SEA of Startups

Decoding the Pulse of Founders, Capital & Conviction in Southeast Asia.
  • 5.0 (1)
  • Entrepreneurship
  • Updated Weekly

Real, raw, relatable takes on Southeast Asian startups. One investor, the week's news, no script. seaofstartups.substack.com

  1. 2d ago

    Just passing through: how Malaysia keeps funding the people who leave

    Over the past few weeks, three Malaysia stories hit the news that, on the surface, have nothing to do with each other. A government pension fund answered in Parliament for nearly RM200 million lost in a fish farming startup that turned out to be a fraud. A celebrity founder and her husband sat in a courtroom over money that came from two of the biggest state funds. And a quasi-crypto commune in Forest City had its license pulled by the local council and announced it was leaving. A fraud, a trial, and a controversy. Three different casts, three different genres. And underneath all three sits one uncomfortable pattern about how Malaysia spends its public money and who actually ends up on the receiving end of it. Here it is in one line, and the rest of this post is me proving it: when the Malaysian state goes looking for the future, it keeps handing its money and its land to people who are just passing through. And the ones who stayed, who put down roots and built something here, are the ones it keeps overlooking. One. Four audit firms, and nobody counted the feeders Start with the pension fund, because this is the one that should make you angriest, and not for the reason you think. The fund is KWAP. It manages the retirement savings of Malaysian civil servants: teachers, nurses, clerks, the people who keep the country running. About RM195 billion under management, more than RM8 billion in investment income last year. A serious, professional institution. In July 2023, KWAP put nearly RM200 million, call it US$47 million, into eFishery, the Indonesian startup that made internet-connected fish feeders and sold itself as the future of aquaculture in Southeast Asia. You already know how this ends. eFishery was one of the biggest startup frauds this region has ever produced. The company kept two sets of books. It claimed roughly 400,000 smart feeders deployed in the field. The real number was about 24,000. The fleet was inflated more than fifteen times over, and revenue was inflated to match. The founder was sentenced to nine years in an Indonesian prison, since trimmed to six on appeal. The story is back in the news because the Prime Minister stood in the Dewan Negara this week to answer for it, and the anti-corruption commission has opened a probe. The easy story, the one a lot of people wanted, is that somebody was lazy or asleep at the wheel. I do not think that is what happened, and the truth is far more useful. KWAP did not skip the diligence. It was part of a consortium that included SoftBank and Temasek, serious money with serious teams. Between them, the investors hired four separate audit and diligence firms: PwC, Grant Thornton, EY, and KPMG. They hired six more firms to survey the market and validate eFishery’s position in it. They hired Kroll, the corporate investigations outfit, to run background checks on the founders. That is millions of dollars of the most reputable professional diligence money can buy. Every single one of them missed it. Because all of that diligence was done on paper. Financial statements verified, documents cross-checked, management interviewed, the numbers in one data room matched against the numbers in another. As far as I can tell, nobody got in a car and drove out to the fish farms to count the feeders. If a single one of those firms had spent one week doing what any private equity analyst is taught on day one, go to the site, walk the floor, talk to the actual customers, they would have found 24,000 machines where the company promised 400,000. The fraud was not hiding in the accounts. It was sitting in plain sight in the fields, where nobody bothered to look. Every founder and fund manager reading this should burn that in: diligence on documents only tells you the documents are consistent. It does not tell you the documents are true. The cheapest, most boring check in the entire toolkit, physically going and looking at the thing, is the one nobody did. Two fairness notes. RM200 million against a RM195 billion fund is a rounding error, one tenth of one percent. Nobody’s pension is at risk, and when you hear the political noise, keep it in proportion. This is embarrassing, not an emergency. But it is exactly because the money was small that the failure matters. This was not a bet that went wrong. It was a bet that was never really examined, waved through on the strength of who else was in the round. SoftBank is in, Temasek is in, the auditors signed off, so we are in. That may pass in public markets. In private investing it is not investing, it is following. And when a Malaysian pension fund follows a crowd of foreign funds into a foreign fraud, you have to ask the question this whole post is about: what did any of it have to do with building Malaysia? Two. They bought the face, not the business I am going to be careful here, because this is a live trial. The founders have pleaded not guilty, and I am not here to convict anyone from a microphone. Everything about the alleged conduct is exactly that, alleged, and the courts will decide in due process. But the investment itself, the money going in and the money coming out, is a matter of public record, and that part is fair game. The company is FashionValet, the Malaysian fashion e-commerce startup founded by Vivy Yusof and her husband. Vivy was, and is, one of the most recognizable entrepreneurs in the country, a genuine influencer before that word got cheap, with a modest wear brand, dUCk, that people genuinely loved. She became the face of a certain kind of Malaysian success story. In 2018, two of the largest state funds invested: Khazanah put in RM27 million and PNB put in RM20 million. Call it RM47 million of public money into a homegrown fashion brand. Here is the number that should stop you. When the two funds eventually sold their stakes, they got back a combined RM3.1 million. A loss of roughly RM44 million, confirmed by the Ministry of Finance. Startups lose money, understood. But the losses were visible before anyone wrote a check. FashionValet lost money every single year it operated, and the losses grew from a few hundred thousand ringgit to more than RM10 million a year. Six straight years of red ink, and the funds looked at that and invested anyway. So the obvious question is why. I was not in the room, and I arrived in Malaysia around that time without the context to judge it then. But I do not think the honest answer has much to do with the business. I think the honest answer is worse, because it is a pattern rather than a one-off. They did not buy a business. They bought a face. A narrative, a following, the magazine covers, the idea that if you back the most famous young founder in the country, some of that shine rubs off, and the national funds get to say they are backing national icons. It feels modern. It ticks the marketing boxes. It photographs well. But an icon is not a business model, and a following is not a balance sheet. When the thing you actually bought is a personality, you bought a fragile asset, because the moment public sentiment turns, and in the influencer game it always eventually can, your investment turns with it. And here is the detail that tells you this is a real lesson and not just hindsight dressed up as insight: Khazanah itself, after the whole thing blew up, publicly warned about the risk of what it called icon-driven businesses. The fund said the quiet part out loud. Betting on personality is a structural mistake. Now put the two stories side by side, because they rhyme. In eFishery, the funds bought a foreign founder’s story and never checked the fields. In FashionValet, they bought a local founder’s story and never respected the profit and loss statement. One was a fraud, one was just a bad business, and those are very different things. But the investor mistake underneath both is the same: fund the narrative, skip the boring verification that would have told you the narrative was hollow. In both cases, public money went in on the strength of a name, not a capability. Three. Three names, one machine Two bad deals is just two bad deals. The reason this is a post and not a shrug is that these are not isolated checks. This is the machine working the way it has always worked. For more than twenty years, the government has tried to build a venture capital scene by pouring public money into funds and programs. Walk the track record. MAVCAP, the state venture arm going back to the early 2000s: over half a billion ringgit committed over the years, and by last reporting somewhere around RM200 million and change had come back. Roughly 38 sen home for every ringgit out. Allow for liquidity and unrealized positions, it is still a very slow, very official way of setting money on fire, and the local venture scene did not become self-sustaining on the back of it. Then came 2020 and Penjana Kapital, a fund of funds worth hundreds of millions, launched to kickstart the sector after the pandemic. Big launch, big numbers. Years later, remarkably little of the money had actually been deployed. The government, the local investors, and the foreign partners all wanted different things, and the apparatus seized up. And after two stalls, it did not stop. It rebranded. The prior attempts were consolidated under the sovereign fund and relaunched in late 2024 under a new name, with a new commitment of capital, as part of an even bigger government mobilization effort. Fresh name, fresh logo, fresh press release, fresh faces. MAVCAP, then Penjana, then the new entity. Three names, one machine, more than two decades. Fairness again, because it matters: real managers got their first checks from these vehicles, some very good local companies were backed, and many of the people involved are smart and sincere. The new iteration is too early to judge, and I genuinely hope it does better. But the machine, as a machine, has not built the thing it was built to build. There is still no self-sustaining venture

