This article first appeared in Digital Edge, The Edge Malaysia Weekly on April 14, 2025 - April 20, 2025

Being an angel or venture capital investor, especially in start-ups, is often seen as glamorous. The media tends to romanticise VC investors who pour millions — sometimes billions — into chasing the next unicorn (a company valued at over US$1 billion). Every time a unicorn is announced, everyone goes ga-ga over the start-up and the investors who they believe will make a gazillion dollars from their investments.

What they don’t realise is that for every unicorn, there are thousands of start-ups that never make it and fail, leading to huge losses for investors. Rarely do the media cover such stories unless there is something juicy, like a well-known investor losing a lot of money or if there is some fraud or mismanagement involved, primarily with well-known start-ups and their founders.

This article is not, however, about founders or any particular investor. It is to highlight the fact that investing in the new, new thing in the technology industry is actually very challenging and the risks that investors take are very high.

Let’s start with the electric vehicle (EV) industry, a sector often seen as exciting and progressive. Championed by sustainability advocates, EVs were promoted as a cleaner alternative to petrol and diesel, which not only emit harmful gases but also harm the environment through oil drilling.

Tesla led the EV charge, becoming the most recognisable brand in the space. Driving a Tesla was once considered glamorous and its founder Elon Musk rose to fame as both a renowned entrepreneur and the world’s richest man.

Among the bigger support industries for EVs are battery manufacturers and charging stations. While Malaysia cannot compete in building EVs, especially with Tesla and the emerging new giants from China like BYD — and we are also not able to compete in battery manufacturing — EV charging stations are one area that has seen growth.

Malaysian investors have limited exposure to the EV industry, with the main opportunity being in EV charging stations, a segment that has seen steady growth. Over the past five years, both locally and globally, numerous companies have entered this space to address key concerns among EV buyers: “range anxiety” and long charging times.

Currently, a fully charged EV can travel about 450km, depending on battery size, and charging can take anywhere from 30 minutes to 10 hours. For example, a round trip from Kuala Lumpur to Johor Bahru would likely require a recharge. As EV adoption increased, so did the demand for charging infrastructure, drawing interest from start-ups and major companies alike.

I was invited to invest in the EV charging space but chose not to. The main concern was the rapid pace of battery innovation, which made it difficult to forecast the industry’s direction even five years ahead.

Moore’s Law, coined by Intel co-founder Gordon Moore, observes that the number of transistors on a microchip doubles roughly every two years, boosting computing power and lowering costs. While originally about semiconductors, a similar trend seems to be playing out in battery tech, with ranges increasing and charging times decreasing rapidly. This trend has already begun to materialise, making long-term investments in charging infrastructure harder to justify.

Currently, most batteries are lithium-ion batteries but both US and Chinese researchers and manufacturers are currently developing batteries using new materials that can last 1,600km on a single charge. That is enough to take you from Penang to Singapore and back with room to spare. In the next six years, if Moore’s Law holds true, then you may be able to travel from Singapore to Shanghai and back (7,600km) on a single charge.

Additionally, BYD has already developed fast chargers that can provide a full charge in just five minutes. In time, it will likely only take one to two minutes to charge a battery (Moore’s Law again). This is as fast as filling up your petrol tank currently.

In the near future, if there are sufficiently large numbers of EVs on the road, I imagine current petrol stations having EV chargers as well to serve EV owners since charging time will be the same as petrol filling time. What would this do to all the charging stations globally? While some will need to upgrade their technology to the new super-fast chargers, most will not be necessary, especially with the extra-long-range batteries as owners can just charge them at home or at petrol stations.

This is emblematic of the difficulty for investors to invest in new technology. Things change so fast that what looks like a great investment today can become a really bad one in just a few years.

The same thing is happening in the artificial intelligence (AI) industry. According to Crunchbase, a total of US$334 billion was invested in AI start-ups in the last five years with 2024 alone accounting for US$101 billion or almost a third of the last five years.

However, trends are changing as most of the larger companies, from Anthropic and OpenAI to Google, Meta, Microsoft and also Alibaba and Tencent in China, are all pumping billions into AI and the largest VCs are now betting on only a few large AI start-ups.

Also with AI, investors will be hard-pressed to select what to invest in, whether it is in the base AI modelling company, AI tools or agents, or in companies using AI in their applications or software products. The vast majority of AI companies are losing money (some analysts predict OpenAI may lose US$5 billion this year) and RAND Corp found that 80% of all AI projects fail. In fact, most AI start-ups don’t actually have a way to make money; even projects by the big boys like Google and Microsoft don’t have a strong monetisation model.

All of this shows us that there are huge risks in investing in the latest technologies. So, what should investors do?

Just like EV charging and batteries, if an investor had considered Moore’s Law and looked at what the future could be, he could have predicted that things could change dramatically and this would mean the sector he was investing in had a much higher risk of failure and hence he could have tempered his investments or, at least, ensured that the companies he invested in were prepared for the future.

Never jump into an investment just because “the big boys” were also investing in the same industry. If you just follow the herd, you will likely be trampled when they panic and stampede. If too many people invest in any particular sector, that likely means there will be too much competition and most investments will fail both because of competition and because when the big companies also pump in money into the same technology, your tiny start-up will never be able to compete.

Most AI start-ups don’t really have a real monetisation model and even if they do, they may not be able to make money for the longer term. The fact that 80% of AI projects fail means that they never found a way to make money and even companies like OpenAI struggle to make real money.

A moat means a defensible position. With most investments, if you cannot build a strong competitive position and be able to defend it from competitors, then you don’t have a moat (like the moat filled with water that surrounds a castle and protects it from attackers). Without a moat, the start-up is highly vulnerable to competition, especially from better-funded start-ups.

In industries like AI, sometimes it’s better to invest in companies that are using AI tools built by others to make their own applications better for customers. It’s like the old story during the gold rush in California. The ones who got rich were not the gold prospectors but those who sold shovels and tools to the prospectors. In using AI for applications, for example, a call centre could use AI agents to take on some work from human call centre staff and this helps create more efficiencies and also cuts costs for the call centre business. AI agents today can replace call centre staff as they can be trained to answer any query a potential user or customer has in a human-like voice and manner. There are many ways to use AI and it’s probably better to invest in such companies than in companies developing tools with no monetisation model.

Investing in new tech is highly challenging and most investors will lose money. A smart and thoughtful investor has a better chance of making money than those who just follow the herd and invest without thinking hard enough about the risks inherent in new tech investments.

Dr Sivapalan Vivekarajah is co-founder and senior partner at ScaleUp Malaysia Accelerator (scaleup.my) and adjunct professor at Sunway University’s School of Science and Technology. He is also the author of Supercharge Your Startup Valuation.

Save by subscribing to us for your print and/or digital copy.

P/S: The Edge is also available on Apple's App Store and Android's Google Play.