By Beatriz Marie D. Cruz, Reporter
PHILIPPINE COMPANIES sit closer to the lower end of the artificial intelligence (AI) adoption curve due to limited investment, cultural and language barriers, and weak scaling strategies, according to Hitachi Vantara, the US-based digital infrastructure unit of Japan’s Hitachi, Ltd.
“From a regional perspective, APAC (Asia-Pacific) overall is moving quickly, with significant investments flowing from countries such as Singapore, South Korea, Japan, and China — in some cases matching the pace of the US,” Jason Hardy, chief technology officer for AI at Hitachi Vantara, said in an e-mail interview with BusinessWorld.
“In contrast, Australia and New Zealand remain slow adopters. The Philippines sits closer to the lower end of the adoption curve — moving faster than Australia and New Zealand, but still behind regional leaders like South Korea, China, and Singapore,” he added.
He said one reason is the lack of comparable commitment to build AI infrastructure locally.
He also cited a 2024 study by the Philippine Institute for Development Studies showing that while 90.8% of enterprises have computers and 81% have internet access, only 14.9% use AI.
At the same time, the country also faces cultural and language barriers in adopting AI since many systems are not optimized for the local context, he noted.
“Unlike markets such as India or China, where governments and enterprises are creating foundational models tuned to local languages and contexts, the Philippines has not yet made a concerted push in this area,” he said.
“Without that foundation, adoption is slower, as systems are not yet optimized for the country’s unique cultural and linguistic landscape.”
Mr. Hardy noted that early adoption of agentic AI — which acts with a degree of autonomy based on real-world inputs — has become a competitive differentiator for companies.
“Companies that invest early will be able to process information faster, make better-informed decisions, and execute more quickly than their peers,” he said.
He added that many firms confuse generative AI with agentic AI which produces content such as text or images based on context.
“This could mean anything from creating marketing material in response to customer behavior to monitoring the power grid and making recommendations based on live system data,” he said.
A common misstep among Philippine companies, he said, is focusing on pilots and proofs-of-concept without scaling the necessary architecture to support autonomous operations.
“This results in stalled projects and limited return on investment (RoI),” Mr. Hardy said, adding that many firms also overlook the importance of governance and auditability processes that could strengthen trust and accelerate adoption.
He said organizations should prepare their environment for agentic workloads, with architectures that support multi-agent coordination and dynamic resource allocation, and prioritize generating high-quality data to avoid the “garbage in, garbage out” problem.
“In reality, these are capabilities and tools designed to accelerate outcomes, not plug-and-play solutions. Expecting instant RoI is flawed; it takes practice, expertise, and a willingness to accept failures along the way before consistent successes emerge,” he said.
A report by Boston Consulting Group estimates that AI and generative AI could contribute around $120 billion to the combined gross domestic product of six ASEAN economies — Indonesia, Malaysia, the Philippines, Singapore, Thailand, and Vietnam — by 2027.