Businesses that are still relying on legacy architecture may soon be left trailing smart rivals that can quickly respond to market opportunities.
Digital-first organizations are building applications in the cloud that leverage AI to fast track what once might have been highly complex undertakings. Doing so will enable them to gain cost savings, increased efficiency, and other competitive advantages.
Data strategy is key to this AI vision, yet there’s work to do. Although large enterprises are increasingly integrating data across silos and/or setting up large data lakes supported by powerful platforms like Microsoft Azure, they often lack a data foundation. For example, recent Foundry research reveals that only 34% believe they have the right data strategy in place to deliver effective AI solutions.1
A robust data strategy is essential to helping find the right data products to integrate or adapt AI in general and Generative AI in particular.
Use cases for Generative AI
Technology firm EPAM is supporting businesses that use huge volumes of data to improve efficiency using Generative AI tools.
For example, organizations in the oil and gas sector are using EPAM’s OSDU Data Platform (Open Source, standards-based, technology-agnostic data platform) to enable their analysts to rapidly query data.2 The solution also has a democratizing effect, freeing business users from having to work with analysts to surface insights, thereby increasing productivity.
In addition, organizations in this sector are digitising highly complex data, such as work and legal contracts, with Generative AI to ensure teams can access information quickly and effectively.
Similarly, EPAM is exploring how Generative AI can improve the underwriting process in the insurance sector. Underwriting is a data-intensive task, which typically requires extensive and costly support from data analysts. Intelligent applications enable underwriters to work more independently and much faster.
The importance of a cloud-first approach
To realize these and other benefits, organizations will need to build their intelligent applications and AI copilots on cloud platforms like Microsoft Azure and ensure the right data foundation is in place.
“Startups are by and large cloud-native, so they will have no problem launching intelligent applications,” says Iqbal Rahmoon, Head of Azure Solutions UK and Ireland at EPAM System. “Larger enterprises that use legacy IT or hybrid capabilities will find it a challenge to build and scale intelligent applications, as on-premises infrastructure isn’t designed to support the high computational requirement of Generative AI. A cloud-first stack is therefore a necessary precursor to success in Generative AI.”
Bringing centralized data platforms into the cloud also requires good communications infrastructure, meaning that CIOs must ensure their enterprise networks are robust and fast.
“Cloud platforms like Microsoft Azure provide the resources businesses need to deploy data and compute-heavy applications like AI,” says Mahesh Kasaragod, Director Technology Solutions at EPAM Systems. “AI is revolutionizing how the cloud providers deliver their services, and we can expect to see huge changes in areas including programming and IT operations in the years ahead, with the aim of making IT entirely zero downtime and zero touch.”
Organizations that embrace cloud-first models earliest will be best placed to achieve these results.
Aligning Generative AI capabilities with business goals
By leveraging AI capabilities in the cloud, organizations can build and scale intelligent applications with ease. “With AI, as with any app, the most important thing is understanding the business problem you are trying to solve and whether AI is applicable,” says Kasaragod. “From there, the framing is key. AI is not about replacing people so much as using automation and insights to make them more productive and optimally drive the required outcome.”
He suggests that technology teams prioritize business problems and then use a mix of human interventions and automation to drive positive outcomes. By aligning AI to business goals, AI models will increasingly become specialized.
Indeed, according to Gartner more than 50% of enterprise Generative AI models will be specific to either an industry or business function by 2027.3 As a result, CIOs should plan to manage a wide range of domain-specific models.
Learn more from EPAM on what the AI future has in store for enterprises.
1 Foundry, AI Priorities Study 2023, October 2023
2 CEGAL, The Open Subsurface Data Universe (OSDU®)
3 Gartner, “3 Bold and Actionable Predictions for the Future of Generative AI”, April 2024