Generative AI offers the potential to transform the delivery of financial services and the way organizations operate, but it must be used responsibly and with understanding. Ensuring all people within an organization are able to explore its potential in
a safe and compliant way is a fundamental consideration and one that should shape every organization’s usage policy.
The UK’s Financial Services Skills Commission recently shared a
report highlighting how skills shortages present a major barrier to AI-driven growth. The Commission calls upon financial services firms to take action on the skills challenge, in order for the sector to reap the growth potential offered by AI.
Given that the pace of change in financial services and fintech is relentless, organizations that understand their resources, proactively address skills gaps, and organize their teams effectively can gain a competitive advantage.
Keeping that in mind, if you’ve not already started, here are four steps to consider in building enterprise fluency in Gen AI.
Involve everyone
The successful deployment of Gen AI demands a robust and comprehensive upskilling program across the enterprise. This requires buy-in across all levels of the organization from senior management through to entry-level staff. A good place to start is by putting
in place a SteerCo of senior stakeholders across all areas of the business. This should include representatives from legal, security, operations, AI/data, enterprise tools, skills development, comms and more.
Key issues to explore include legal and ethical considerations, responsible usage, benefits and applications, and the creation of introductory materials to share with all. Involving legal teams upfront is essential in developing rigorous legal and compliance
frameworks and policies, while the role of internal comms and HR teams is also vital in communicating with everybody effectively.
Fluency in generative AI is a requisite life skill that goes beyond what’s required in anybody’s current job role. But as with any significant event, humans can be resistant to, or fear, change. That’s why it’s imperative to provide reassurance that Gen
AI is not about replacing people but can be used to handle some of the manual and repetitive administrative tasks that currently restrict them from exploring their full potential. Investing in a robust training program, secure tools and a compliant operating
environment, can help attract and retain the best people.
Train and upskill
Set out a global, phased skills program for all employees. This should cover the basics across the organization to start building enterprise fluency in the tested, enterprise-grade technology and AI-chat tools being deployed. In parallel to this, firms with
software development and engineering teams, should have a separate focus on developer specifics, diving into usage by job family, with clear use case testing and rollout.
As well as focusing on skills development, also consider practical ways to explore, interact with experts and share knowledge. Consider broadening the learning by creating a platform for both internal and external experts to share their insights. This includes
leveraging partners to come in and contribute to learning. As well as conducting training and running workshops online, also consider the value of running events in person to help dial up the interactive, community element, which is fundamental to enterprise-wide
adoption and enjoyment.
Supplement events and tutorials with a range of other tools and materials including: cheat sheets on industry terminology, advice on prompt engineering, use cases for getting started, technical and business-focused whitepapers, and more. Gamified competitions
can also be used to encourage uptake of AI chat and other tools. It’s important to maintain an ongoing cadence of communications, from organizational-wide Town Halls to weekly newsletter features and to keep people updated on successful use cases and shared
learnings they can seek to emulate.
For developers and engineers a phased learning approach is extremely important. Even though this audience has more technical expertise than the average enterprise employee, they need to be able to understand in detail the technology, architecture, and infrastructure
that they are working with. Help accelerate their learning by sharing templates and tutorials, designed around common use cases that enable them to experiment and learn quickly from the experience of their colleagues. The goal is to make everything as seamless
as possible, enabling them to create their own proofs of concept and production models.
Create a safe place to work
All employees need access to the right tools – including enterprise AI-chat with commercial data protection so that chat data is not saved without consent, nor used to further retrain any underlying models. It’s also vital to implement a comprehensive set
of policies for usage, including legal and compliance, so everyone is clear on their responsibilities and the best practise to follow.
Having a variety of internal channels and communities, where employees can ask questions and share learnings, helps everyone feel like they’re in this together and it’s safe to seek help.
For developers it’s important to create a secure development environment, with the key functionality, tools, and resources they need to prototype, experiment, develop and deploy production-grade solutions. Tools like Github Copilot are also worth considering,
perhaps testing with small groups of engineers first to demonstrate the value before a wider roll-out.
Get started!
You’ll soon be able to tell if your upskilling policies are working by the level of attendance in training sessions, the duration people stay for, and the number of participants going on to onboard and start using the tools at their disposal.
Use cases for Gen AI across the enterprise can range from reducing administrative work for HR, finance, sales and marketing professionals through to empowering developers and engineers with the tools they need to accelerate the production of software solutions.
A sales and marketing example could be using Gen AI to create an internal text-to-speech model that generates high quality and voice-cloned audio files for video scripts. This can save time spent preparing product materials and demos for clients.
To help developers streamline common Gen AI initiatives organizations can create production-grade templates for internal and client-facing use cases. Developing a standardized roadmap for product development and for integrating Gen AI into customer-facing
solutions is essential. This should cover everything from ideation, prototyping, and validation through to production-grade deployments. Creating a standard roadmap template that details each of the four stages provides vital assistance to teams working on
solutions for customers – ultimately creating a production powerhouse that accelerates production and commercialization.
Building enterprise fluency in Gen AI can help improve productivity, enhance creativity and augment human capabilities at scale. While it’s not a silver bullet, used in the right context and in a safe and responsible manner it can transform ways of working
across financial services.
