The new Stanford Institute for Human-Centered Artificial Intelligence (HAI) AI Index 2026 report shows enterprise AI adoption expanding. Yet readiness remains uneven as those capabilities move into systems that support finance, supply chain, and decision-making.
In ERP environments, where transactions, controls, and reporting structures define how data is interpreted, that gap becomes harder to manage as AI outputs must align with established processes, audit requirements, and business context.
Organizations can prepare ERP data for analytics and AI, but the report shows that outputs capable of supporting business decisions still depend on how systems preserve context, enforce controls, and maintain traceability from input data to outcome.
AI Is Everywhere, but It’s Not Enterprise-Ready
The AI Index 2026 shows that enterprise AI has moved beyond experimentation, but not into mature, repeatable deployment. AI is now present in most organizations, with 88% reporting use in at least one business function, while generative AI adoption continues to rise.
Yet broad adoption does not translate into deep operational change. Much of today’s enterprise AI activity remains limited to augmentation, productivity support, and targeted workflow improvements rather than fully autonomous business processes.
The practical question is whether organizations can validate, govern, integrate, and scale AI. A model that performs well in a benchmark or pilot can still struggle with customized processes, fragmented data, exception-heavy workflows, and regulated operating conditions.
Analysis
What This Means for ERP Insiders
AI use is expanding faster than operational maturity. Organizations are applying AI in workflows before establishing controls, creating uneven outcomes as usage scales.
Trust and Governance Are Becoming Core Business Challenges
As enterprise AI use expands, so do the risks associated with it. The AI Index 2026 tracks 362 notable AI incidents in 2025, up from 233 in 2024, reflecting issues such as model failures, misuse, bias, and security exposure as AI is deployed more broadly across organizations.
Public sentiment reflects the same imbalance. Nearly 60% of respondents believe AI’s benefits outweigh its drawbacks, up from 55%, yet 52% express concern about AI-powered products and services. The gap widens further between experts and the public, with 73% of experts expecting AI to improve work compared with just 23% of the public.
That imbalance carries into enterprise systems, where organizations can deploy AI across workflows yet face a different constraint: whether outputs can be validated, explained, and aligned with business controls. Governance needs to be embedded into operating models and system-level controls so that AI-assisted decisions can be tested, traced, and defended.
Analysis
What This Means for ERP Insiders
Trust divergence is reshaping enterprise risk exposure. Public skepticism and rising incidents create external and internal pressures that redefine how organizations manage AI accountability.
AI Infrastructure and Sovereignty Are Now Board-Level Risks
AI is becoming a board-level concern because compute capacity, data center availability, semiconductor supply, energy demand, and cloud geography shape what enterprises can deploy. The AI Index 2026 highlights the concentration of data centers and chip supply chains, which places growing dependency on a small number of providers.
At the same time, AI sovereignty is moving from policy debate into enforcement. Governments are increasing scrutiny of where data is stored, where inference occurs, which models are used, and which vendors control critical infrastructure, which raises new considerations for compliance and operational control.
AI strategy now has to account for data residency, regulatory exposure, and regional cloud choices, where decisions about infrastructure and vendors affect how systems operate. The ability to expand AI use depends as much on infrastructure strategy as on model capability.
Analysis
What This Means for ERP Insiders
Infrastructure choices are becoming strategic lock-ins. Early platform commitments will shape long-term flexibility, constraining how enterprises adapt AI across jurisdictions and vendors.
Talent Development Will Determine Whether AI Delivers Value
The AI Index 2026 points to measurable productivity gains, including a 14% improvement in customer service productivity and a 26% improvement in software development productivity. However, it also notes a 20% decline in the number of young software developers in the US, which raises questions about how AI is reshaping talent pipelines.
AI changes how organizations build and sustain skills. Productivity gains do not translate directly into organizational capability. Teams still need to redesign processes, evaluate AI outputs, manage exceptions, interpret risk, and connect model behavior to outcomes.
Organizations that invest in tools without developing these capabilities risk creating isolated pilots rather than scalable transformation, while those that embed AI talent development into operating models are more likely to sustain value.
Analysis
What This Means for ERP Insiders
Reskilling and entry-level training determine sustained AI value. Without rebuilding entry pathways, organizations risk short-term gains but weaken the pipeline that sustains long-term capability.
About Us
ERP Today covers how ERP, cloud, and AI change the way businesses run. Our editors speak with practitioners, vendors, and analysts to surface the technology, contracts, and risks that matter for enterprise leaders.
Alongside our newsroom coverage, we run in‑person summits where ERP leaders compare notes on programs like yours, and a research practice that turns reporting like this into organization‑specific briefings and content.
SAPinsider first published a version of this article on April 22, 2026.
