Indosat’s Heavy Investment in AI Could Be the Key to Accelerating AI Adoption in Indonesia

By Benjamin Chan | 3Q 2024 | IN-7522

Registered users can unlock up to five pieces of premium content each month.

Log in or register to unlock this Insight.

One of Indonesia's telecoms giants, Indosat Ooredoo Hutchison (IOH), has been actively promoting the adoption of Artificial Intelligence (AI) in Indonesia. In August 2024, Indosat partnered with Huawei to launch Indonesia's first AI Experience Center, aiming to demonstrate the benefits of AI in various industries and provide training for enterprises. In addition, Indosat also collaborated with Google Cloud to explore a joint go-to-market strategy, which aims to enable Small and Medium Businesses (SMBs) to leverage new technologies for growth. By forming strong partnerships with major tech and AI players, Indosat is strengthening its AI presence in Indonesia. It is set to play a significant role in influencing and integrating AI adoption in enterprises, signifying potential growth in the Indonesian region.

Unlocking critical enterprise use cases through partnerships with infrastructure and technology specialists in AI and the cloud is a viable strategy for accelerating AI adoption and growth in the country. Indosat's steps toward AI integration and adoption in Indonesia perfectly align with the broader Southeast Asian (SEA) region's push to become an AI powerhouse, encouraging bigger global tech giants to invest in partnerships with major local players to offer AI services to the region's enterprises.

ABI Research observes strong growth in AI adoption in Indonesia, with major telecoms players like Indosat and other startup adopters attempting to establish their AI niche-companies like Yellow.ai and WIZ.AI have built multilingual and bilingual Large Language Models (LLMs) focusing on supporting local Indonesian languages in translation, content generation, and information retrieval. With continued strong support from major technology and infrastructure players internationally, like Google Cloud's partnership with domestic players, Microsoft's planned data center construction, and local technology conglomerates like Indosat, the barriers to entry for AI adoption will drop significantly, allowing an accelerated increase in uptake and implementation. As major local technology players continue to explore avenues of partnerships in Indonesia and its surrounding SEA region, local enterprises can continue to reap the benefits of ease of access to various enabling AI technologies, such as Natural Language Processing (NLP), Machine Learning (ML), and algorithmic-driven insights.

Enterprises around the region should prepare for an impending AI boom in SEA. The accelerating influx of AI is not only derived from international tech giants like OpenAI, Google, and Meta's interest in the region, but also from locally-driven AI innovators like Sea Labs and other startups in leading AI countries like Singapore, Indonesia, Thailand, and Vietnam. A comprehensive understanding of AI applications and how they can impact an enterprise's business model would be critical in understanding the extent to which AI and ML can transform and accelerate businesses. Enterprises expecting to explore, adopt, and implement AI workflows into their business processes should understand that the key to unlocking AI-enabled ROI is strongly tied to the utilization of data and data processes. These data considerations are as follows:

• Clean Data: The potential for Generative Artificial Intelligence (Gen AI) and LLMs lies in data hygiene. Data-driven insights and recommendations on opportunities gleaned from an enterprise's data must be built on a single source of truth and point of accuracy, rather than on a mess of duplicated, incorrect, formatted incorrectly, or even missing data entries, which could distort data observation and analysis.
• Relevant Data: Like clean data, relevance and reliability of data should be a key consideration in leveraging AI and ML-enabled processes. Data recency and relevance will significantly improve the LLM's reliability and its ability to maximize ROI from generated insights. Outdated data may not reflect the current trends, while missing data and incorporating outliers could clutter the models and result in inefficient and misleading outputs.
• Proprietary Data: Due to confidential and proprietary data, enterprises must consider appropriate AI integration or deployment models. When managed carefully, enterprise proprietary data could be the key to gaining a competitive advantage against other organizations. However, proper AI governance policies should be in place that carefully manage the sensitive nature of consumer data and the extent to which the data are exposed to the wider Internet. Otherwise, enterprises could quickly lose their advantage as their proprietary data are unintentionally fed into public ML models.

Additionally, enterprises should understand that the extent to which AI and ML-enabled workflows can unlock ROI is not always sky-high. Successful ROI will depend on the alignment of use cases to the applicability of the business, along with integration within existing systems and workflows that help expedite adoption and value realization. Enterprises that adequately manage expectations of realistic returns and continually refine their data governance policies would likely see better ROI from AI implementation, as well as higher-quality insights gleaned from AI models.