    Just passing through: how Malaysia keeps funding the people who leave

  2. Jul 15

    The $7.4 Billion Lie

    You have seen the number this week, probably five or six times, from five or six people who all copied it from the same report. Southeast Asian tech funding hit 7.4 billion dollars in the first half of 2026. More than double last year. Recovery is here, the drought is over, break out the good coffee. It is true. It is also one of the most misleading true things I have read all year. Because 4.5 billion of that 7.4 billion went to a single company. One. A data-centre operator. Take that one company out, and on the exact same set of numbers, the region did not double. It went sideways, and depending on how you count, slightly down. And while we are here: when did we start counting data centres as startup funding at all? That is a genuine question, and it is going to matter more than it sounds. One landlord, not a region Here is the full picture, because the detail is where the headline falls apart. First half of 2026, 7.4 billion raised across Southeast Asia, against 3.2 billion in the same six months last year on the same source. On paper, up 130 percent. Now pull the thread. Of that 7.4 billion, 4.5 went to DayOne, a Singapore-registered data-centre operator, across two Series C rounds to fund a build-out. That is more than 60 percent of everything that flowed into the entire region, in one company, for concrete and cooling and racks. This is not a knock on DayOne. They did nothing wrong. Raising four and a half billion dollars is not a crime, it is a very good year. The problem is not the company. The problem is that we take their balance sheet and hand it to founders across five countries as if it were their momentum. Strip DayOne out and the region raised roughly 2.9 billion in six months, which is less than the 3.2 billion it raised the year before. The honest headline is not “funding doubled.” It is “one landlord had a great six months, and everything else went slightly backwards.” It gets worse when you look at where the money sat. Singapore captured 6.9 of the 7.4, over 90 percent, and still climbing. So this is not a Southeast Asian story. It is a Singapore data-centre story. And even that is a little bit of a fiction, because much of the physical build is not in Singapore at all. It is in Johor, across the causeway in Malaysia. The concrete goes up in Johor, the capital gets booked in Singapore, and the statistics tell you Singapore is booming. The map and the money have stopped agreeing with each other. One caveat to hold onto, because it trips people up. Around the same time, KKR and Singtel bought ST Telemedia’s data-centre business for about 5.2 billion. Huge, and real, but that is mergers and acquisitions. One company buying another. It is not venture funding and it is not in the 7.4 billion. If someone stacks the two and tells you data centres pulled in ten billion, they are double-counting. The money went into concrete. Whether a founder in KL, Jakarta or Ho Chi Minh City ever sees a cent of it is a separate question, and so far the answer is no. And here is the part that should sting. Fintech. Payments. The thing this region was supposed to be about, the super-apps and the wallets and the great Southeast Asian consumer story we told for a decade. Fintech raised 685 million dollars in the first half. Not a slow year. A sector that is basically over as the headline act, and nobody held the funeral. So the founders leave, into a narrower door Now widen the lens, because the timing matters. The same six months that Southeast Asia congratulated itself on 7.4 billion, global venture funding hit a record 510 billion, a record half driven almost entirely by the AI hype. Of that 510 billion, two companies, OpenAI and Anthropic, raised 217 billion between them. Two American AI labs pulled in 43 percent of all the startup funding on Earth in six months. Put the numbers side by side. All of Southeast Asia raised 7.4 billion, and ex-landlord, call it 2.9. Two AI labs in San Francisco out-raised our entire region by something like 75 to one. We are a young market, I get that. But 75 to one, two companies against a region, is not a gap you shrug off. So what does a smart, ambitious founder do with that information? Some of them are already answering it. They are leaving. Founders who launched in Singapore in 2025 packed up in April and May and moved to the Bay Area. This has always happened, but it is becoming a steady trickle, which is worse, because a trickle does not make the news. It just quietly drains the pool. Here is where I want to be careful, because there is a lazy version of this story. The lazy version is: the money is in San Francisco, so move there and get funded. That is not true anymore. The money in the US has concentrated too, and not just by geography. It has concentrated by story. Look inside that record US number and 86 percent of it went to AI. The same brutal filter is running there, just on a different axis. In Southeast Asia the filter is one landlord. In the US it is one narrative, and if you are not telling it, the cheque book stays shut. Think about what that does to the bar. There used to be a respectable way to raise. You grew triple, triple, double, double, double. You built a business that compounded, showed durable revenue, and that was a clean Series A. That founder today walks into a room in San Francisco and gets a polite no, because the person across the table is not looking for durable. They want a thousand-x. They want the AI story that eats a category in eighteen months, and a healthy business that doubles every year sounds boring next to it. You did not escape the filter. You swapped a filter you understood for one that is even harder to clear. And I want to be fair, because it would be easy to turn this into a loyalty test, and that is not honest. The founders who leave are not traitors. They are moving toward the center of gravity, and San Francisco genuinely is the center of gravity for building right now, especially in AI. But nobody should sell you the fairy tale that the flight to SFO ends with a term sheet. The center of gravity is also the most crowded, most selective room on the planet, and this year it is writing cheques for exactly one kind of story. Whether the founder stays or goes, the answer is the same shape. Here, the money went to a building, not a founder. There, the money goes to one narrative, not a founder. Either way, the ordinary, good, growing company, the backbone of any real startup scene, is the thing nobody is funding. We built a region that funds the warehouse and exports the talent, and the place we export it to only wants that talent if it can promise a miracle. Fewer deals, but not better ones There is a comeback I always get here, and it is a fair one. Deal count is down, sure, but that is discipline. The market matured. Fewer, bigger, better deals. Quality over quantity. This is healthy. I would love to believe that. In the first half of 2026 there were 127 funding rounds across the region, down from 153 a year earlier. Fewer deals, yes. But look at where the money inside them went. Six billion of the 7.4 went into just twelve rounds of a hundred million dollars or more. Twelve rounds took six billion. The other 115 rounds, every seed cheque, every Series A, every founder not raising nine figures, split roughly 1.4 billion between them. That is not discipline. Discipline is looking at a hundred good companies and carefully backing the best thirty. This is a hundred companies looking up at twelve giants eating almost everything, and scrapping over the crumbs. When the top twelve deals take 80 percent of the capital, that is not a mature market. It is a bare cupboard with one very full shelf. And before anyone tells me last year was some golden baseline we have fallen from, no. Last year was the same shape. In the first half of 2025, fintech was carried by three deals that made up more than half of all fintech funding, and Singapore took over 90 percent of the pie even then. The concentration is not new. It is not a one-off. It is the structure. Southeast Asian venture has run on “one or two deals carry the whole region” for at least two years straight. The only thing that changed in 2026 is that the one deal got bigger, so the number got louder, and the lie got easier to tell. Read the middle of the list Let me be clear about what this is and is not. This is not doom. I am not telling you the region is dead, or that nobody should build here, or that we should all give up and move to California. Plenty of good companies are being built here right now, quietly, with real revenue, and they deserve better than to be background noise behind a data-centre headline. Which, again, I still do not understand why we file under startup funding at all. What I am asking for is honesty about the number. Stop reading 7.4 billion as a sign of health. It is not. It is the balance sheet of one landlord plus a rounding error for everyone else. If you want to know how Southeast Asia is actually doing, do not look at the top deal. Look at deal number three, and deal number fifty, and deal number 127. Look at whether a seed-stage founder in Kuala Lumpur can raise a real round without moving to Singapore first. Look at whether the best people are staying or leaving. Right now, on the honest read, the top of the market is a landlord, the middle is thin, and the sharpest founders are heading to the airport. Until the number without the landlord starts going up, we are not narrating a recovery. We are narrating a story we would like to be true. Real. Raw. Relatable. If this one annoyed you, good. That means you were paying attention. Tell me where I am wrong. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit seaofstartups.substack.com

