Sinch research: 60% of executives are confident their AI programs are succeeding. Only 43% of the teams delivering them agree

The survey findings help explain why many organizations may struggle to move AI for customer communications from pilot projects to reliable production deployments. This is a familiar challenge for Sinch as enterprises scale conversational AI across messaging, voice, email, and other customer channels.

Overview of findings

      62% of enterprises already have AI agents live in production

      74% have rolled back or shut down a deployed AI agent

      81% rollback rate among organizations with mature governance frameworks

      75% report investment in trust, security and compliance, higher than AI development itself (63%).

      84% of AI engineering teams spend at least half their time on guardrails

      55% have to build custom infrastructure for cross-channel context

      86% have evaluated or are considering new communications providers

      98% are increasing AI investment in 2026

"For years, the AI conversation has largely focused on whether organizations can get AI into production. Our research suggests another challenge is emerging: making sure leadership and operational teams are working from the same reality," said Sophie Cheng, Chief Marketing Officer at Sinch. "Executives see investment, deployment, and strategic progress. The teams responsible for delivering AI for customer communications are more likely to see the operational challenges, from governance to communications infrastructure, that ultimately determine whether those initiatives succeed in the real world."

The research suggests that the disconnect stems from differing day-to-day experiences rather than strategic priorities. While executives track investment and deployment, implementation teams are more likely to confront the governance, reliability, and communications infrastructure challenges that determine whether AI for customer communications succeeds in production.

"The teams closest to AI deployments experience these operational challenges firsthand. When infrastructure falls short, they're the ones responsible for keeping systems running despite those limitations," said Daniel Morris, Chief Product Officer at Sinch. "At Sinch, we believe this reveals a deeper reality. The challenge is more about production readiness than AI capability. We're focused on building the infrastructure layer that helps teams move from experimentation to trusted, reliable, production-scale AI systems for communication."

One of the study's clearest findings was that communications infrastructure satisfaction was the strongest predictor of confidence in AI deployment. Organizations with stronger infrastructure foundations consistently reported greater confidence in scaling AI safely and effectively, outperforming factors such as governance maturity, deployment experience, and investment levels.

Full Report: The AI Production Paradox

Blog: The AI Production Paradox: What 2,527 enterprise leaders told us

Methodology

The study was conducted between January and February 2026 by Sinch in collaboration with an independent research institute. A total of 2,527 senior decision-makers participated across the United States, United Kingdom, Australia, Brazil, Germany, France, India, Singapore, Mexico and Canada. Respondents represented large enterprises across financial services, healthcare, telecommunications, technology, retail and professional services. Participants included C-suite executives, vice presidents, directors and managers responsible for AI strategy, implementation and operations.

For more information, please contact:

Frerik Hallstan

Director Corporate Communications

Mobile: +46 761 15 38 30

E-mail: fredrik.hallstan@sinch.com

  Shannon Hames
  PR & Communications Manager, North America

E-mail: shannon.hames@sinch.com

Mia Nordlander

SVP Investor Relations & Sustainability Mobile: +46 73 511 53 95

E-mail: mia.nordlander@sinch.com