Indonesian Political, Business & Finance News

AI Makes It Easier to Build Start-ups, but Business Models Remain a Challenge

| | Source: kompas.id | Technology
AI Makes It Easier to Build Start-ups, but Business Models Remain a Challenge
Image: kompas.id

So-called AI startups are often just wrappers for AI products from global platform companies.

22 Okt 2026 11:32 WIB · Ekonomi & Bisnis

AFP

DENPASAR, KOMPAS—The ease of developing applications using only artificial intelligence (AI) technology is opening up opportunities for more start-ups. However, start-ups that rely solely on AI technology without strong product advantages and business models risk struggling to survive the competition.

Kejora Ventures Partner Andreas Surya stated that the AI industry consists of several layers (subsectors). The lowest layer is infrastructure, followed by original models, platforms, and aggregators. The top layer is business applications.

The layer of AI infrastructure is evident from the construction of data center facilities, electronic circuits specifically designed to accelerate graphics and video processing (GPU), and central processing unit (CPU) components.

In Indonesia, infrastructure in the form of data center facilities is currently being widely developed. These are sourced from both global and national investors. However, for infrastructure in the form of GPUs and CPUs, the majority relies on the United States (US).

As one moves up the layers, an increasing number of actors from regional and local areas become involved. They are developing application businesses related to AI to address the everyday problems of citizens and corporations.

“So-called AI startups are often just wrappers for AI products from global platform companies. We don’t want that to happen to the startup scene in Indonesia,” said Andreas, who is also the Co-Founder of Compedium, at the AI Impact Summit, part of the third day of the Bali Tech Week 2026 forum, Wednesday (10/22/2026), in Denpasar, Bali.

According to Andreas, they should develop business AI applications that truly serve as solutions for regulated industries, such as banking, healthcare, and agriculture.

With this approach, Indonesia’s AI startup landscape can protect itself from the dominance of global players. This approach doesn’t aim to compete with the AI platforms created by global tech giants, but rather focuses on building practical applications and strengthening talent.

Some AI-based startups in Indonesia include Nexmedis, Jala Tech, and Algobash. Nexmedis helps hospitals automatically screen social security claims, cutting time by 60–70 percent, and transcribing local dialects into medical standards.

Jala Tech develops AI-based solutions to detect diseases in shrimp, helping farmers reduce the risk of crop failure.

Meanwhile, Algobash is developing AI-based solutions to support a recruitment process that is more focused on the skills and competencies of applicants. However, the AI technology being developed is not designed to make final decisions regarding which candidates are accepted or rejected. That authority remains with humans, specifically the company’s human resources team conducting the recruitment.

The Director of The Founder Institute Jakarta, Andy Zain, acknowledged that the development of AI is also changing the way individuals start businesses. This technology allows individuals to establish companies without the need to first form large teams and pursue significant funding.

With the help of AI technology, an individual can start a business more quickly. They do not need to first have a team or raise funds to establish a venture. Founders can also utilize AI agents to assist with various tasks that are being carried out.

However, he believes the term “one-person company” is more of a marketing term than a fully realistic depiction of how a company grows. The idea that one person can build a multi-million dollar company is appealing, but it shouldn’t be the primary assumption.

“As an investor, I actually have a somewhat dilemmas perspective on that concept. Would I invest in a company that is run by only one person? That is quite concerning, because if something happens to that person, the entire value of the company could be affected,” he said.

Eddi Danusaputro, Chairman of the Indonesian Venture Capital and Startup Association (Amvesindo), also met on the sidelines of Bali Tech Week 2026, shared a similar view. The development of AI-based startups in Indonesia is beginning to show progress.

However, most of them are still focused on utilizing or implementing AI technology, rather than creating basic technologies like large language models (LLMs). The development of basic AI technology is still dominated by companies from countries like the US and China.

According to Eddi, the market opportunity for AI startups in Indonesia remains significant because the technology can help solve various problems while improving business operational efficiency. The use of AI has reached various sectors, including healthcare, education, banking, and financial services. However, he does not yet have data on the most dominant sectors in AI development and use in Indonesia.

In terms of funding, Eddi acknowledged that investment in AI startups in Indonesia is still relatively small. Developing AI technology at the infrastructure, original model, or platform levels would require significant investment, making it difficult to compare with global technology companies that have already entered these business segments.

Citing a Crunchbase report (June 15, 2026), the surge in investment in AI-based startups has not been evenly distributed across the globe. Based on investment data collected by Crunchbase, nearly 80 percent of global startup funding, from seed to growth, has gone to US-based companies.

This contrasts with the period before the AI investment boom, when US companies typically received less than half of all global startup funding.

Most of the funds are even concentrated in two AI platform development companies, namely OpenAI and Anthropic. This concentration of investment indicates that the surge in global AI funding is primarily supported by a small number of large technology companies in the United States.

Outside the US, several countries are still seeing growth in startup funding. China, for example, has secured more than US$33 billion in startup investment throughout 2026 as of mid-June. This figure surpasses the total funding for all of 2025.

The United Kingdom also recorded funding of 16.5 billion US dollars, approaching a total of 19.5 billion US dollars throughout 2025, with AI and financial technology as the main sectors. Meanwhile, India, Japan, South Korea, Germany, France, and Spain showed relatively stable or moderately increasing funding developments.

Crunchbase also highlighted the potential for an investment bubble to emerge due to the massive concentration of AI funding in the US, which is home to just over 4 percent of the world’s population.

Writer:

Caecilia MedianaEditor:

Budi Suwarna

More from 22 October 2026 →

Cite this article
Source: jawawa.id (https://jawawa.id/newsitem/ai-makes-it-easier-to-build-start-ups-but-business-models-remain-a-challenge-1792655411)
Originally published by kompas.id (https://www.kompas.id/artikel/building-a-startup-is-getting-easier-because-ai-business-models-are-an-issue?open_from=English_Page)
Print