The adoption of AI in software development is accelerating, with businesses recognising its transformative potential for building new software experiences. Recent Forrester research reveals that companies implementing AI-powered development, security, and operations (DevSecOps) platforms are seeing substantial returns, including up to 483% ROI over three years.
Yet technology leaders face significant challenges: integrating AI effectively and responsibly across the development lifecycle, ensuring security and compliance, and building sustainable frameworks for the future.
GitLab’s EMEA vice president, Michel Isnard, discusses how AI can augment software development and shares insights on creating lasting value through strategic implementation.
Q: What’s driving the urgency around AI adoption in software development?
A: In customer conversations, I’m hearing that technology leaders can see AI’s massive potential for software development, but they’re wrestling with a few obstacles.
Three challenges come up: intense pressure to ship secure software faster, the requirement for rock-solid security guardrails around their AI, and many, especially in regulated industries, need AI solutions that work in air-gapped environments.
While the immediate benefits of code assistance and AI are clear, the long-term impact on organisations will be game-changing. That makes this an exciting time—we’re not just implementing AI; we’re using it to transform how software gets built and delivered.
Q: How significant is the impact of AI on development teams?
A: The impact is profound, but people often get caught up asking the wrong question: “Will AI replace developers?” The short answer is “no.” While AI is freeing developers from mundane, repetitive tasks, it’s also empowering them to focus on creatively solving problems and innovating.
We’re seeing this play out in a few key ways. Development cycles are becoming dramatically faster and more efficient, code quality is improving while bugs are decreasing, and teams are collaborating more effectively than ever.
GitLab research found that developers spend about a quarter of their time on code generation, while the remainder is spent on testing, maintenance, and identifying security vulnerabilities. There’s a huge opportunity to leverage AI across the software development lifecycle, allowing developers to focus on bigger-picture priorities.
We’re not just implementing AI; we’re using it to transform how software gets built and delivered
The real question isn’t whether AI will replace developers but how businesses can harness AI across the entire development lifecycle to accelerate innovation, drive tangible results, and achieve business value.
Q: How should decision-makers approach AI implementation?
A: The key is to be purposeful in your approach. Technology leaders should build a data-driven AI foundation, which requires rethinking how data is handled across the entire development lifecycle.
It is critical to take a measured, strategic approach rather than transforming everything overnight. Leaders must be intentional about where they start, choosing the areas that will deliver immediate productivity gains and building from there.
Significant changes mean altering human behaviours in technology teams, which can be the most challenging part of implementing a new approach to enterprise software. Training and upskilling are vital in the roll-out stage’s design.
Q: Why should software development leaders consider a unified approach to their AI strategies?
A: As AI models are integrated into applications, cross-team collaboration ensures seamless software development, deployment, and maintenance. Using a complete platform like GitLab instead of piecing together separate AI tools allows you to see your entire software development process in one place. This means AI isn’t just helping with isolated tasks. When AI has that complete context, you can deliver secure software much faster and with better insights than you’d get from disconnected tools.
When handling enterprise-scale software development, this kind of unified intelligence isn’t just nice to have—it’s essential for moving fast while staying secure.
Q: What real-world impact are leaders seeing from digital transformation in development?
A: We aren’t seeing just incremental improvements but a fundamental shift in software development.
New efficiencies extend beyond just coding teams, which are experiencing smoother onboarding, faster deployment cycles, and more robust security protocols. Perhaps most importantly, organisations are reducing system downtime. For example, space exploration company Intuitive Machines achieved a 99% reduction in downtime while accelerating their release cadence tenfold.
Q: Where is the future of AI headed in software development?
A: We’re at a turning point in how AI helps us build software. The future of applications is shifting toward intelligent, adaptable AI agents surpassing traditional software’s limitations. Rather than interacting with fixed interfaces and preset workflows, users will engage with AI agents that respond intuitively and learn over time.
As AI assistants move beyond reactive prompt-based interactions to proactive problem-solvers, they will anticipate developers’ needs and offer real-time suggestions to help streamline the entire SDLC.
We’ll also see a shift toward on-premise AI deployment. As open-source models become more cost-effective and accessible, organisations will increasingly opt to run customised versions within their own data centres.
Open source or open core providers like GitLab are used by highly regulated companies because, increasingly, these complex organisations need bespoke versions of software and still want the benefits of community collaboration behind the safety of their own firewall.
Q: What should technology leaders consider now to future-proof their businesses?
A: To adopt AI responsibly, organisations should develop AI best practices in low-risk areas before allowing additional teams to adopt AI.
Success in today’s AI landscape demands proactive planning. The foundation must be data-driven—fundamentally restructuring your development lifecycle to centralise data collection and management. When effectively implemented, this becomes the lifeblood of your AI initiatives, supercharging output quality while enabling precise measurement of success.
Governance can’t be an afterthought. Leaders should establish AI steering committees with authority and clear operational boundaries. This isn’t about restricting innovation but creating guardrails that enable teams to move quickly and confidently.
Tech leaders need to take steps now to ensure their technical systems are ready to support AI tools and prepare for multi-model approaches.
How we adopt AI now will significantly impact its future role in business and society.
For more information please visit: about.gitlab.com