    The $7.4 Billion Lie

  3. Jul 8

    Who Owns the Scarce Thing?

    This week the two biggest stories in Southeast Asian tech were not a funding round or somebody’s ninth super app pivot. They were a government chip program in Penang and 3,600 kilometres of fibre being dropped on the seabed between India and Singapore. Two boring stories. Laid side by side, they are the most honest picture of this region you will get right now. Both are asking the same question, the one I ask in every partner meeting at Indelible, the one that decides who gets rich over the next ten years and who just gets used: Who owns the thing that is actually scarce? Malaysia tries to climb a rung On 1 July, MTDC, the Malaysian Technology Development Corporation, launched the first cohort of Semicon Start Malaysia. Ten companies picked from 39 applicants. A pot of RM10 million for the first phase, up to RM1 million per company, call it US$250k apiece, with Khazanah money in the mix. If you have been in this region as long as I have, your first reaction to “government launches program to build high-tech industry” is a small, tired sigh. We have seen this film. Malaysia has a graveyard of these: grand corridors, MOU signings, innovation valleys, state venture funds that wrote checks into slide decks and got slide decks back. Big announcement, ribbon, photo, handshake. Two years later you go looking for the companies and nobody is home. I had that sigh ready. Then I stopped, because this one has the potential to be different, and the reason why is the whole point of this piece. This time there is a real industry underneath the program. Penang is not a hopeful press release. Penang has been doing semiconductor assembly and testing for decades. A serious slice of the world’s chips passes through Malaysian hands on the way to being packaged and tested. That is not a pitch. That is payroll. Factories that have run for thirty years, and a workforce that already knows the difference between a good die and a bad one. So the bet is not “let’s conjure a chip industry out of nothing.” The bet is much narrower, and potentially much smarter: we already own one rung of this ladder. Can we climb one step up into design, where the money actually sits? The climb has already started without the program. SkyeChip, a homegrown Penang design house doing genuinely hard work (high bandwidth memory, chiplets), listed on Bursa’s Main Market. Before recording this week’s episode I saw a report suggesting Cerebras, the US chip company that also just went public, may be tapping SkyeChip for design work. I have not verified that, so hold it loosely. But the proof point stands either way: a local company has already climbed the exact rung the government now wants ten more companies to climb. Add the National Semiconductor Strategy from a couple of years back, Penang’s own chip design academy, and Selangor standing up a state fund, and you have something rarer than a press release. You have momentum with an industry underneath it. The timing is as good as it has ever been, too. The world wants to diversify where its chips come from. Nobody wants every advanced part made in one strait that could close on a bad Tuesday. Malaysia is neutral, capable, and already in the supply chain. If there was ever a decade to attempt this climb, it is this one. Now the hard part, out loud, because that is what this show is for. Money was never the thing missing here. What has been missing, every single time, is patience and expertise arriving in the same envelope as the cash. A million ringgit and a short program do not build a chip design house. Chip design is a long-term sport played by people who have failed at it a few times first. If Semicon Start is a check and a demo day, it joins the graveyard. If it comes with real design mentors, real customer introductions, and follow-on money that does not vanish when the photo op ends, it has a shot. So the thing to watch is not the RM10 million. It is whether anyone attached to the program has real operating expertise. Money is easy. Knowing what to do with it is the scarce part. Hold that thought. The cable, and what it actually is Now to the seabed. This week it was reported that Microsoft, together with Singapore’s Lightstorm, is leading a consortium building a new subsea cable called I2C: roughly 3,600 kilometres of fibre linking India to Malaysia to Singapore, targeted to go live around 2029, built for AI and data centre demand. Standard disclaimer, because I read these announcements the way I read a pitch deck: this is a 2029 project, consortium details on these things move around, and I have not seen final paperwork, just a news story. Treat the specifics as direction, not gospel. But the direction is what matters. Every few weeks now there is a story like this. A new cable, a new hyperscaler campus, somewhere with cheap power and a friendly minister. And every one of them gets written up as billions pouring into Southeast Asian digital investment. Celebrations all round. Here is what I actually see, and maybe I am a bit cynical: the region being wired up as a very good place to host other people’s compute. The fibre lands here. The data centres sit here. They use our power and our seabed. That is real economic activity and I am not pretending it is nothing. But ask the only question that matters. Who owns the compute? Who owns the demand sitting on top of that cable? Generally, not us. The demand is offshore, the models are somebody else’s, and the margin, the part where value actually compounds, is in Seattle and San Francisco, not Johor. We are the landlord renting out the ground floor, being told to feel grateful for the rent. I am a capitalist. Rent is not a dirty word. It is a perfectly good business, and Singapore has run that playbook for fifty years. But do not confuse being the landlord with owning the building. A region cannot tell itself it is climbing the value chain when what it is actually doing is leasing the basement to the people who own the value chain. This is where the cable and the chips rhyme. Same story, pointed in opposite directions. Malaysia’s chip program is a country trying to own more of the building. The cable is the region agreeing to stay one rung down. One is a strategy. The other is a lease dressed up as a strategy. What is actually scarce Value flows to whoever controls the scarce thing. It always has, AI or no AI. Find what is scarce, own it, and the money flows to you. Own something abundant and you compete it down to nothing. So: in Southeast Asia right now, what is actually scarce? I will tell you what is not. The technology is not scarce. The model is not scarce. Models are commoditizing in front of us, between the big labs’ price war and open source, and they will get cheaper and better every quarter whether you do anything or not. Building your moat on the model is building your house on the tide. Here is what is scarce. The customer who already trusts you. The physical network that took years and real pain to build. The license from a regulator who does not hand them out twice. Distribution into the towns and small shops that no hyperscaler in the world will ever bother to map. The workflow nuance that took ten years of unglamorous work and cannot be copied in a weekend of clever prompting. That is the scarce layer. That is the thing worth owning. Where the winners come from Look back at the two stories through that lens and they light up. Malaysia is trying to move from an abundant thing (cheap, capable labour, which everyone has) to a scarce thing (design capability, which very few have). Right instinct. Own the scarce rung. The founder version of the same move: the winner is not the one who owns the AI and goes hunting for a customer. The winner is the one who already owns the customer and quietly adds AI on top. The lending business that already has the borrowers and now underwrites them better. The logistics operator that already owns the trucks and the routes and now runs them tighter. The distributor who already reaches 10,000 shops and now forecasts demand for them. Those companies will never put AI in the headline. They do not need to. They already own the scarce thing. The AI is just a sharper tool in a hand that already knows the work. I know that is not a fashionable thing to say in 2026. Every second founder I meet opens with the model they are building on, the AI-native this, the agentic that. The funding tallies love it: somebody counts up the AI startups that raised this quarter, puts out a chart, and everyone nods. But that chart measures ambition, not durable revenue. Those are very different things, and the gap between them is where founders and their investors go to die. And here is the uncomfortable part I want founders to sit with. Every wave of cheap capital, every shiny new tool, every drop in the price of intelligence does not close the gap between those two kinds of companies. It widens it. When the tool gets cheap and everyone has it, the tool stops being the difference. The only difference left is the position underneath: the distribution, the trust, the scarce layer. Cheap AI makes owning real distribution worth more, not less. Be honest about what you own This is where Indelible puts its money, and I will say it plainly so you can hold me to it. We back people who own the scarce layer, or are credibly climbing one rung towards owning it. Not people standing on top of somebody else’s scarce layer with a nicer logo. (None of this is investment advice. It is simply where my money already is.) So the homework this week, if you are a founder: be honest about what you actually own. Not what is in your headline. What is in your foundations. If the answer is a really good wrapper around somebody else’s model, it is better to know that now. Using a commodity as an input is perfectly fine. Every company will. The question is what you own on top of it. A chip program in Penang. A cable on the seabed.

