IDC is forecasting Asia/Pacific (excluding Japan) spending on AI systems to rise from US$17.6 billion in 2022 to around US$32 billion in 2025. Businesses invest in artificial intelligence (AI) to gain a competitive advantage through improved customer insight, increased employee efficiency, and accelerated decision-making.
Dr Chris Marshall, associate vice president, responsible for data, analytics, and AI at IDC Asia/Pacific, says AI adoption and maturity vary. At a country level, Singapore, Australia, and Japan are leading in the use of AI, quickly followed by Malaysia, Thailand, Vietnam, Indonesia, and certainly places like South Korea and Hong Kong.
Success breeds success
IDC predicts that over the next five years, the banking industry will continue to invest the most in AI solutions. Risk mitigation would be vital for the banking industry's AI investment through augmented threat intelligence and fraud analysis applications.
State/Local government is the second-highest spender on AI solutions, focusing on public safety and emergency response, augmented threat intelligence, and prevention systems.
For the next five years, the next top spending industry is professional services, growing fast at 26.6% (CAGR). The key focus area is augmented customer service agents, which help resolve customer issues. Smart business innovation and automation will optimize and streamline complex and repetitive business tasks to support organizational decision-making.
“The top 40% of companies that spend more on AI and make additional investments as they see success and gain value out of it. The other 60% are not doing so much in this area so the gap across industries is widening,” he continued.
Hurdles to AI adoption
In a July 2022 roundtable in Singapore, FutureCIO noted the momentum for AI adoption is straddled by concerns over changing business landscape. Similar discussions outside of a formal setting reveal other issues hampering AI adoption including data quality, data governance, as well as ethical considerations.
For Marshall, one of the hurdles is the limited use cases in the market.
“The biggest problem is identifying the use cases that are going to be useful businesses. We have smaller companies imitating the big firms as they do not have the luxury to experiment, but they may not be relevant.”
Dr Christopher Lee Marshall
“In fact, they look for safe use cases as they do not intend to establish new business models which is rather a shame,” he opined.
Asked whether the pandemic has influenced, in any way, the adoption and direction in which AI is being integrated into the enterprise, Marshall cited remote or hybrid working as one obvious outcome of the pandemic that is impacting how businesses operate.
“There’s also a whole set of customer experience and transaction-based systems, and AI sits on top of those transactional systems. The link between AI and digital transformation, more than anything else, is the fact that, once you start digitally transforming your transactions, and your business processes, a lot of data is thrown out of that. And, once we have data, AI becomes viable,” he elaborated.
Positive influence on AI adoption
Marshall acknowledged that success in integrating artificial intelligence can embolden organisations to continue their AI journey. He cited the three stages of AI development starting with coming up with the idea and getting seniors on board.
“Second is finding resources to focus on the AI projects, and lastly, bringing people closer who will ultimately apply the models. The organisations need to understand the importance of AI and how the projects will add value before embarking on that journey,” he opined.
Use cases that build momentum
Marshall noted that Singapore parks are one of the biggest users of digitalisation technology. The data, technology, IoT, screening work processes, specifically AI, all come together to be much bigger than the sum of their parts in managing these parks and facilities. It is part of the bigger system, which involves people, processes, assets, technologies, and all sorts of different things and there are many exciting use cases.
Friction in the AI journey is good
Renée Richardson Gosline, a senior lecturer and principal research scientist at the MIT Sloan School of Management, suggests that hurdles like human bias present friction that is important in the AI journey. Writing for Harvard Business Review, she suggested “executives who seek to improve customer experiences must embrace “good friction” to interrupt automaticity in the application of “black box” AI systems.”
She believed that AI holds tremendous promise.
“but if we are to be truly customer-centric, its application requires guardrails, including systemic elimination of bad friction and the addition of good friction. Friction isn’t always a negative — the trick is differentiating good friction from bad and auditing systems to determine which is most beneficial.”
Dr Renée Richardson Gosline
Tips for optimising AI adoption
When rationalising the adoption (or not) of AI (or any other technology), Marshall is of the belief that leadership must realise what is important about this technology.
“Companies that realise that AI is not an end onto itself are much more likely to be successful because they will align the use of the technology with particular, real business problems. Identify the business challenges and choose the right problems to solve with AI to be successful. If the problems are not important, no amount of success will make a difference,” he continued.
So, how should enterprises chart their AI roadmap?
For Marshall, the starting point could be a Centre of Excellence (CoE). Having a CoE opens the opportunity to scale the adoption of AI.
Also, “get the key players on board and make sure that they are supporting. Understand that AI depends on data. Without that data platform, the infrastructure, data quality, data management, it is very difficult to have a successful AI program,” he concluded.
Click on the PodChat player and listen to Marshall share his observation and recommendations on how to get the best from AI.
- Give us a snapshot of the state of AI adoption in Asia today.
- What are the three most significant hurdles impacting AI development in Asia?
- In what areas has the pandemic influenced both the adoption and direction in which AI is being integrated into the enterprise?
- What factors will most influence the maturation/adoption of AI?
- In the near, what AI use cases will likely see commercial adoption (at scale)?
- For enterprises to optimise their adoption of AI, what must they undertake first?
- There are many use cases of AI/machine learning today, how should enterprises chart their AI roadmap?