PHL AI guardrails must be in line with regional standards, focused on data

By Aubrey Rose A. Inosante, Reporter

SINGAPORE — The Philippines should draw up governance and ethics guidelines or guardrails for the use of artificial intelligence (AI) that are aligned with those in the Association of Southeast Asian Nations (ASEAN) and prioritize data privacy amid a booming digital economy, industry experts said.

“There is no point drafting your own [AI] guardrails or guidelines when not in concert with the ASEAN guidelines,” Raju Chellem, editor-in-chief of the AI Ethics and Governance Body of Knowledge, told ����ý at the sidelines of the annual IBM Think Singapore 2024 conference held on Aug. 14-15.

“It’s better to work together so that we are all together to work for the common good because many countries are looking at AI in isolation and not including data as part of it,” Mr. Chellem added.

The ASEAN Guide on AI Governance and Ethics was released during the 4^th ASEAN Digital Ministers’ Meeting chaired by Singapore in February. This is meant to serve as a guide for organizations in the region that wish to design, develop, and deploy traditional AI technologies.

In the Philippines, the Department of Trade and Industry this year released its National AI Strategy Roadmap 2.0, which consists of seven strategic imperatives: building a robust connected and networked environment; improving data access and data value extraction; transforming education and nurturing future AI talents; upskilling and reskilling the workforce; building an AI ecosystem “conscience”; mastering and pushing the boundaries of AI; and accelerating innovation with AI.

According to an IBM Institute for Business Value study, 60% of executives surveyed said they believe the adoption of generative AI (GenAI) comes with major ethical risks that would be difficult to manage without new or at least more mature governance structures.

Mr. Chellem said without AI guardrails, the finance and healthcare sectors, which are considered “supercritical industries,” will be at high risk as both need “high regulation for data privacy, data security, and the use of personally identifiable information.”

“Hackers can do phishing, probably destabilize governments, or ransomware as a service,” he said.

The digital sector has a “massive” contribution to the gross domestic product (GDP) of ASEAN countries, and “a huge chunk of that is because of the power of AI or the power of companies using GenAI,” Mr. Chellem said.

“The current GDP of the Philippines is about $472 billion, and the ICT (information and communications technology) sector contributed roughly about 9.4% to the GDP in 2022,” he said.

To educate people on AI, the Philippines needs to apply the same model it used to boost the nursing sector, which is integrating it in schools, Mr. Chellem added.

“The number one lesson is that the Singapore government does not work in isolation. It always includes the community, industry, and academia. All three work together to form policy and set guidelines, set governance frameworks [in AI],” he said.

Catherine Lian, general manager and technology leader for IBM ASEAN, told ����ý that governance is key in the adoption of GenAI.

“The GenAI adoption that we talk about is all about governance. We have a watsonx.governance model that enables appropriate design principles across governance and building up the enterprise-wide application of multiple AI models,” she said at the sidelines of the same event.

The model ensures that the adoption of an AI-first approach does not compromise the data provided and follows the code of ethics of an organization, Ms. Lian said.