    Who Owns the Scarce Thing?

  4. Jul 1

    One Winner, Six Shipwrecks

    Since 2017, Southeast Asia has produced exactly one tech IPO that made public investors real money. One. And this week, the Philippines is getting ready to bet its entire year on the next one. So this week I want to talk about who is buying, who is selling, and which side of that trade you actually want to be standing on. Four stories, and they braid into one. We open with the good news, because there usually is some. Then we follow the money all the way to the part nobody puts on the deck. The smart money showed up twice in one week Start with the hopeful, because it is real and it is specific. This week two of the most serious institutions on the planet made their first proper bet on Southeast Asia. Not a press tour. Not a memorandum of understanding. Actual money into actual companies. The first: MIT, the university, joined the cap table of a Singapore company called PVX Partners. Not a flashy name, I had not heard of them before this. They do cohort-based financing for user acquisition. In plain terms, they fund the marketing spend for mobile games and consumer apps, and they get paid back out of the revenue those users generate. It came on the back of a ten-plus-million-dollar round with names like General Catalyst, and I think a DraftKings vehicle in there too. As far as I could find, this is MIT’s first major disclosed startup bet in the region. The second, and this one landed the day before I recorded: Pfizer Ventures, the drug giant’s venture arm, made its first Southeast Asian startup investment into a Singapore biotech called Engine Biosciences. Engine does AI-driven precision oncology, hunting cancer drugs with machine learning. They just opened a Silicon Valley office to go with the Singapore base. Here is why this is not just a funding roundup. When an elite American endowment and Big Pharma’s investment arm both pick Singapore companies for their opening move, in the same week, that is not a coincidence. That is a signal about where sophisticated capital now thinks the edge is. These are not tourists chasing a hot round. PVX is unglamorous infrastructure. Engine is deep science. Both are the kind of bet you make after you have done the work. Hold that thought, because the rest of this is about what happens to the money that was already here when it tries to leave. The Philippines is betting its whole year on one listing On the 27th, Mint, the company behind GCash, filed its registration with the Philippine SEC and its listing application with the stock exchange. The number: up to 92.3 billion pesos, roughly 1.5 billion US dollars at up to ten pesos a share, targeting a fourth-quarter debut. If it prices at the top, it is the largest IPO in Philippine history. Sit with the context. The Philippines’ IPO count for 2026 before this filing was zero. Nothing. So the country’s first listing of the year is also the biggest it has ever had. And it is a fintech, which if you have listened before you know is my home-turf bias made concrete. GCash put financial services into something like 90 million pockets. It is the rare regional company that is genuinely profitable. The pitch writes itself: the people who made GCash a habit can now own a piece of it. I want this to work. Let me say that plainly. Now the part that worries me, out loud, because that is the point of these episodes. The float is about 12%. Twelve percent of the shares go to the public market. The public is being sold a fairly thin slice while insiders keep the rest. And to fit GCash into its main index, the exchange is now considering cutting its own minimum public float rule from 20% down to as low as 12%. Take that in. The benchmark is bending its own rules to accommodate one company. When a market reshapes itself around a single listing, and that listing is carrying the whole nation’s IPO year on its back, that is not a recovery. That is concentration risk wearing a party hat. The real question: does GCash trade well enough to reopen the pipeline for everyone waiting behind it, or does one wobble set the Philippine market back another two years? To answer that honestly, you cannot just look at GCash. You have to look at what happened to the last batch of regional champions that rang the bell. Indonesia got a stay of execution, not a clean bill of health While Manila is opening a door, Jakarta is trying to keep one from closing. On the 24th and 25th of June, MSCI, the index provider whose decisions quietly move billions in passive money, deferred its decision on whether to downgrade Indonesia from emerging-market status to frontier. They kicked it to November. Indonesia keeps the badge, for now. Why was it even on the table? MSCI said, in effect, that it cannot trust the market. Lack of transparency in who actually owns the shares. Suspected coordinated trading that makes it hard to know what a fair price even is, or how much stock is genuinely free to trade. And the market rallied on the news. Here is where I get off the celebratory bus. That rally is celebrating a delay, not a fix. When the index provider tells you it cannot work out who owns the shares or what they are really worth, that is not a paperwork problem. That is a governance warning about the entire market. And look at the response. Indonesia is leaning on Danantara, the sovereign fund, plus insurance and pension money, to add buying support and prop up the exchange. Think about what that means. To pass a test about transparency and genuine free float, the answer is to bring in state and pension money to hold the market up. That is close to the opposite of the thing they are being asked to prove. A frontier downgrade is not abstract. It would force passive funds to sell Indonesian equities mechanically, which raises the cost of capital for every late-stage founder in the country dreaming about an IPO on that market, especially now without the hype cycle. November is closer than it sounds. Manila might be opening up, maybe. Jakarta is one review away from being pushed out. Hope on one side, risk on the other. So let me put some numbers on which way this bet usually goes. The receipts I promised you a number at the top. Here it is with the receipts. Since 2017, this is how Southeast Asia’s big tech IPOs have actually treated the public investors who bought in. SPAC valuations are listing marks, not day-one closes. Dollar figures are dragged by weak pesos and rupiah. Current values approximate. One winner. Sea Limited went out at a $4.9 billion valuation and trades somewhere in the $56 billion range today. Everything else is a shipwreck. Grab is down around 60% from its listing cap. GoTo lost roughly nine-tenths of its value. Bukalapak is trading below the cash it raised. Converge, the one Philippine name I could pull, is the cautionary tale sitting right next door to GCash. Now the caveats, out loud, because the show runs on honest data. The SPAC valuations were listing marks, not day-one closes, and several fell on the open. Currency matters too: weak pesos and rupiah drag the dollar figures down. On a per-share basis the returns are often worse than the market-cap numbers suggest, because of share issuances along the way. But the base rate for this region is brutal. If you bought the Southeast Asia tech IPO story over the last eight years, with one exception, you lost money. What actually breaks the curse Here is the thing that matters. Almost every one of those shipwrecks went public unprofitable, floated at the very top of the cheap-money window on a growth-at-all-costs story. GCash is not that. GCash actually makes money. That is the one real thing that could break the curse. The curse was never the business. The risk is the entry price. GCash is reportedly chasing a valuation around eight to nine billion dollars, against roughly five billion in the private market just a couple of years ago. That is the exact same “premium to the last round” framing that came right before every name on the shipwreck list. History says it is not company quality that determines whether public investors win. It is the price on the day they are let in. Buy low, sell high. If Mint prices for perfection at the top of the range, the regional base rate says the valuation compresses toward fundamentals first and compounds later, if you are patient. Converge, down 40%, is what impatience looks like. Who holds the pen Here is the thread that ties the week together. This was the week Southeast Asia’s public markets stopped pretending to be a pure growth story and started behaving like state-managed plumbing. A fintech bends an exchange’s rules to get listed. A country leans on its sovereign fund to keep its emerging-market badge. And underneath all of it, the smartest new money in the world, MIT and Pfizer, is quietly buying into private companies at the early stage, where the value actually gets made, long before any of this public-market theater begins. Notice where the sophisticated capital is putting its chips. Not into the IPO. Into the cap table, years earlier. So my filter for all of it, and yours, should be the same question: who actually holds the pen here? Who decides what gets built, what gets listed, what gets propped up? More and more in this region, the answer is governments and sovereign funds, not founders and not public investors. If you are a founder who is not a conglomerate heir or a sovereign-fund favourite, that should tell you exactly where to aim, and exactly who to raise from. That is the week. If it was useful, the most useful thing you can do is send it to one founder who is about to get excited about an IPO. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit seaofstartups.substack.com

