You need not give up your AI vision, but be practical in implementation
Here are four strategies for effective generative artificial intelligence deployment, in order to realise its true potential
GENERATIVE artificial intelligence (GenAI) promises transformative capabilities across industries. Yet, SoftServe’s latest research with Forrester found that only 22 per cent of global organisations have successfully unlocked its value.
Amid the excitement, a sobering truth emerges: There is a disconnect between the hype for GenAI and the practical realities of deploying it.
The root cause is that many companies do not understand what it takes to deploy GenAI effectively. They rush into the implementation without a clear strategy or understanding of their readiness.
Here are four strategies for effective GenAI deployment.
1. Start with a clearly defined purpose
The first step to good problem-solving is a clearly defined problem statement. Many companies, in their rush to not miss out on the GenAI hype, hurry through developing use cases instead of assessing their actual needs.
No wonder 51 per cent of Singapore companies encounter challenges optimising GenAI production at the prototype deployment stage, and 41 per cent during the version 1 roll-out. A solution’s implementation and scalability will encounter resistance if it does not address the right problem.
BT in your inbox
Start and end each day with the latest news stories and analyses delivered straight to your inbox.
Successful GenAI deployment starts with researching, assessing and identifying a clear, company-specific use for GenAI.
Leaders can conduct focus groups across all functions and levels of employees to hear first-hand the day-to-day pain points and map possible GenAI use cases as solutions.
Anyone would be able to tell you the general benefits GenAI could bring: increased productivity, optimised operational efficiencies, data analysis with real-time insights.
What does that mean for your organisation? What specific problems can GenAI solve for your company, and how does solving those problems align with your broader business goals?
2. Understand your company’s readiness
Conducting assessments to understand your company’s pain points also helps you gauge how prepared your business is to integrate GenAI.
It is key that leaders keep a realistic view, to see both the potential and pitfalls of deploying and scaling GenAI. It may promise transformation, but that will not happen if leaders are not sensitive to making use cases stick.
No two companies are the same, but here are some starting points based on the top three factors Singapore companies identified as areas of improvement in their GenAI strategy:
- Large language model (LLM) accuracy – Gen AI models are only as good as the data they are trained on. Companies should reassess their data quality and IT architecture to ensure they can support the extensive computing needs of GenAI.
- Capability awareness/use – This is fundamentally a human factor. Leaders need to train their workers to work alongside and with GenAI applications, while simultaneously assessing if their organisational culture is ready for AI-driven changes.
- Strategy roadmap – GenAI is an exciting new frontier, but this also means companies have little to no precedent for projecting the road ahead. From operations to regulatory compliance, leaders must source the knowledge and materials necessary to have a clear, holistic picture of what is up ahead for their business, and what it would take to get there.
3. Form partnerships with the right experts
Two kinds of help companies have identified as useful for overcoming production scale challenges are having a consulting partner (43 per cent) and having a software solutions partner (41 per cent).
Implementing GenAI is complex and often requires specialised expertise. Agile external partners would have the technical expertise necessary to guide businesses on:
- Industry-specific best practices for GenAI deployment.
- Methods of data preparation and LLM model training.
- How-tos for integrating gen AI into existing workflows, across functions and departments.
- Tools for monitoring and improving GenAI performance over time.
By outsourcing expertise to partners, companies can supercharge their evolution – leading to faster time-to-value, reduced risk and more effective use of resources.
4. Know your industry to offer such solutions to customers
Everyone knows that GenAI offers exciting possibilities; but for it to be truly effective, its applications must be industry-specific. Business leaders should ask themselves what gaps exist in their market, and how GenAI can fill them.
Consider the banking and financial services industry, for example. GenAI offers exciting opportunities for personalised customer experiences, with customer service avatars and pathfinders powered by it.
The effectiveness of such solutions hinges on the AI’s ability to pull data from all the financial services a customer uses – from savings to loans to investments – which is challenging due to siloed legacy data.
In fact, 51 per cent of all Singapore companies – not just those in banking and financial services – face challenges by the prototyping stage.
GenAI’s true potential will only be unlocked by those who approach its implementation with ambition and pragmatism, addressing internal gaps while connecting with the wider industry.
Andrew Tan is the Asia-Pacific enterprise solutions lead for banking, financial services and insurance (BFSI) at software company SoftServe. Dipen Mehta is head of BFSI Apac at SoftServe.
Copyright SPH Media. All rights reserved.
