{"id":228779,"date":"2024-11-20T16:24:15","date_gmt":"2024-11-20T08:24:15","guid":{"rendered":"https:\/\/www.grab.com\/sg\/?post_type=editorial&p=228779"},"modified":"2025-12-05T11:43:27","modified_gmt":"2025-12-05T03:43:27","slug":"harnessing-ai-for-public-good-grabs-approach-to-ai-governance","status":"publish","type":"editorial","link":"https:\/\/www.grab.com\/sg\/inside-grab\/stories\/harnessing-ai-for-public-good-grabs-approach-to-ai-governance\/","title":{"rendered":"Harnessing AI for public good: Grab\u2019s approach to AI Governance"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t

Artificial Intelligence (AI) has been integral to Grab\u2019s operations for over a decade, with more than 1,000 AI and machine learning models built and deployed to support our core businesses. From optimising trip allocations to making rides safer, our AI systems enable real-time automated decision-making that enhances customer experiences.<\/p>

Now, with the rise of Generative AI, we\u2019re adding powerful new capabilities to our toolbox. Its ease of use and accessibility allows everyone to experiment, innovate, and develop solutions faster and at a larger scale.<\/p>

At Grab, we aim to be a leader of AI innovation in Southeast Asia. We embrace this technology with a peace of mind because we have a robust AI governance framework in place, ensuring that we pursue opportunities fervently while managing risks prudently.<\/p>

Developing AI ethics principles<\/h5>

As an early adopter of AI in Southeast Asia, we quickly recognised that leveraging AI\u2019s potential responsibly would require a proactive approach.<\/p>

This foresight led us to define our stance on AI ethics, aligning our use of the technology with Grab’s core values. This ethical framework ensures that our AI initiatives are not only innovative but also grounded in responsibility and transparency.<\/p>

To establish clear principles for AI ethics, Grab formed a dedicated working group to explore how governments, industry leaders, and international organisations approach the ethical use of AI.<\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t

The group reviewed evolving frameworks for AI ethics and governance from well-regarded bodies, such as the OECD, UNESCO, and WEF, as well as from countries including Singapore, the EU, Australia, Japan, and China. This assessment provided insight into societal concerns about AI and the importance of governance to address these issues.<\/p>

Based on this research, the working group proposed a set of AI ethics principles that were then aligned across Grab, ensuring that our approach to AI remains both socially responsible and in line with global best practices.<\/p>

AI risk framework based on existing foundations<\/strong><\/p>

Building on our ethics principles, we developed an AI risk framework that offers a systematic, practical approach for identifying, assessing, and mitigating AI-related risks at Grab.<\/p>

As we deepened our understanding of these risks, it became clear that they weren\u2019t entirely new; rather, AI amplifies their scope and scale. This underscores the importance of carefully managing AI\u2019s influence within our operations.<\/p>

We found that each type of AI-related risk aligns closely with an existing risk function at Grab, allowing our specialised teams to contribute their expertise to AI risk management.\u00a0<\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t

To leverage this alignment, we formed an interdisciplinary AI governance task force, drawing on the in-house knowledge and experience of well-established functions to create a comprehensive and adaptable approach to AI oversight.<\/p>

The task force\u2019s mission is to guide Grab toward safe, responsible, and effective AI integration. By uniting these teams, we can act swiftly and cohesively, ensuring that our AI governance is both strong and responsive to the organisation\u2019s evolving needs.<\/p>

Putting AI risk management into practice<\/h5>

Grab has now built essential structures to promote safe and ethical AI use across the organisation.<\/p>

We have already discussed the AI Ethics Principles, as well as our AI Risk Framework and our interdisciplinary AI Task Force, which brings together expertise from integrity, data governance, cybersecurity, legal, data privacy, and public relations teams to guide AI adoption aligned with Grab’s standards and values.<\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t

Based on these foundations, we implemented Pre-production Safeguards \u2013 controls in place to assess AI risks prior to the launch of any new features with AI integration. Any applications with potential risks are flagged\u00a0 by the AI Task Force and vetted for controls to ensure a successful launch and minimise issues related to AI.<\/p>

To encourage a culture of awareness around responsible AI governance practices, we conduct workshops and gamified quizzes, incentivizing engagement from a wider group of AI practitioners. This collaborative approach to learning about AI risks and responsible AI usage makes it more interactive and enjoyable, and simultaneously offers insights into the organisation\u2019s collective competence, empowering our teams to confidently navigate the evolving AI landscape.<\/p>

The next step: scaling AI governance with automation<\/h5>

As Grab\u2019s AI capabilities expand, our governance structures will scale accordingly, incorporating greater automation and metric-driven oversight.\u00a0<\/p>

Our vision is to embed governance rules directly in the development lifecycle, enabling scalable, real-time AI governance that adjusts dynamically to evolving risk levels. This setup will enhance our time to market and agility by delivering instant insights into AI performance and compliance, which feed directly into our governance process for timely adjustments \u2014 allowing for active monitoring, addressing areas for improvement, as we continually uplift the standards of our AI products.<\/p>

As a pioneering technology company in Southeast Asia, we aim to lead as champions of AI by fostering a mindset that embraces its potential, without fearing its limitations. By achieving a balanced benefit-risk perspective, supporting our business to confidently adopt generative AI through calculated, informed decisions. Our goal is not only to anticipate and mitigate risks but to enable rapid recovery and adaptation when challenges arise, ensuring resilience while advancing innovation.<\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"parent":180237,"menu_order":0,"template":"grab21-default","acf":[],"_links":{"self":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/228779"}],"collection":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial"}],"about":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/types\/editorial"}],"version-history":[{"count":43,"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/228779\/revisions"}],"predecessor-version":[{"id":255176,"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/228779\/revisions\/255176"}],"up":[{"embeddable":true,"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/editorial\/180237"}],"wp:attachment":[{"href":"https:\/\/www.grab.com\/sg\/wp-json\/wp\/v2\/media?parent=228779"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}