  5. Jun 24

    The Mirage and the Fork in the Road

    Start with two numbers and a question. In May, startups in this region raised $472 million. More than double what they raised in April. Read only that line and you would think the drought had broken. Now the second number. That doubling was built almost entirely on two checks. Take those two out and May was thin, still down on the year before. So here is the question I want to sit inside. When you are a founder in Kuala Lumpur, or Bangkok, or Manila, which numbers are actually telling you the truth? Because two of the loudest numbers in this market, the funding headline when you raise and the IPO pipeline when you want out, are both unreliable. And they are unreliable in different ways. The money coming in is inflated. The money going out is uneven. In between sits a real company, your company, trying to make decisions on top of figures that flatter and figures that lie. The mirage: headlines that flatter The funding rebound is a perfect little lie. Not a dishonest one. A statistically true one, which is worse, because it is harder to argue with. May 2026: $472 million across 31 deals, per DealStreetAsia. Up 104% on April. The kind of line that gets screenshotted into a pitch deck by Tuesday. Look underneath it. The jump came from the return of mega deals, transactions worth $100 million or more. A data center. An AI hardware platform. April had none. May had two. Two checks did the heavy lifting for an entire region. And even with them, May still came in 18% below the same month a year earlier. Strip the two big ones out and what you have left is quiet. This is not new, and that is the point. We saw the same shape in the first quarter: about $2.8 billion across 98 deals, the lowest deal count in at least eight years, with a single data center raise accounting for more than 70% of all that capital. Once you see the pattern you cannot unsee it. The total goes up. The number of companies actually getting funded does not. The aggregate is being inflated by hardware and data centers, while the count of real operating companies catching a check stays flat. Here is why that matters to you, and it is not academic. If you are raising right now and you benchmark yourself against the headline, you will conclude that capital is flowing and you are simply being passed over. That is the wrong lesson, and it will make you do desperate things. The right lesson is that the deal count, not the dollar total, is the honest gauge. And the deal count says fewer companies, higher bar, slower checks. The honest number is in the margin So if the aggregate is a mirage, what is the real one? What is the number on a Southeast Asian cap table that does not lie? It is the margin. Which brings me to one of the genuinely good stories in the region this month. Respond.io, a Malaysia-based company, raised a $62.5 million Series B led by Camber Partners, with Endeavor Catalyst and existing backers coming back in, off the back of going through the Endeavor selection network. Big round. But the round is not the story. The story is what was true before the round. $35 million in annual recurring revenue. Growing over 100% a year. At a decent profit margin. Read that again, because they were already profitable. They raised growth money from a position where they did not strictly need it. That is the exact opposite of the burn-first, find-the-model-later playbook the last cycle rewarded and then punished. They run an AI-agent-powered customer messaging platform, the layer that lets a business actually hold a conversation and close a sale across the channels where commerce in this region happens. Billions of messages a quarter, more than 10,000 businesses, over 180 countries. The new money is going west, into North America and Europe, with the possibility of some acquisitions. A profitable company, quietly compounding, raising on its own terms and going on offense into the biggest markets in the world. Take one thing from this. Stop reading the league tables. Read the profit and loss. In 2026, the only honest number on a Southeast Asian cap table is the margin, because it is the one figure nobody can dress up with a single big check. The asterisk Malaysia should be honest about Let me complicate my own happy story, because I am not here to wave the flag. This one is close to home, and KL should be proud of it. The founder is not Malaysian. The company did not start here. It was brought here. That should be a feature, not a footnote. A founder who could base anywhere chose to base in KL, and that decision creates things you can touch: engineering jobs, payroll that gets taxed, corporate tax, office leases, local lawyers and accountants, the cafe downstairs, and a signal to the next founder weighing where to land that says people build serious companies here. Malaysia should bank that credit fully and without an asterisk. But the timing is almost too on the nose, because there is an asterisk. At the same moment, the rules on foreign talent are leaning the other way. The salary floor on the employment pass has jumped. Pass lifespans are changing. To me, though, the salary number is not the headline. The harder one is the requirement that you have a replacement plan in place for foreign talent, and some of those plans are short. Detail has been scant, but one person closer to the interpretation told me the employment is treated as tied to the company, not to the title or the role. So if you bring in a foreign hire to fill, say, a junior developer seat, and that person does well and gets promoted, it does not matter that their title has grown. What matters is that they are still there, and the requirement is that you replace them so that they no longer are. Sit with that from the talent’s side. What highly capable person takes a role knowing there is a clock on it? If they have a family, will they uproot to a market that is effectively saying we want you temporarily but not forever? I understand the intent. We do need to build local capability, and you should not let companies park expats in seats indefinitely. Fair enough. But here is the tension I cannot get past as an investor. You cannot run a “come build your global company here” pitch and a “here is your countdown timer, please train your replacement” policy at the same time. The open-door version of this works. There are countries we can point to that prove it. This is a competitive sport. The founder who chooses KL had other options, because Singapore wanted him, Hong Kong wanted him, Tokyo, Bangkok and Manila all wanted him. The risk is that Malaysia celebrates this win in the very quarter it makes the next one harder to land. If attracting mobile founders is how a small market punches above its weight, and it is, then the policy and the pitch have to point in the same direction. For this month at least, they did not. The fork in the road Now the way out. Every founder eventually asks the quiet question. If this works, how do I get out, and where? Every investor asks it less quietly. In Southeast Asia the answer used to be a shrug. This month, three companies gave three different answers, and together they tell you more about this region than any funding total. Thailand sends its champion abroad. LINE MAN Wongnai, the app more than 10 million Thais use for food, rides and payments, is weighing an IPO, and the venues it is looking at are Hong Kong and New York, not Bangkok. The reporting cites weak domestic conditions and political volatility, with a decision expected as soon as the end of this month. Sit with that. The most-used app in the country looked at its home exchange and decided it could not get a fair hearing there, so it is shopping for a listing 8,000 kilometers away. A market that cannot list its own champions does not have a sentiment problem. It has a plumbing problem. The pipes that turn a great company into a liquid, locally owned public outcome simply have not been built. The Philippines builds a house worth staying in. In the same window, the opposite answer. Mint, the parent of GCash, the finance super app tens of millions of Filipinos live inside, has authorized the filing to go public: a registration with the regulator, a listing application with the Philippine Stock Exchange, an offer of around 12% of the company, targeting the second half of this year and possibly the fourth quarter. It is shaping up to be the largest IPO in the history of that exchange. And it is listing at home. Not Hong Kong. Not New York. The biggest fintech outcome the country has produced is choosing to be a Philippine public company. It is not alone. Maya, the digital bank, is weighing its own listing on a dual track, the local exchange plus NASDAQ, after its first profitable year. One foot at home, one foot abroad, a hedge. Look at the fork honestly. Thailand’s champion is leaving the list. The Philippines has one champion committing to the home exchange outright and another hedging across both. That is not the region as a single sound story. That is the region splitting in real time over the same question: is it worth building a venue people want to stay for? Right now, this quarter, the Philippines is making the bigger bet that the answer is yes. The caveat, because I promised it. Do not let anyone sell you Mint and Maya as a scrappy-startup miracle. Mint sits behind Globe and the Ayala group, with AMP alongside. Maya sits behind PLDT. These are conglomerate and telco children going public, which rhymes with what I said recently about Vietnam, where the giants raise and the startups starve. Hold both thoughts. The optimism is earned: a deep local public market is the single thing this region has always lacked, and the Philippines is genuinely building toward it. But the homegrown-founder fairy tale is not the right frame. Incumbents are listing. That is still good. It is just not the legend. And here is the constructive next move, the one I would want a Filipino policymaker or oper

