The Moment of Reckoning for AI in Financial Services
Across financial services, AI is everywhere, and yet, true transformation is rare.
Banks, asset managers, and insurers are actively piloting tools, launching copilots, and building use cases at unprecedented speed. There is no shortage of innovation. But beneath the surface, a more uncomfortable reality is emerging: much of this activity is not fundamentally changing how these organizations create value.
Instead, firms are caught in a familiar pattern, one that mirrors the early days of digital transformation. Technology is being layered onto existing systems, improving efficiency at the margins, but leaving the core business model largely untouched.
It is in this context that guest speakers Alyson Clarke, Principal Analyst at Forrester; Deep Srivastava, Chief AI Officer at Franklin Templeton; and host Pranay Agrawal, CEO of Fractal, come together to explore a critical question: what does it actually take to scale AI in a way that transforms the enterprise, not just optimizes it?
When technology adoption masquerades as transformation
One of the most persistent misconceptions in the AI journey is the belief that deploying technology equals transformation.
Organizations often point to successful pilots or agentic systems as proof of progress. And while these efforts can deliver tangible benefits, cost savings, faster workflows, and incremental productivity, they rarely touch the heart of value creation.
Alyson Clarke draws a sharp distinction:
When I look at AI transformation, I see parallels with everything that happened for the last decade or so in digital transformation. I’ll be honest. For 99% of firms, at least in financial services, digital transformation failed because they didn’t actually transform. They digitized what the organizations already did, didn’t really change their business model or outcomes.
Alyson Clarke
Principal Analyst, Forrester Research
The same mistake is now repeating itself with AI.
At the center of this issue is what Clarke calls “use-case chasing, “a pattern where firms continuously experiment with isolated applications without stepping back to rethink the broader system. The result is a proliferation of initiatives that improve parts of the organization but fail to move the whole.
Alyson offers a simple but powerful test:
If you remove the AI, would the business model still fundamentally work the same way? If yes, it’s just an IT upgrade. If no, then you’re looking at real transformation.
Alyson Clarke
Principal Analyst, Forrester Research
By that standard, most AI initiatives today fall short.
True transformation demands something far more ambitious: reimagining end-to-end processes, redefining decision-making frameworks, and placing AI at the center of how the business operates. It requires leaders to move beyond task automation and toward decision-centric organizations, where intelligence is embedded into every layer of the enterprise.
The seduction and danger of early success
If the path to transformation is so clear in theory, why do so many organizations struggle in practice?
Part of the answer lies in the early success of AI itself.
Deep Srivastava has seen this dynamic play out firsthand. In the early stages of AI adoption, each successful use case generates momentum. Teams become energized. Stakeholders gain confidence. Investment increases.
“There’s a lot of euphoria… a high that happens every time you get a use case right.”
This momentum is not inherently bad. In fact, it plays an important role in building organizational buy-in. But it can also create a dangerous illusion that scaling more use cases will naturally lead to transformation.
At Franklin Templeton, this approach led to an explosion of activity. At one point, the organization was managing nearly 400 AI use cases simultaneously.
On the surface, this seemed like success.
But over time, a deeper realization emerged: while these initiatives were improving individual workflows, they were not shifting the business model.
“They created productivity, but they were all sitting on top of the existing model.”
This moment forced a difficult but necessary decision. The team dramatically narrowed its focus from 400 use cases to just four.
The transition was not easy. It introduced uncertainty and required a fundamental shift in thinking.
“At the beginning, you cannot describe what the end state looks like… that creates a completely new set of challenges.”
The initial excitement gave way to ambiguity, and even fear. But this is precisely the phase where real transformation begins. Moving from experimentation to impact requires letting go of volume in favor of value.
From Use Cases to Outcomes: A Shift in Leadership Thinking
The turning point in many AI transformations comes not from technology, but from leadership conversations.
As Srivastava describes, engaging in deeper, ROI-driven discussions with the CEO helped reframe the entire approach. Instead of focusing on the number of initiatives, the conversation shifted to outcomes: customer experience, investment performance, and long-term value creation.
This shift is subtle but profound.
Use-case-driven thinking asks: What can we build?
Outcome-driven thinking asks: What must we change?
Alyson Clarke reinforces this distinction:
“Right now, we’re seeing the firms that are chasing use cases, not outcomes… that’s all fine in the short term, but eventually they’ll need to focus on transformation and revenue growth.”
Use cases can generate quick wins, improve share price, satisfy boards, and demonstrate progress. But without a clear connection to the business model, they rarely scale in meaningful ways.
Moreover, the way these initiatives are measured often undermines their long-term impact. Proofs of concept are typically optimized for feasibility, not integration. They are built quickly, tested in isolation, and rarely designed for reuse.
“Proof of concepts are optimized for feasibility, not necessarily business integration.”
The result is fragmentation, systems that cannot scale, data that cannot be reused, and operating models that struggle to align across legal, risk, and data functions.
To break out of this cycle, leaders must prioritize fewer, larger initiatives, ones that are directly tied to strategic objectives and capable of reshaping the enterprise.
Re-centering AI Around the Customer
While much of the AI conversation focuses on internal efficiency, Pranay Agrawal emphasizes a different starting point: the customer.
In his experience working with Fortune 500 organizations, the most impactful AI strategies are those that drive revenue growth, customer experience, and product innovation.
Yet many firms still approach AI through a narrow lens, using data primarily to improve targeting or increase product sales.
This approach misses the larger opportunity.
