The AI Remediation Bill: What Ungoverned AI Coding Really Costs

The News

At KubeCon North America 2026, ActiveState discussed the findings of its 2026 research report, The AI Remediation Bill, which examines the downstream costs organizations are absorbing as AI-assisted software development accelerates. Based on a survey of 307 enterprise security, governance, platform and engineering leaders, the report found that while 93% of respondents reported higher engineering productivity from AI-assisted development, engineers are spending at least 139 hours annually fixing or revisiting AI-assisted work. For an organization with 100 engineers, that represents more than 13,900 hours of engineering capacity each year.

The research also found that the costs extend well beyond development. Seventy-six percent of respondents experienced a software release delay related to unmanaged open source dependencies, 58% received an open source-related compliance or audit finding, and 32% confirmed a security incident traceable to an unmanaged dependency within the previous two years. The discussion connected these findings to broader industry challenges, including software supply chain governance, the EU Cyber Resilience Act (CRA), agentic AI accountability and the organizational changes required to make AI-assisted development sustainable.

Analyst Take

AI Productivity Is Real, but So Is the Remediation Bill

ActiveState’s AI Remediation Bill identifies an increasingly important distinction in enterprise AI adoption: generating more code is not the same as delivering more software value. Although 93% of respondents reported higher engineering productivity from AI-assisted development, engineers are spending at least 139 hours annually fixing or revisiting AI-assisted work. For an organization with 100 engineers, that represents more than 13,900 hours of engineering capacity.

The underlying issue is the growing volume of open source dependencies entering enterprise environments through AI-assisted workflows. AI coding tools are designed to solve problems quickly, often introducing libraries and dependencies that developers might not have independently evaluated. Without sufficient guardrails, those components can create additional security exposure, audit findings and remediation work. ActiveState found that 76% of respondents experienced release delays related to unmanaged dependencies, while 58% received an open source-related compliance or audit finding.

The financial impact extends beyond development. Engineering absorbs the rework, operations deals with delayed releases, and security and compliance teams manage the resulting risks. Only 31% of respondents said their organizations had formally quantified the total cost of unmanaged open source risk. That makes it difficult to understand the actual return on AI investment when productivity gains appear in one department while remediation costs accumulate elsewhere.

Software Supply Chain Governance Is Becoming an AI Requirement

The ActiveState findings highlight a structural weakness in how organizations manage open source software. Sixty-three percent of respondents said they address open source vulnerabilities only after scanners or security tools identify them, while half reported that new dependencies can reach production without prior review or approval. That reactive approach becomes increasingly difficult to sustain as AI coding assistants accelerate the introduction of new packages and their transitive dependencies. The EU Cyber Resilience Act adds urgency to the problem. Its phased requirements around software security, vulnerability reporting and software bills of materials (SBOMs) are pushing organizations toward greater supply chain transparency and accountability. Enterprises need visibility into the components used to build their applications, along with policies governing which dependencies can enter production environments.

The conversation also raised an important question about accountability as organizations expand agentic AI adoption. Agents increasingly have the ability to interact with enterprise applications, development environments and production data. When an agent takes an action that creates a security or operational problem, the organization deploying that technology remains responsible. Governance therefore needs to extend beyond code generation to include agent permissions, access controls and policy enforcement. The opportunity is to move security upstream rather than treating it as a final checkpoint. Automated dependency scanning, approved package sources, SBOM generation and policy enforcement need to become integrated components of the software development lifecycle.

AI Is Exposing the Organizational Gaps DevOps Never Solved

One of the more interesting threads in the ActiveState conversation was the comparison between today’s AI transition and the digital transformation movement that began more than a decade ago. Enterprises spent years pursuing cloud-native development and continuous delivery, but many never fully implemented the organizational practices needed to make those models successful. Development teams were measured on velocity, QA teams on quality, and security often remained a final gate rather than an integrated part of the delivery process.

Those organizational silos continue to affect software delivery. ECI Research found that 24% of organizations want to release code hourly, but only 8% currently achieve that cadence. The challenge is not simply having the right CI/CD tooling. It requires development, security, quality and operations to work toward shared outcomes rather than competing performance metrics. The conversation pointed to service-level objectives (SLOs) as one approach organizations are increasingly adopting to bring those teams into alignment.

AI is accelerating the need for that organizational change. ActiveState highlighted the growing preference for generalist engineering skills, while ECI Research found that 67% of organizations are hiring generalists over specialists. Engineers increasingly need to work across disciplines, understand AI-assisted development and apply the technical judgment necessary to evaluate generated code.

The transition will not happen uniformly. Organizations recognize the competitive pressure to adopt AI, but many are still figuring out how to change their workflows and governance structures. The companies that benefit most will be those that treat AI as an operating model transformation, bringing development, platform engineering, security and operations together around shared objectives.

Looking Ahead

ActiveState’s The AI Remediation Bill provides an important framework for measuring the next phase of enterprise AI adoption. As organizations expand AI-assisted development, engineering leaders will need to reconcile productivity gains with the costs of remediation, dependency management and security validation. Vendors that can demonstrate improvements across the full software lifecycle, rather than focusing exclusively on code generation, will have a stronger position in this market.

The CRA will add urgency to software supply chain governance, while the expansion of agentic AI will make accountability and operational controls increasingly important. Organizations investing in automated dependency management, policy enforcement and security earlier in the development process will be better positioned to scale AI adoption without accumulating additional remediation costs.

The larger opportunity extends beyond compliance. AI is creating an opportunity to revisit the organizational structures and delivery practices that limited earlier DevOps and cloud-native initiatives. The convergence of development, security and operations, supported by shared SLOs and more AI-capable engineering teams, could help enterprises finally achieve the software delivery model they have been pursuing for more than a decade. Generating more code is becoming easier. Building the engineering discipline to deliver that code securely, reliably and continuously is where the real competitive advantage will emerge.

Authors

  • Paul Nashawaty

    Paul Nashawaty, Practice Leader and Lead Principal Analyst, specializes in application modernization across build, release and operations. With a wealth of expertise in digital transformation initiatives spanning front-end and back-end systems, he also possesses comprehensive knowledge of the underlying infrastructure ecosystem crucial for supporting modernization endeavors. With over 25 years of experience, Paul has a proven track record in implementing effective go-to-market strategies, including the identification of new market channels, the growth and cultivation of partner ecosystems, and the successful execution of strategic plans resulting in positive business outcomes for his clients.

    View all posts
  • Samantha Weston

    With over 15 years of hands-on experience in operations roles across legal, financial, and technology sectors, Sam Weston brings deep expertise in the systems that power modern enterprises such as ERP, CRM, HCM, CX, and beyond. Her career has spanned the full spectrum of enterprise applications, from optimizing business processes and managing platforms to leading digital transformation initiatives.

    Sam has transitioned her expertise into the analyst arena, focusing on enterprise applications and the evolving role they play in business productivity and transformation. She provides independent insights that bridge technology capabilities with business outcomes, helping organizations and vendors alike navigate a changing enterprise software landscape.

    View all posts