After many years of research and development, mostly confined to academia and projects in large organizations, artificial intelligence (AI) and machine learning (ML) are advancing into every corner of the fashionable enterprise, from chatbots to tractors, and financial markets to medical research. But corporations are struggling to maneuver from individual use cases to organization-wide adoption for several reasons, including inadequate or inappropriate data, talent gaps, unclear value propositions, and concerns about risk and responsibility.

This MIT Technology Review Insights report, commissioned by and produced in association with with JPMorgan Chase, draws from a survey of 300 executives and interviews with seven experts from finance, health care, academia, and technology to chart elements which are enablers and barriers on the journey to AI/ML deployment.

The next are the report’s key findings:

Businesses buy into AI/ML, but struggle to scale across the organization. The overwhelming majority (93%) of respondents have several experimental or in-use AI/ML projects, with larger corporations prone to have greater deployment. A majority (82%) say ML investment will increase in the course of the next 18 months, and closely tie AI and ML to revenue goals. Yet scaling is a significant challenge, as is hiring expert employees, finding appropriate use cases, and showing value.

Deployment success requires a talent and skills strategy. The challenge goes further than attracting core data scientists. Firms need hybrid and translator talent to guide AI/ML design, testing, and governance, and a workforce technique to ensure all users play a job in technology development. Competitive corporations should offer clear opportunities, progression, and impacts for employees that set them apart. For the broader workforce, upskilling and engagement are key to support AI/ML innovations.

Centers of excellence (CoE) provide a foundation for broad deployment, balancing technology-sharing with tailored solutions. Firms with mature capabilities, normally larger corporations, are likely to develop systems in-house. A CoE provides a hub-and-spoke model, with core ML consulting across divisions to develop widely deployable solutions alongside bespoke tools. ML teams needs to be incentivized to remain abreast of rapidly evolving AI/ML data science developments.

AI/ML governance requires robust model operations, including data transparency and provenance, regulatory foresight, and responsible AI. The intersection of multiple automated systems can bring increased risk, comparable to cybersecurity issues, illegal discrimination, and macro volatility, to advanced data science tools. Regulators and civil society groups are scrutinizing AI that affects residents and governments, with special attention to systemically vital sectors. Firms need a responsible AI strategy based on full data provenance, risk assessment, and checks and controls. This requires technical interventions, comparable to automated flagging for AI/ML model faults or risks, in addition to social, cultural, and other business reforms.