Financial Services is the comprehensive term for a vast industry with some of the world's largest global companies, dealing with some of the most risky data sets and assets in the general public.
For years, decades, and even centuries, these traditional institutions have built consumer trust by ensuring that the money of the masses is safe, financial investments are sound, and their assets are rooted in the future.
Financial services regularly treat people on the worst days of their lives, dealing with the greatest vulnerability of people in often disrupted economic situations.
FS agencies will not treat this burden lightly. They are often known in the technology world for the troublesome legacy systems and massive datasets that need to be fought across siloed systems. Their approach to conservative and stable innovation means instilling trust from customers, but times are changing.
The rapid rise in AI and the increased investment, adoption and deployment of AI puts pressure on more conservative industries.
The advent of AI offers more risk to more risky and disadvantaged industries that are tasked with approving and managing mortgages, addressing the emotional and financial consequences of succession, and dealing with the product of the livelihood of people who advise the grantor and first-time parents' financial plans.
But AI is here, already adopted by financial institutions, pushing up sectors and systems at a vitality and a pace that doesn't seem to show signs of slowing down.
However, given the FS' interests, speakers at Digit's first FS Technology Summit in Edinburgh encouraged hundreds of representatives present to show pause, attention and critique as they launched AI to the organization.
“Things are moving so fast, you have to be able to be agile, adapt and innovate. That's especially difficult for financial services organizations.”
“We need to manage risks and build trust,” he said.
Jumping to the AI bandwagon is not a viable option for FS companies. The risks and lack of trust that this can arise are too expensive to pay.
But expectations for AI are very high, says a study by analytics companies McKinsey and Bloomberg. UK GDP is added to the global economy every year due to Genai from the 2020s to the 2030s. McKinsey predicted.
Similarly, Bloomberg predicted 10% to 12% of all technology spending in an organization is directed towards generating AI.
McMahon was important, he admitted. Especially what does this actually look like?
“There are people who manage the budget of an organization. Can you imagine 10% of that budget, everything you see, engraving? And can you imagine putting it on a new, new technology that just hit the scene? It's so dramatic!”
McMahon encourages others to be “very important” in investing in AI. This is an enthusiasm that covers almost all of AI adoption, this terror left behind, the terror of freeing investors, the fear of losing to competitors.
When implementing AI in businesses, McMahon said it needs to be treated like an investment in other new technologies. In fact, you can't push it into the system.
Returning to basics
Looking at companies that have best integrated AI into their systems, it's very different from providing AI chatbots for basic customer service queries. Generation AI is a general purpose technology that can be applied to a variety of domains.
When companies find a place where they can use this technology in a particular vertical “not just following hype,” they distinguish them from those jumping on the bandwagon.
“Giving AI assistants to everyone is not going to drive ROI automatically,” says McMahon. “The driving ROI is when you drill down to a specific vertical within your organization. We need to have such specific questions in the organization and find out where the value of AI will be translated for you.”
So how do you identify positive use cases for AI?
McMahon proposes maintaining six key capabilities of AI, as a central framework for figuring out whether it creates a compelling adoption case.
These include search, summarizing, analysis, generation, translation, and transcription.
“If this isn't these six from Genai, you're just trying to put a square peg in the round hole,” he said.
It is also important to go beyond these fundamental principles. Just because AI can summarise things doesn't mean there's a business case for this integration.
AI is also touted as a productivity improvementr, but again, McMahon is critical of the data.
“Since adopting AI, I've said, 'I'm 10 times more productive than me,' and I want to try this, where is the data? ”
Is this just a feeling, or is it people really productive?
“Maybe they're faster. They don't necessarily mean they're really putting out better quality codes.”
It is important to note that McMahon has not beaten AI. He is simply critical of the mid-term implementation efforts of companies that throw AI into systems without proper foresight or adequate foundational foundation.
And in reality, when it comes to AI racing, companies are still waiting at the starting line.
a Report from BCG It turns out that around 40% of companies in different sectors have not yet taken action on AI adoption, but only 10% are in the stage where they can scale. Half of them were in the planning stages of an AI journey. Rush may be more stories and promises than reality.
More than that, it shows that there is a strong barrier to translating enthusiasm for AI into actual investment returns.
Ensuring this is achievable, you're back to the basics of being ready for AI and ready for business.
At the start of the digital transformation journey, organizations may struggle to implement AI effectively without this foundation.
“We need to think very strategically about how to make the right investment,” McMahon said.
I recommend reading
Building trust by not trusting AI
Financial services should now know that trust takes time to build trust with customers. The same can be said for trusting new systems, especially when it comes to AI.
“Don't assume that the shelf solution is trying to solve what you need to do,” McMahon insisted.
AI skills also play a role. Let staff use AI to not only complete it, but also create a world of difference.
Research from Git Clear It shows that AI-generated code from Copilots leads to increased duplicate code and erosion of code quality. Workers may be doing more, but the quality of this work may be degraded.
While experts in their field may be able to use AI to find errors and fix them, this also shows the requirements for having a strong basic understanding and ability
AI models have bias, hallucinating abilities, and misleading abilities. It is important to understand what you want from the AI model and whether what they generate is quality.
“It's just about chasing and balancing the hype and then critically assessing it,” he said.
“We go back to basics and what is the business case, what are the opportunities I'm chasing? What metrics do I say are successful? What are they for?”
Determine your use cases and create metrics to measure this return – not just what people feel, but what is actually happening from a productivity perspective, is essential to ensuring that AI investments are done responsibly.
Trust must be built even when trust in AI is responsibly built and hype means IT will win funds for AI transformation or integration.
“We're still going to face the same challenges. That's what we'll have when I implement something like this. You know business. You know business.
McMahon has found a board that sincerely encourages investment in AI, but the organization is still asking questions about return on investment.
“The worst thing you can do as an organization is to spend millions of dollars.
Implementing the right metrics and measures is essential for organizations to trust that AI implementation and investment is worth it.
However, this is easier than that. The AI skills gap is the highest concern for businesses as they are now trying to implement new technologies, and refund-of-investment data and metric measurements can be difficult to spread and explain to boards that don't fully understand technology.
McMahon focuses on integration initiatives across different departments, suggesting some kind of central AI department (e.g., AI Center of Excellence) that can explain and track the progress of AI projects and reflect these to leaders.
This helps financial services organizations maintain a balance of innovation while maintaining trust within the organization and with their customers.
“Legacy Financial Services companies have built a very righteous reputation and trust in being conservative,” admits McMahon. “But now we're in an age where we have to break away from some of them. I've never seen anything move so fast, so we have to think, 'What's the balance we're trying to attack?” ”
