover. Data from the S&P Global Corporate Sustainability Assessment (CSA) shows that only 36% of responding companies have a dedicated or integrated AI policy. Research from BSI paints a similar picture: just 27% of global businesses currently factor their AI footprint into their broader sustainability strategies, a notable gap given the surging energy demands of AI data centers.

Unlocking sustainable growth

Ethical AI goes beyond risk mitigation and regulatory readiness-it drives tangible economic value. By ensuring fairness, transparency, and robust data privacy, organizations protect their brand reputation while empowering their workforce to perform at their highest level.

When AI coordinates, executes, and optimizes routine business processes within an ethical context, employees are freed from manual tasks to focus on strategy, creativity, and human connection. This synergy between human potential and artificial intelligence forms the backbone of sustainable enterprise growth.

Organizations that understand this will reap the rewards as regulations tighten and societal expectations increase. However, ethical AI only becomes a reality when we work together. We look forward to continuing to work alongside our customers and partners as we embrace one of the most important challenges and competitive advantages in the AI era.

Read the SAP Global AI Ethics policy version 3.0. Find out more about the Autonomous Enterprise.

Matthias Medert is head of Sustainability at SAP.