Three barriers hindering Gen AI adoption in APAC businesses
By Deepak Ajmani, VP of APAC Digital Native and Emerging Markets Business at Confluent
If there’s one question on every leader’s agenda today, it’s this: How can our organization unlock AI’s transformative potential?
No longer a novelty, Generative AI (Gen AI) has become embedded in the fabric of modern business – and everyday life – across APAC. Its rapid ascent is being fueled by fierce market competition and aggressive digital transformation, with emerging economies leading the charge.
But while the pace of adoption varies across the region, organizations face a common set of challenges. Three critical barriers stand in the way of successful implementation: legacy systems, a widening skills gap, and regulatory uncertainty.
Let me explain.
Barrier #1: Legacy data systems
Nearly every established organization in APAC faces the challenge of integrating legacy systems with gen AI applications. Data extraction, transformation and loading become complex and costly, with the additional re-engineering of technical debt adding to the burden. Then there’s the issue of data governance – especially in heavily regulated industries like finance and healthcare – as a lack of suitable frameworks exposes data to mismanagement.
Legacy systems pose a particular issue as organizations begin to scale. It becomes clear that the mess of disparate systems cannot be effectively managed to meet new demands. In a world where we expect products and services to be delivered instantaneously, real-time data streaming is essential to harnessing generative AI effectively. Older infrastructures tend to operate in batch processing cycles and simply weren’t designed to handle these needs.
At some point, the imperative to switch becomes vital to competitive success. But the substantial investment required at the start (and for ongoing maintenance) can be prohibitive, especially where IT departments are being mandated to lower costs.
Barrier #2: The skills gap
There’s a wide pool of talent in Asia ready to adopt AI tech. Nevertheless, implementing gen AI can introduce significant changes to existing workflows and collaboration between teams. Leaders need to make sure their people have the knowledge across areas including machine learning, natural language processing, and real-time data streaming. Businesses either need to upskill their current workforce or bring in specialized talent that is often costly. And with AI models continually evolving, that training needs to be continuous, adding pressure to employees who may already be managing heavy workloads.
The anxiety around AI also needs to be addressed. How much is it perceived as a threat to jobs? Employees may fear that adopting these tools could reduce their value, leading to a resistance to using the tech and potentially affecting morale.
To overcome this challenge, we need robust training programs bolstered by ongoing support. This can help with shifting the mindset across the company towards a more data-driven, innovative culture where AI is a tool to enhance, not replace, roles. In this regard, strong leadership is key. Senior executives who advocate for AI adoption and identify “change champions” – employees who are enthusiastic about AI – can drive change and increase motivation.
The good news is that there’s a younger cohort of employees dubbed “generation AI” in APAC who are using the technology at extremely high rates. According to a study by Deloitte, 62% of young people (aged 18 to 24 years old) are using AI, with 43% doing so at work. The study shows how the accessibility of AI is a key driving factor behind its adoption, creating a wave of early career professionals set to bring the technology into the companies they’re working for. This signals a need to work with these early adopters to foster innovation at all levels and build frameworks for AI adoption across various departments.
Barrier #3: Regulatory uncertainty
Many organizations are hesitant to invest in AI due to the fear of regulation changing as the technology evolves. We know regulation is coming, we just don’t know what it will look like.
APAC is also diverse in its regulatory approaches. China has strict regulations related to AI that prioritize government oversight, while Singapore promotes a more flexible approach designed to cultivate innovation. The inconsistency in how guidelines are applied compounds the uncertainty around regulation. Without clear answers, businesses are wary.
Privacy concerns are a critical issue, especially in countries with strict data localization laws. The massive volumes of data required for AI training have brought data sovereignty to the forefront, with many businesses hesitating to shift data to the cloud. Regulations like Japan’s Act on the Protection of Personal Information (APPI) impose stringent restrictions on cross-border data transfers, requiring approvals that can significantly delay – or even halt – AI projects.
As APAC governments increasingly scrutinize where and how data is processed and stored, many organizations are opting for on-premises or private cloud solutions over fully public cloud models. However, these options are costly, and smaller companies often lack the resources to make such investments.
Hybrid, BYOC (bring-your-own-cloud) or multi-cloud solutions are proving an effective compromise. Sensitive data can be kept on-premises or within local servers while non-sensitive tasks can be carried out over public cloud services. While this setup can increase complexity, it also improves flexibility. Businesses don’t want to be locked into rigid, long-term contracts anymore, and are turning to open-source and agile SaaS solutions to achieve the flexibility they need. In fact, working with the right technology partners can help significantly streamline the transition to a compliant, AI-ready infrastructure.
Fostering AI adoption doesn’t have to be painful
Overcoming these challenges doesn’t have to be a painful process. A shift in mindset is key – the first step to embracing AI as the future. Leaders can then look to implement a phased approach, blending legacy systems with cloud-based or hybrid architectures, developing a clear data strategy, and investing in workforce upskilling.
With this foundation, businesses across APAC will be well-positioned to unlock the full value of AI as it advances, establishing the region as a global leader in AI innovation.
