LocalStack released the 2026 State of AI in Cloud Development report conducted by Dimensional Research. It is available for download at: https://www.localstack.cloud/state-of-ai-cloud-development-2026.
Dimensional Research Principal Diane Hagglund said: “The impact AI innovations have had on writing code is well established but the broader implications of the role of AI in testing, staging and deploying that code has not been part of those conversations. This research extends our understanding of how AI is impacting teams and how they can get even more from their investments as they move through all stages of the software development lifecycle.”
The survey results found that enterprise software development teams grant AI agents access to cloud environments supporting various stages of the software development lifecycle even though they express concerns about how doing so might affect security, costs and application reliability.
Key survey findings include:
- 79% of respondents give AI agents access to cloud environments for developing, testing or staging code at least occasionally and 18% said agents always have access
- 97% have concerns about granting that access for AI agents, citing leaked credentials, misconfigurations, rising cloud costs and failures in production
- 26% said AI agents can access production environments, and 55% said agents can access staging
- 65% generate at least twice as much code since adopting AI but developers estimated only 61% of AI-generated code is released to production
- 26% of developers said AI-generated code has significantly more bugs than code created by humans, while only 6% of development team managers said the same
- 58% said leadership expects them to control token and cloud costs without giving them the information to do so
“These responses align with what we hear from our customers every day,” said Waldemar Hummer, LocalStack co-founder and CTO. “AI is generating code faster than teams can confidently ship it, making testing and validation the new bottleneck. We help developers and AI agents close this gap with fast, safe, isolated cloud environments that enable rapid iteration without the risks and costs of traditional cloud infrastructure.”
