Adam Lieberman, head of artificial intelligence and machine learning at Finastra.
getty
Since ChatGPT took the world by storm, we have seen impressive use cases of generative AI’s ability to unleash creativity, approach problem-solving from new angles and analyze data through different lenses.
In the face of its potential applications, organizations across different industries are considering how to deploy generative AI rapidly in both internal operations and client-facing solutions.
Beginning to implement generative AI comes with unique challenges and considerations—particularly for large organizations. In addition to addressing issues such as data quality, ethical and legal concerns, computational resources, interpretability and explainability and security (subscription required), organizations must also consider how to educate employees across functions while ensuring that utilizing this new technology remains a priority internally.
As a tech leader, exploring ways to scale generative AI is one of my top priorities. Here are some initial strategies and considerations for large organizations that want to deploy generative AI across facets of their operations:
Prioritize education for all teams—not just developers.
Generative AI is becoming a household term, but putting this technology to use at a large organization requires more than just surface-level familiarity. The field of AI extends far beyond the large language models (LLMs) we’re now becoming familiar with, actually dating back to the 1950s, with decades of research contributing to the likes of today’s ChatGPT.
It’s not only engineers and product teams who need to be educated. Even at this nascent stage, generative AI shows an aptitude for driving efficiency into a wide range of data, communications and administrative functions. For this reason, large organizations that want to make use of this new technology at scale should ensure all employees, not only developers, are given learning opportunities.
Rolling out generative AI training programs for everyone—across sales, communications, legal, HR, engineering and more—can empower teams to ideate new use cases that deliver market value, streamline workflows and free up bandwidth for strategic priorities. For example, research analysts spend a big chunk of their time reading content. Tools like ChatGPT could accelerate this process, giving them more time to generate original ideas. Internal customer support teams can leverage generative AI models to group similar incoming support tickets together, generating answers at a rapid pace.
In my experience, successful generative AI educational programming requires cross-company collaboration. My team, which focuses on AI and machine learning, is working closely with colleagues in other departments, including HR, security and legal, to develop customized learning plans with different levels of certification appropriate to an employee’s job function. The first level of our certification, intended for all employees, covers what generative AI is and how to identify potential use cases for it. The second level, designed with developers in mind, gets into more technical areas such as data collection, prompt engineering and model fine-tuning.
The entire world is still learning about generative AI—its potential and its risks—and immense business value can be derived from bringing your entire team along for this journey.
Create space for innovation.
Once employees have a nuanced understanding of generative AI, the focus naturally shifts to how they can apply the innovation to operations and product offerings.
To kickstart this process, I recommend creating spaces for teams to experiment with potential generative AI applications. Secure experimentation zones can empower employees to come up with world-class ideas. For example, equipping developers with the tools and resources required to prototype and deploy production-grade generative AI solutions can go a long way in building out a pipeline of new future offerings.
Internal hackathons structured around creating "GenAI" applications are a perfect illustration of such a space for creativity. This kind of initiative can be incredibly valuable for getting teams to think about exciting client solutions while fueling their desire to learn more about the technology.
Review internal processes and adapt for faster innovation where possible.
Developments in generative AI are coming rapidly, and companies that want to stay ahead of the curve will need to act fast. Particularly in financial services, institutions must follow highly regulated processes when building and launching new products that can be time-intensive, so moving quickly is paramount.
Finding ways to streamline product development processes where possible while providing adequate time for testing and due diligence is critical in competitive industries when the market is this hot. Organizations can start by conducting internal evaluations of current processes and having senior leaders meet regularly throughout this audit to discuss how company policies can evolve to support the safe development of innovative solutions.
Finally, in deploying new policies, organizations should not forget to stress the continued importance of human oversight. Employees still need to leverage their expertise and domain experience to validate the output of machine learning tools, and they should feel empowered to do so. We cannot rely on machine learning models alone. We must keep humans at the center by approaching these tools as a means of assistance and not replacement.
Generative AI holds great promise, and savvy organizations that want to use this innovation to their competitive advantage should start laying the groundwork now. While we may not know everything about where this technology is going, it’s clear that it will become part of our work lives and the broader world.
As we enter the era of “GenAI,” education, creativity and strong, streamlined governance will be key.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