    The Mirage and the Fork in the Road

  6. Jun 10

    Ep 31 - Oil, iron, and idle money: what the war is really doing to Southeast Asia

    Start with a sliver of water between Iran and Oman. On a normal day, roughly a fifth of the world’s oil moves through the Strait of Hormuz. This year it stopped being normal. When the strait seized up, Brent jumped 10 to 13 percent in a single session into the low 80s and kept climbing to the highest level since 2022. The International Energy Agency, which does not deal in drama, called it the largest supply disruption in the history of the global oil market. I am not here to cover the politics. I am here to follow the power. Because that one shock shows up three times in the Southeast Asian startup story this quarter, wearing three different costumes. It sold electric cars. It raised the price of the electricity our data centre boom depends on. And it gave every cautious LP one more reason to keep the chequebook shut. Energy, iron, and idle capital. Follow the power and you follow the whole region. One. The war that sold a million electric cars The lazy version of this story is “war happened, everyone bought an EV.” That is not what happened. What happened is that a fuel shock landed on top of a shift that was already moving fast, and poured petrol, pun intended, on the fire. The scale first. In 2025, EV sales in Southeast Asia more than doubled year on year to more than half a million vehicles, and more than 90 percent of those were full battery electric, not hybrids. The demand was already there. Then the petrol queues showed up. One Thai market report described long lines at filling stations on the same days that EV displays pulled the biggest crowds at the Bangkok motor show. That is the whole story in one image. One queue for the old thing, one crowd for the new one. Go around the region and the averages hide the real story. Vietnam is the outlier nobody outside Asia talks about: EV share of new cars hit close to 40 percent in 2025, ahead of the UK and the EU, almost entirely on the back of one company, VinFast, which targets 300,000 deliveries this year after 175,000 last. Thailand is the cleanest fuel link, with EV sales tripling year on year to over 44,000 units in January 2026 alone, and logistics fleets switching specifically to cut their exposure to fuel cost swings. When the fleet operators move, it is about the spreadsheet, not the planet. Indonesia crossed 15 percent EV share and passed the United States, with Chinese brands taking more than 75 percent of the market. This is not a Western EV story. It is a Chinese supply story with a Southeast Asian buyer. And Malaysia, my home market, is earlier and more honest: adoption up 14-fold since 2022, but still only about 5.5 percent of cars sold, held back by roughly 5,000 public charge points. You cannot fuel-shock your way past missing infrastructure. None of this is just consumers being noble. It is policy and cheap money. Thailand cut excise on passenger EVs from 8 percent to 2, and to zero on electric pickups. The Philippines went further, putting forward an incentive package worth around 60 billion pesos while ending subsidies for combustion engines, with the reporting tying the move directly to the oil shock. Read that again: a government using an oil crisis as cover to stop subsidising petrol and start subsidising electrons. Then the banks did the quiet part. In Singapore, UOB ran a green car loan at 1.5 percent, DBS at 2.48. When a bank prices your electric car loan below your petrol one, the moral argument is over. The maths makes the decision. The part that matters for operators is the fleet. Grab signed with BYD to put up to 50,000 EVs into its fleets across the region, with an eco-friendly toggle in Singapore and Thailand. GoTo took the other lane, going after two wheelers with a pledge to electrify Gojek’s motorbike fleet by 2030. On autonomy, be honest: the robotaxi headlines are a US and China story. Out here the fundable shift is the powertrain under the existing driver, not removing the driver. If you are pitching autonomous ride-hailing for Southeast Asia this year, the oil shock did not help you. The EV swap did. Here is where I land, and it is not the clean version. The war did not invent this boom. China did, with cheap good cars and a supply chain nobody here can match, and governments did, with subsidies written before anyone fired a missile. The shock just compressed years of slow behaviour change into a few quarters. And demand pulled forward by a price spike can snap back. If Hormuz reopens and Brent drifts back to the 60s, some of this 2026 surge was borrowed from 2027 and 2028. The companies that survive that are the ones building real local supply, financing, and charging, not the ones riding a fear premium. Two. Twenty billion lands in Johor, and DayOne raises four and a half We have covered the Malaysian data centre build before, so I will not reread the brochure. I want to follow the money one step further than the headlines do. Announced data centre capex across the region now runs past 20 billion US dollars over the 2024 to 2028 window, and that is committed, not deployed. AWS around 9 billion into Singapore, Google 5 billion plus 2 for its first Malaysian site, Microsoft a couple of billion more into Malaysia and Indonesia. On top of that, private money: AirTrunk alone is putting 12 billion ringgit into two new Johor campuses, taking its Malaysian commitment to roughly 27 billion ringgit, call it 7 billion dollars. And just this month DayOne, the Singapore-domiciled operator that flipped out of China’s GDS, closed a 4.5 billion dollar Series C led by Coatue and Hillhouse with Indonesia’s sovereign fund alongside. Hold that name, because it comes back in the third act. Now the question nobody asks: what is that money actually buying? Land, concrete, power, cooling, and imported chips. A hyperscale data centre is a real estate and energy project wearing an AI t-shirt. The single biggest cheque inside it goes to Nvidia. Very little of that 20 billion touches a local software founder. This is not venture capital landing in the region, it is construction capital. So what is the secondary effect on the rest of us? Three things, and I want to be balanced. First, cost. These campuses pull on the same grid and water local businesses use, and Malaysia stopped approving non-AI data centre proposals back in 2024 to keep the power for AI builds. The state is rationing power and choosing hyperscalers. When your tariff drifts up in three years, this is part of why. Other parts of the world now require operators to reinvest into the local energy and water network to offset that pressure. I have not seen that proposed seriously in Malaysia yet, and I would like to. Second, jobs. A hyperscale campus employs a crowd for eighteen months of construction, then a skeleton crew. It is not a founder-jobs engine. Third, and this is the genuine prize: if the build is done right, founders get cheaper, closer compute and local data residency, the thing that lets a regulated fintech or health startup build on sovereign infrastructure without stitching together a compliance workaround. The roads analogy is the honest one. Infrastructure is an enabler, not the destination. The data centre boom only pays off for the domestic economy if we generate the demand to use it: enterprises and government going properly digital, and a real layer of AI-native startups creating the load these campuses were built for. Lay the road, then you still need the trucks. Capital keeps flooding the iron. Whether it earns its return depends entirely on who drives on it. Three. The lowest deal count in eight years, sitting on a mountain of cash Two facts that should not be true at once. In the first quarter of 2026, Southeast Asian startups raised about 2.8 billion dollars across 98 equity deals, the lowest quarterly deal count in at least eight years, and even that is flattered by one or two giant infrastructure cheques of the DayOne variety. Meanwhile APAC investors sit on roughly 240 billion dollars of dry powder, down from a 2023 peak near 315 but hardly an empty tank. So which is it, drought or hoard? Both, and the contradiction is the story. The money exists. It is just not moving into Southeast Asian early stage. The last clean read on region-specific dry powder was around 7 billion dollars, a couple of years old and probably overstated, but the direction is the point: funding here fell about 70 percent from the 2021 peak while the cash pile barely moved. That is not a region that ran out of money. That is a region whose investors went on strike. Where did the new money go instead? Peak XV, the old Sequoia India and Southeast Asia team, closed 1.3 billion late last year, labelled India Seed, India Venture, and APAC. India now runs hundreds of active early-stage funds and has climbed from roughly 9 percent of APAC capital markets volume toward 20. The APAC money is concentrating into India for growth and Japan for buyouts, not Southeast Asian seed. So when a Singapore GP tells you the market is tough, hear it precisely. It is not that Asia has no money. It is that the money is choosing India’s depth and Japan’s stability over our fragmentation and our weak record in the asset class. Capital is being selective, and Southeast Asia is the one being un-selected. Then layer the war back on. In March the reporting was blunt that the Iran conflict threatened to deepen Asia’s worst private equity fundraising slump in a decade. An oil shock spikes uncertainty, and uncertainty is the enemy of a new fund commitment. The same barrel of oil that sold an electric car in Bangkok made a pension fund in the West, and a high-net-worth backer here, think twice about a new Southeast Asian VC. Cash gets more cautious exactly when founders need it to get braver. So do not buy the clean drought story, and do not buy the clean abundance story either. The honest version: the tank is full, the driver is scared, and the road out, meaning exits, still looks rough. 98 deal

