Falling Behind on AI? Here’s How Banks and Credit Unions Can Catch Up Fast
By Keith Fulton, Chief Data Officer, Jack Henry™
Simple Subscribe
Subscribe Now!
Generative AI is poised to transform the world as we know it, but it’s no longer a futuristic concept. It’s already reshaping how financial institutions operate.
From enhancing efficiency, improving customer and member service, and driving innovation, AI is becoming a competitive necessity.
But here’s the good news: it’s not too late to get in the game.
Maximizing these benefits will require a practical approach and the agility to adapt to ongoing technological advancements and innovations. With the right mindset and strategy, you can take practical, strategic steps today to unlock immediate value while laying the groundwork for long-term success.
Let’s explore four key insights to help you catch up (and jump ahead) along your AI adoption journey.
Want more insights like these? Check out Jack Henry’s content hub: Maximize the Impact of AI to Ignite Innovation
1. Innovation Takes Time, and AI is No Exception
While AI’s potential feels limitless, history reminds us that transformative technologies often take time to find their most impactful applications. But when it comes to AI, waiting isn’t an option. That’s why early adopters are experimenting now to learn, fail, and refine their approach.
When motion pictures first emerged, the initial use case was simply filming a live play and projecting it onto a screen. It took 16 years before filmmakers realized they could shoot from multiple angles and edit scenes together – unlocking the true storytelling power of the medium.
AI is following a similar path.
As capabilities grow, so will the creativity behind how we apply it – meaning many of the most valuable use cases haven’t been invented yet.
Just as early steam engines were met with skepticism over safety concerns, AI is facing similar scrutiny today. But safety shouldn’t be a reason to avoid AI. It should be something we build into its use from the start.
Think of AI like a dog park: within clearly defined boundaries, playful pups have the freedom to explore, play, and innovate. Establishing clear guardrails and usage policies for your teams ensures responsible use of AI while keeping risks in check.
Key takeaways:
- Err on the side of allowing experiments and pilots (subject to guardrails).
- Embrace failure as part of the innovation process.
- Build a culture that supports experimentation.
2. Understanding the Capabilities and Limits of Generative AI
Generative AI uses giant neural networks – often called large language models (LLMs) – to create content.
Understanding how LLMs work, what they’re good at, and what they’re not good at is essential to using them effectively.
LLMs excel at reading and writing, but that’s essentially all they do. In many ways, Generative AI is just a sophisticated prediction engine, trained on millions of words and pixels – predicting what should come next. And while it’s very good at prediction, it doesn’t actually “think.”
Because of this, today’s AI models can also get befuddled by complicated logic.
While AI might be able to follow some basic “if-then-else” rules, AI is easily confused by complex or nested rules, overwhelmed by large amounts of data, and struggles with basic math.
For big data tasks, it’s best to use machine learning (ML) or other statistical methodologies that are designed to handle large-scale, structured data.
Key takeaways:
- Use AI for content creation, summarization, and accountholder communication. Until it’s more advanced, stick to asking it for image analysis, reading, and writing.
- Generative AI’s weaknesses with numeric data suggest sticking to statistical methods and traditional ML for large-scale numeric analysis.
- Stay informed about emerging agentic AI, but treat it as experimental for now.
3. The Rapid Decline of AI Costs and What It Means for You
AI capabilities are advancing at an extraordinary pace, with models leapfrogging each other month by month.
In just three years, the cost of deploying generative AI and LLMs has dropped by a factor of 1,000. As energy requirements decrease and chip performance improves, the cost to serve AI approaches zero.
This shift has major implications for financial institutions.
While AI may feel new and worth paying a premium for today, it’s quickly becoming a standard feature, and standalone AI tools are losing their edge – tech giants like Microsoft and Apple are already embedding AI into their products at no extra cost.
To enable low-cost switching to better, cheaper solutions, products you build or buy should be provider-agnostic whenever possible. Avoid locking into multi-year contracts with AI providers (regardless of the discount) so you can take full advantage of future cost efficiencies.
It’s also important to recognize that charging separately for AI features may not be sustainable. As AI becomes a standard part of most technology offerings, its value will be measured by how well it’s embedded – not by extra costs.
Key takeaways:
- Choose provider-agnostic solutions whenever possible to stay flexible.
- Avoid multi-year contracts with AI providers – no matter the discount, to ensure you can benefit from the coming cost efficiencies.
- If you’re building AI-based tools, don’t count on being able to charge for them separately forever. Focus on embedding them into existing workflows.
4. AI Will Soon be Everywhere, So Don’t Get Left Behind
Just like spellcheck or Excel formulas, AI will become a standard part of every tech product. Think Apple phones and how they include AI at no additional cost. Banks and credit unions must shift from viewing AI as a luxury to treating it as a baseline capability.
The key takeaway is clear: AI will soon be everywhere, powering both small and large use cases.
Your goal shouldn’t be to charge more for AI. It should be to use AI to deliver better service, faster decisions, and smarter operations to ultimately enhance your accountholder experience.
Key takeaways:
- Generative AI will eventually be inside most of our products, for small and large use cases.
- Think of AI as a teammate, one that boosts productivity across your organization.
The Bottom Line: Start Now, Start Smart
If you haven’t engaged with AI yet, the time to act is now. Your competitors are already learning, adapting, and gaining efficiencies. But it’s not too late to catch up and even jump ahead.
Remember:
- Don’t buy AI solutions for statistical machine learning problems.
- Seek providers who are embedding AI in their products at low-to-no extra cost, rather than special purpose AI providers.
- AI assistance will level up your employees, like having a helpful teammate at everyone’s fingertips.
Make sure your financial institution isn’t just keeping up but is leading the way when it comes to AI.