Alyson Clarke highlights the risk:
“We saw this with digital transformation and digital banking. When everybody got their digital mobile app up to par, it is no longer a differentiator that drives loyalty.”
Basic AI capabilities, chatbots, transaction automation, and onboarding tools are rapidly becoming table stakes. They are necessary, but not sufficient.
The real competitive advantage lies in building systems that deepen customer understanding and create meaningful, long-term relationships.
This requires a shift from transactional interactions to experience-driven engagement, where AI helps organizations anticipate needs, deliver relevant solutions, and build trust over time.
The Promise of Anticipatory Personalization
One of the most transformative opportunities in AI is the ability to move beyond reactive personalization toward anticipatory experiences.
In financial services, this has profound implications.
Traditionally, sophisticated financial advice and tailored products have been reserved for high-net-worth clients. AI changes that equation.
There’s no reason why this industry cannot cater to everybody.
Deep Srivastav
Chief AI Officer, Head of AI and Digital transformation, Franklin Templeton
With AI, firms can deliver:
Highly personalized recommendations at scale
Lower-cost access to advanced financial tools
Context-aware interactions in real time
This democratization of capability has the potential to reshape the entire value chain, from investment research and decision-making to product design and portfolio advice.
But realizing this vision requires more than better algorithms. It demands a rethinking of workflows, data systems, and organizational structures.
AI helps organizations develop new products faster, reduce friction in customer engagement, drive efficiencies, create disruption, and make insights available to all decision-makers.
Pranay Agrawal
Co-founder and Chief Executive Officer, Fractal Analytics
These capabilities are not incremental; they are foundational to a new operating model.
Rethinking Data and Infrastructure for an AI-First World
As AI becomes more deeply embedded in decision-making, the role of data is fundamentally changing.
In the past, humans queried systems to retrieve information. Today, AI agents are increasingly making those calls autonomously.
This shift introduces new requirements:
Greater speed and scale
More dynamic, interconnected data systems
The ability for data sources to communicate with each other
Srivastava describes this as a move toward a new paradigm, where data is not just a passive asset, but an active participant in the system.
“You cannot just place an LLM on top of old systems. Intelligence must be embedded into the data systems.”
This has significant implications for infrastructure. Organizations must move beyond siloed data environments and treat data pipelines and models as shared, enterprise-wide assets.
At the same time, many firms are investing heavily in infrastructure without a clear link to outcomes.
Deep observes a common pattern:
“The foundations, innovation, and business are often siloed.”
Without alignment, even the most advanced infrastructure will fail to deliver meaningful value.
Measuring What Matters
A critical barrier to transformation lies in how success is measured.
Many organizations continue to rely on metrics such as cost reduction and productivity gains. While important, these indicators capture only a fraction of AI’s potential value.[GG1]
Ultimately, AI investments must be evaluated based on their ability to drive both business value and customer value.
Governance as a Continuous Discipline
As AI systems become more autonomous, governance becomes increasingly critical.
But traditional approaches, focused on one-time approvals and retrospective reviews, are no longer sufficient.
Instead, governance must be:
Continuous
Embedded
Proactive
Organizations need to establish clear guardrails around autonomy, explainability, and escalation, while ensuring accountability for outcomes.
“It’s really important that you have clear accountability for outcomes, not just models.” — Clarke
Srivastava adds that governance should not be an afterthought, but a central capability that evolves alongside the transformation.
The Future Will Belong to the Bold
AI transformation is not a technology initiative. It is a leadership challenge.
It requires organizations to align strategy, data, technology, and culture around a shared vision of value creation. It demands difficult choices, fewer use cases, deeper investments, and a willingness to embrace uncertainty.
The divide between leaders and laggards is already beginning to emerge.
Those who continue to chase incremental gains will fall behind.
Those who commit to transformation, who rethink their business models, prioritize customer value, and embed AI at the core will define the future of financial services.
In the end, success will not be determined by how much AI an organization deploys, but by how fundamentally it changes the way the organization works.
Written by Vanessa Thompson
In-person
Alyson Clarke helps financial services leaders understand and prepare for the future. She is a highly skilled expert with extensive industry experience in both banking and wealth management. Alyson has global expertise, having worked in Sydney, London, San Francisco, and now New York. She works closely with retail banking and other financial services executives to develop and execute on strategy, deliver superior customer experiences, and digitally accelerate their business. With more than 19 years of financial services industry experience, Alyson has held senior positions with several leading banks and wealth management firms. She has been quoted in many publications, including American Banker, Bloomberg, CNBC, The New York Times, and NPR. Alyson challenges thinking to help financial services firms drive change and generate new business propositions that create value for customers and shareholders.
Deep Srivastav is the Chief AI Officer, head of AI and digital transformation for Franklin Templeton. He is responsible for AI research and innovation, for managing the firm's AI driven digital platform and for driving digital transformation for the enterprise and clients.
In his prior role, he drove the development and deployment of Franklin Templeton's investment solutions through digital channels. He oversaw the digital product roadmap bringing personalized investment solutions to market. These solutions won the Barron's/MMI Industry Disruptor Award in 2021.
Mr. Srivastav is the co-author of "A New Approach to Goals Based Wealth Management", which was published in the Journal of Investment Management and won the Harry M. Markowitz Award in 2018. He is the recipient of "Innovator of the Year" award by WealthTech Americas for 2023. Mr. Srivastav holds a bachelor of engineering in electronics from The M.S. University of Baroda in India and an MBA with a focus in quantitative research and marketing from Indian Institute of Management Ahmedabad.