    Ep 31 - Oil, iron, and idle money: what the war is really doing to Southeast Asia

  7. Jun 3

    Ep. 30 - We Called It a Funding Winter. I Think We Built for an Exit That Was Never There.

    Singapore just released its report on venture funding for 2025, and almost every write-up reads the same way. Funding winter. Capital’s gone quiet. Hold the line, it’ll come back. I think that’s the wrong story. I’ve been sitting with these numbers for a few days, and the more I look at them, the more I’m convinced we’ve been telling ourselves the comfortable version. The comfortable version is that the money left and the money will return. The harder version, the one I actually believe, is that the region made a strategy bet a decade ago, the bet didn’t have an exit attached to it, and 2025 is just the year the math stopped hiding. We’ve had a few of these years where the math stops hiding. This is another one. So let me do a bit more opining than usual. This one’s a little spicy. * * * The number everyone read The headline is genuinely rough. In 2025, Singapore recorded 472 venture deals, down 35 percent from the year before. Total capital raised came in at 4.6 billion US dollars, down 34 percent year on year. And Singapore is the strong one. Across the ASEAN-6, both deal value and deal volume hit a four-year low. Now hold that next to the United States in the same year. Silicon Valley deal value nearly doubled, to around 160 billion dollars. A lot of that was two rounds: OpenAI at 40 billion, Anthropic at 15 billion. Two companies, in one country, raised more than ten times what the entire island of Singapore raised across 472 deals all year. The easy conclusion is that capital is concentrating into American AI and starving everyone else. That’s true as far as it goes. There’s real gravity pulling allocators toward the bleeding edge, and that gravity sits in Silicon Valley. But that’s a description of the weather. It doesn’t tell you why our house is the one with the leak. For that, you have to go back further than last year, and look at what we actually spent the money on, and what we expected to get out the other side. * * * The bet we made Here’s the part that doesn’t get said enough. For most of the last decade, Southeast Asia poured its venture money into consumer. Ride-hailing, e-commerce, food delivery, the super-app. The big, beautiful, blitzscaled consumer story where you capture a young, mobile-first population of 700 million and become the thing they open twenty times a day. I’m not mocking it. I lived through the optimism. Grab, GoTo, Sea, Lazada, Shopee. These companies built the rails the whole region runs on now. Digital payments are everywhere because of them. That’s real, and it was needed. Consumer is the precedent layer. Most maturing markets start there, build the rails, then transition. That part is natural. But look at the allocation. In 2023, more than a third of Southeast Asian venture deal value went into consumer. The honest caveat is that “consumer” is a fuzzy line, depending on whether you fold in consumer fintech, so treat the exact figure loosely. Even on the conservative read, you land somewhere north of thirty percent. Run the same count in the US that year and you’re in single digits. The number I keep landing on is around three and a half percent. Read that again. We put an order of magnitude more of our capital into consumer than the most mature venture market on earth did. And we weren’t growing out of it. We were accelerating into it. Consumer’s share of regional deal value kept climbing while software’s share fell. So while the US was doing the boring, durable thing, funding enterprise software and infrastructure, we were doubling down on the consumer copycat play right as the cheap money drained out. Why does that matter? Because of what happens at the end. * * * The door that was never there Every venture dollar is a bet on an exit. Money goes in, and somewhere down the line it has to come out bigger, through a sale or a listing. No exit, no returns. No returns, no next fund. So how did the region do on exits? Here’s the number that should be tattooed on every term sheet. Since 2015, the entire Southeast Asian venture market generated roughly 70 billion dollars in exit value. Sounds fine until you look underneath. More than 55 billion of that came from three exits, all in 2021. Stretch it out and nearly 87 percent of all exit value since 2015 came from six companies. Take it to the top twenty and you’re at 96 percent. Yes, there’s always a power law. Concentration is normal. But strip out a handful of unicorns and the regional market has returned almost nothing to almost everyone. The investment-to-exit ratio has run consistently above twenty to one. Twenty dollars in for every dollar that found its way out. It’s been a trap. The Hotel California of venture. You can check in, but you can never leave. And here’s the part that connects the dots. The few giant exits we did get didn’t happen here. Grab went out via a SPAC on the Nasdaq. Sea listed on the New York Stock Exchange. They had to leave to get out. The Singapore Exchange, the biggest in the region, ranks only ninth by market value in Asia-Pacific, and several regional exchanges still carry listing rules strict enough to keep a cash-burning consumer company out entirely. For a blitzscaled consumer business, the local IPO was a closed door. So put it together. We funded consumer companies built on the growth-at-all-costs playbook, and that playbook only pays off through a big public listing. We never built the public markets to list them on. We built companies for a door that, at home, was never there. That’s not a winter. Winter ends. This was a design flaw. * * * Consumer is the hardest thing to sell, everywhere This is the part I want founders and investors to chew on, because it goes beyond us. Consumer is one of the hardest categories to exit anywhere in the world. Think about who actually buys companies. In enterprise software there’s a deep, permanent bench of buyers who do this all day. 2025 was the most active year on record for software M&A, with strategic buyers alone accounting for around 42 percent of deals. The most active software acquirers in 2024 included IBM, Cisco, Autodesk, Nvidia. There were 22 firms that each made at least five acquisitions in a single year. That’s a machine. A standing market of people whose job is to buy companies. What are they buying for? Recurring revenue, mission-critical, sticky, hard to rip out. Now ask who the standing buyer is for a