The ongoing wave of AI adoption has ignited high hopes for breakthrough innovations in enterprise operations. However, the actual implementation of AI still faces numerous barriers. This highlights the need for businesses to recognize the foundational role of core technologies in standardizing processes—laying the groundwork for AI to be truly effective.
Table of Contents
The AI Hype and Unrealistic Expectations
Artificial Intelligence is being hailed as a “universal key,” fueling a tech frenzy with promises to revolutionize business operations. In reality, investment in AI is surging: total global spending on AI in 2023 is estimated at $20–25 billion, with projections reaching nearly $300 billion by 2026.
Yet, AI is far from a magical solution. Experts emphasize that successful AI deployment requires massive volumes of data, robust computing infrastructure, and highly skilled personnel—all of which come at a significant cost. Moreover, concerns over data security, algorithmic transparency, and bias remain top of mind for many organizations.
Gartner has warned that most AI projects fall short of expectations if a business lacks well-standardized and automated processes. A recent analysis revealed that up to 85% of AI initiatives fail to deliver expected value—primarily due to dirty data and weak foundational workflows. As James McKay, Founder and CEO of VEN, cautions: “If you feed garbage data and outdated processes into AI, it will only accelerate your failure.” In short, without an optimized foundation, AI tends to “amplify the chaos” rather than deliver miracles.
Even more concerning is that the AI hype has caused many businesses to delay or freeze investments in foundational digital technologies. Caught in a “wait and see” mindset, some organizations have postponed or canceled existing IT projects to cut costs.
According to Larry Walsh, CEO of research firm Channelnomics, overblown expectations surrounding AI are making companies hesitant to invest in essential areas like cloud computing, cybersecurity, and enterprise software—causing a decline in vendor revenues. He warns that betting everything on a future “AI miracle” could stall the tech industry’s momentum in the short term, especially as critical infrastructure upgrades are being put on hold.
What Businesses Must Focus on When Applying AI
Digital transformation is not just about adopting new technologies—it’s about reshaping how organizations work, think, and operate. To implement AI successfully, companies need a strong foundation across multiple dimensions.
First, AI initiatives must be tightly aligned with the organization’s business strategy. Leaders must clearly define the specific goals AI is meant to support—be it revenue growth, enhanced customer experience, or cost optimization—and build a structured roadmap to get there. According to Vietnam’s Ministry of Information and Communications, 59.6% of Vietnamese businesses face leadership challenges in digital transformation. Human resources are another concern, with 64.7% lacking specialized talent, and 71.3% without effective change management plans.
Therefore, strong leadership commitment, a clear digital strategy, and robust training programs for employees are key to ensuring successful, synchronized AI implementation.
Moreover, streamlined and stable business processes are essential for AI to deliver meaningful results. Before deploying AI, organizations should review and standardize existing workflows, then automate repetitive steps. In fact, implementing Robotic Process Automation (RPA) is widely regarded as one of the most effective foundational technologies to pave the way for successful AI adoption.
RPA forces organizations to eliminate redundant tasks, process bottlenecks, and manual errors—resulting in consistent data inputs and smoother workflows, which are critical for downstream AI applications.
“Many businesses mistakenly expect AI to deliver instant automation. In reality, the proper roadmap involves standardizing and optimizing business processes using tools like RPA first. Only then can AI achieve its full potential on top of a stable automation framework,” said Mr. Ngo Quy Kien, Head of Solution Consulting at Akabot, FPT Corporation.
The Trend of Enhancing RPA with AI Capabilities
Among leading banks in Vietnam’s digital transformation journey, a proven strategy has been to first optimize operations with foundational technologies like RPA and then integrate AI to achieve full intelligent automation.
For example, in 2020, TPBank partnered with FPT to deploy 45–50 bots on the Akabot platform across five business units. This laid the foundation for process standardization and created consistent data flow across systems, before scaling to hundreds of processes across nine core functions.
By early 2021, TPBank had 75 “virtual assistants” automating large-scale repetitive tasks. By 2022, the number of bots had grown to over 300, upgraded to handle more complex operations, integrate with multiple software systems, and support centralized governance—paving the way toward a “digital workforce” operating model.
With standardized processes powered by RPA, TPBank further integrated AI components (chatbots, natural language processing, OCR/computer vision, big data analytics) into its service and operations channels—such as ChatPay, Paste to Pay, and LiveBank 24/7. These tools underpin its vision of becoming an “intelligent bank,” delivering personalized experiences and seamless, end-to-end processing with minimal errors.
TPBank’s journey offers a valuable lesson: instead of standing still waiting for a perfect AI solution, visionary leaders can invest in process standardization and automation technologies now to gain immediate operational improvements.
While AI holds vast potential for the future, RPA provides a strong digital foundation today. When combined effectively, the two enable businesses to harvest the real fruits of automation—both now and in the long run.
By Ánh Dương
Thanh Niên Việt
Source: CafeF
