AI's Database Dilemma: Redgate Unveils a New Layer of Control

CAMBRIDGE, England – July 21, 2026 – The enterprise world is in an arms race, but the weapon of choice isn't hardware—it's artificial intelligence. From automating workflows to generating code, AI is being woven into the fabric of modern business at a staggering pace. Yet, beneath this frantic push for innovation lies a critical vulnerability: the database. As AI-powered agents begin writing and deploying their own code, they are threatening the very foundation of data integrity and stability. Addressing this existential risk, Redgate Software today announced a major evolution of its Flyway Enterprise platform, introducing a change control layer designed specifically for the age of AI.

"AI is the biggest opportunity in a generation to accelerate software creation, and enterprise leaders want to seize it with confidence," said Jakub Lamik, CEO of Redgate. "The organizations that win will be the ones that pair AI speed with strong foundations." The company's new Flyway Enterprise MCP Server aims to be that foundation, providing a governed bridge between AI coding assistants and the sensitive database environment.

The Governance Gap: AI's Unseen Risk

The rush to leverage AI has created a significant governance vacuum. Recent industry data reveals a startling trend: while AI adoption in database management has nearly tripled year-over-year to 44%, a staggering 77% of organizations admit they lack the formal control and data governance processes to manage it. This disconnect between ambition and oversight is a ticking time bomb.

AI coding assistants like GitHub Copilot and Google Gemini can generate database schema changes in seconds, a task that once required careful human deliberation. While this accelerates development, it also floods CI/CD pipelines with code that often lacks context, bypasses established policies, and goes unaudited. The result is an escalating risk of deployment failures, data corruption, and costly production incidents. The problem is so acute that Gartner projects by 2027, 40% of enterprises will be forced to decommission autonomous AI agents due to critical governance gaps exposed after such incidents occur.

"AI without strong foundations doesn't just fail slower, it fails faster, and in ways much harder to debug, fix and explain to the business," one industry advocate noted. This new reality demands a paradigm shift from simply enabling speed to ensuring safe, controlled velocity.

A New Guardrail for Agentic Workflows

Redgate's answer to this challenge is the Flyway Enterprise MCP Server. The term 'MCP' (Model Context Protocol) signifies an emerging standard for secure AI-to-system interaction, and this server is designed to be the definitive guardrail for AI-driven database modifications. It functions as an intelligent intermediary, intercepting any change proposed by an AI agent and subjecting it to the same rigorous validation as human-authored code.

This 'agentic approach' allows developers to leverage AI for speed while ensuring every change is captured, validated, and auditable. The platform automatically enforces a library of over 95 customizable policy rules, blocking non-compliant changes before they can ever reach a production environment. It generates deterministic, dependency-aware migration scripts to ensure consistency and continuously monitors for 'drift'—manual or un-pipelined changes that deviate from the expected state.

This move places Redgate within a nascent but critical market focused on taming AI's interaction with core infrastructure. Competitors like Liquibase are promoting similar frameworks, signaling a broader industry recognition that AI-generated code cannot be trusted without a specialized layer of governance. By building a complete audit trail directly into the workflow, these solutions transform compliance from a reactive, after-the-fact burden into a proactive, automated process.

Extending Control Beyond the Transactional

Perhaps most strategically, Redgate's vision extends beyond traditional transactional databases. The new release introduces advanced capabilities for Databricks, with similar support for Snowflake on the roadmap. This is a crucial acknowledgment that modern AI and analytics initiatives are increasingly built upon data lakehouse architectures, not just operational databases.

The pipelines feeding AI models and business intelligence dashboards are just as critical, and often more complex, than production systems. Without disciplined schema management in these analytics estates, data quality erodes, and the reliability of AI-driven insights becomes questionable. By extending its governed approach to Databricks and Snowflake, the company aims to provide a single, consistent control plane for an organization's entire data landscape.

This ensures that the data fueling machine learning models is as rigorously managed as the data processing customer transactions, addressing a key barrier to successful AI adoption: the lack of 'AI-ready' data. It's a holistic approach that recognizes that in the age of AI, governance cannot be siloed.

From Theory to Practice: The Tangible Impact

While the threat of AI-driven chaos is significant, the benefits of taming it are concrete and measurable. The automation framework underpinning Flyway Enterprise has already demonstrated its value for customers navigating complex deployments. Telematics giant Verizon Connect, for example, reported slashing operational support requests by 90% and reducing change lead times by up to 98% after implementing Redgate's solutions.

These dramatic improvements stem from targeting the root causes of deployment failure: manual steps, inconsistent environments, and a lack of pre-deployment validation. By automating these processes, the platform not only reduces risk but also frees up development and operations teams to focus on innovation rather than firefighting. The integration of Flyway Enterprise with Redgate Monitor, an observability solution, further enhances this by providing leaders with estate-wide insight into database health and change activity.

As enterprises continue to embed AI deeper into their operations, the ability to automate, govern, and audit every database change will no longer be a best practice but a fundamental requirement for survival and success. Solutions that provide this critical control layer are poised to define the next era of sustainable, AI-powered innovation.