Session Summary

The Evolving Build-Versus-Buy Decision in Financial AI

Technological Advancements Reshaping Strategic Choices

Andrei from BNP Paribas illuminated how rapid technological advancements are fundamentally transforming the build-versus-buy calculus for financial institutions. The reduced costs of AI tools and significantly improved capabilities of pre-built models have made AI implementation more accessible than ever before. However, he cautioned against premature deployments, advocating instead for iterative testing and rigorous data integrity protocols before full-scale implementation.

This measured approach reflects a broader industry trend: whilst AI tools have become more affordable and powerful, successful deployment requires careful planning and robust testing frameworks to ensure reliability in high-stakes financial environments.

The Critical Foundation: Data Engineering

Hira from Visionet emphasised a fundamental truth often overlooked in AI discussions: generative AI cannot resolve foundational data issues without robust data frameworks already in place. She stressed the critical role of data engineering, arguing that financial institutions must establish:

• Hybrid platforms that combine multiple data sources seamlessly 
• Deterministic frameworks ensuring predictable, auditable outcomes 
• Robust data governance structures tailored for regulated environments 

For regulated financial institutions, these foundational elements aren't optional—they're essential prerequisites for successful AI implementation. Without proper data infrastructure, even the most sophisticated AI models will produce unreliable results.

The Decline of Traditional SaaS Models in Enterprise AI

In-Housing AI Capabilities: A Strategic Shift

Leon from Elsewhen predicted a significant decline in traditional SaaS models as companies increasingly bring AI capabilities in-house. This shift is driven by two powerful forces:

• Decreasing software development costs making custom solutions more economically viable
• Strategic value of bespoke solutions tailored to specific organisational workflows 

Leon advocated for custom AI platforms designed around unique business processes, warning against fragmented agent-driven solutions that lack proper governance structures. This perspective challenges the conventional wisdom that purchasing off-the-shelf solutions always represents the most efficient path forward.

BlackRock's In-House Platform Strategy

Mari from BlackRock provided a compelling case study of this trend in action. Her firm's decision to develop an in-house AI platform was driven by several strategic imperatives:

• Unifying disparate processes across the organisation
• Enhancing innovation through proprietary data leverage
• Maintaining centralised governance and control Implementing agentic workflows securely
• Transitioning from experimentation to production environments safely

Both Mari and Leon noted growing demand for scalable, enterprise-grade AI platforms as organisations recognise the limitations of off-the-shelf solutions for complex, regulated environments. The message is clear: whilst pre-built solutions may suffice for simple use cases, sophisticated financial institutions increasingly require bespoke platforms that align precisely with their unique requirements and regulatory obligations.

Governance, Collaboration, and Strategic Partnerships

Agile Development and Startup Partnerships

Andrei described BNP Paribas's approach to platform development, emphasising agility and iterative methodologies. The bank's partnership with startups like Mistral to co-develop cutting-edge solutions exemplifies a hybrid strategy: leveraging external innovation whilst maintaining strategic control over core capabilities.

This collaborative approach allows established financial institutions to access innovative technologies and methodologies from agile startups whilst ensuring solutions meet stringent regulatory and operational requirements.

Multi-Layered Governance Strategies

Mari elaborated on embedding governance through comprehensive, multi-layered strategies:

• Platform-level guardrails ensuring baseline security and compliance
• Process-level reviews involving cross-functional teams
• Continuous monitoring and adjustment mechanisms
• Clear accountability structures across the AI lifecycle 

These governance frameworks aren't merely bureaucratic obstacles—they're essential enablers of responsible AI adoption in regulated environments. Proper governance allows financial institutions to innovate confidently whilst maintaining regulatory compliance and managing risk effectively.

Opportunities for Startups and Overcoming Institutional Inertia

Targeting Niche Problems

The panel reached consensus on strategic advice for startups entering the financial AI space: target niche, underexplored problems rather than attempting to solve the largest challenges already addressed by major firms. This approach allows startups to:

• Differentiate their offerings in crowded markets
• Demonstrate value quickly in specific use cases
• Build credibility before expanding to broader applications
• Avoid direct competition with well-resourced incumbents 

Addressing Governance Challenges and Institutional Inertia

Panellists identified governance challenges and institutional inertia as significant barriers to AI adoption within financial institutions. The path forward requires striking a delicate balance between innovation and regulatory compliance.

Financial institutions must cultivate organisational cultures that embrace experimentation whilst maintaining rigorous risk management. This balance—though challenging—is essential for achieving meaningful progress in AI adoption without compromising the stability and trustworthiness that define successful financial services organisations.

Conclusion: Strategic Imperatives for Financial AI Success

The panel discussion revealed that successful AI implementation in financial services requires far more than simply purchasing the latest technology. Financial institutions must carefully evaluate whether to build, buy, or wait based on their unique circumstances, capabilities, and strategic objectives.

Key takeaways for financial services leaders include:

• Prioritise robust data engineering and governance frameworks before deploying AI solutions 
• Consider in-housing critical AI capabilities as development costs decrease and strategic value increases 
• Embrace iterative, agile approaches to platform development 
• Establish multi-layered governance structures that enable innovation whilst ensuring compliance
• Collaborate strategically with startups and external partners to access cutting-edge capabilities
• Balance innovation ambitions with regulatory requirements and risk management imperatives 

As AI technologies continue evolving rapidly, financial institutions that successfully navigate these strategic decisions will gain significant competitive advantages. The future belongs to organisations that combine technological sophistication with robust governance, strategic vision with operational excellence, and innovation with responsibility.

Ready to transform your financial institution's AI strategy? Contact industry-leading consultancies and technology partners to develop a bespoke roadmap aligned with your unique requirements and regulatory environment.