regional food-delivery app, or who’s lining up to roll up consumer brands in a market where customers switch the second someone else runs a discount. There isn’t a bench. Consumer internet leans almost entirely on the IPO. And we just covered what happened to that door. Let me be fair, because the honest version is more interesting than the cheap one. Enterprise exits aren’t easy either. Only about ten percent of companies tagged as software ever get acquired. IPOs are about six percent of software exits. The median software acquisition went for roughly three times revenue, not the eye-watering multiple people imagine. B2B is not a golden ticket. What enterprise has is a functioning market of repeat buyers. Consumer mostly has the IPO. It’s a difference in optionality, in how many doors are actually open. We bet the region on the category with the thinnest exit options, and didn’t build the one exit that category depends on until recently. If you wanted to design a liquidity crunch on purpose, that’s how you’d do it. * * * The people who built it are now saying it What makes this report worth reading past the headline is the back half, where they ran candid pieces from a row of the region’s investors. To their credit, the honesty is right there. Vishal Harnal at 500 Global names liquidity as the clearest challenge facing the region, pointing straight at underdeveloped exit markets and the long holding periods that wear founders and investors down. Angela Toy at Golden Gate is just as direct, conceding the region still lacks depth in both M&A and secondaries to get people their money out. The one that stuck with me is from Cyril at SOSV, who lays out the question every Singapore founder eventually asks out loud. If the place you ultimately have to go for capital, scale, and an exit is San Francisco, why not just start there on day one? Why build here at all? That’s a tough one to sit with. It’s not a critic on the sidelines. It’s a GP at an active global fund saying the quiet part into a government report. Then there’s Antler. They’ve raised about 1.5 billion dollars globally, from dozens of institutions and sovereign funds. The amount that came from Singapore institutions was around 10 million. The US allocates roughly five percent of its capital to venture as an asset class. Singapore sits well below one. So even the domestic money, the money that’s right here, mostly doesn’t back the local market. The capital sits in the city. It just doesn’t believe in the thing the city keeps saying it wants to be. When this many people who built the market all point at the same missing piece, it stops being a complaint and starts being a diagnosis. * * * So what do we actually do To be clear, Singapore isn’t sitting still. The response is substantial: an extra billion dollars into Startup SG Equity for growth-stage companies, a new 1.5 billion dollar anchor fund aimed squarely at strengthening exits, and a Singapore Exchange and Nasdaq partnership we’ve talked about here before. Almost all of it is about building the exit door now, after a decade-plus of funding companies that needed it and didn’t have it. I’m not saying that to dunk on the policy. The policy is correct. Real liquidity, a working M&A culture, a credible place to list, that is exactly the right thing to spend on. My point is that we’re building the staircase after everyone already jumped. The companies that needed this in 2018, 2021, 2023 are gone or got out somewhere else. T

    Ep. 30 - We Called It a Funding Winter. I Think We Built for an Exit That Was Never There.

  8. May 27

    EP 29 - Chatbots to Agents and where Liability Lands

    I hopped into a taxi in Bangkok last week and the driver, a man north of fifty, spent the ride telling me what he was building with AI. Not complaining about the economy. Not asking where I was from. Telling me about his project. I’ve been turning that over ever since, because it isn’t an isolated thing. For weeks now I’ve been scanning event listings in whatever city I land in, and the pattern is hard to miss. It isn’t pitch nights anymore. It isn’t another fireside with a fund manager. It’s vibe coding meetups, agentic AI sessions, AI trainings. Paid attendance, no walk-ins, speakers who’ve shipped real apps. KL has them. Singapore has them. Bangkok and Manila have them. Go on Lu.ma or Eventbrite right now and there’s probably one happening in your city this week, maybe two. I know this firsthand because I run some of them. I host AI salon events in Bangkok, and I’ve watched the rooms change. So while the rest of the startup world argues about whether funding is back, looking at numbers that are frankly pretty dismal, there’s this whole other thing happening in cafes and malls across Southeast Asia. Regular people are learning to build software by talking to a machine. Why I trust this one I dismiss most AI hype on reflex. My feed is littered with slop, articles that read like they were generated by the thing they’re describing, people calling everything the future. I scroll past it. This is different, and the reason is simple. People are paying to show up. And it’s a different crowd than I’m used to seeing at startup events. University students and fresh grads who can see the job market tightening and are choosing to get ahead of the curve instead of waiting it out. Founders who can’t afford a dev team. Marketers. People with an idea and no technical co-founder, who a year ago would have been stuck with that idea trapped in their head, never seeing daylight. This is the no-code, low-code movement, upgraded and supercharged into the current AI era. The category has a name now: “vibe coding”. I’m not a fan of the term, all that talk of vibes and feel grates on me, but it’s the vernacular, so I’ll use it. You describe what you want in plain language and the AI writes the code. That’s the whole thing. I do it myself. I’ve used AI coding to replace most of our software stack. Thinking back to the friction of a couple of years ago versus how good this is now, and then projecting forward to how good it’ll be as the models keep improving, is genuinely one of the more interesting arcs I’ve lived through as an operator. From apps to agents, which is where it gets serious Building an app is one thing. The next rung up the ladder is building an agent, and agents are a different animal. Most people, once you get out of the tech bubble, still picture a chatbot. You type, it types back. You ask, it answers. A better Google. That’s generation. It makes text, images, words. An agent acts. It doesn’t tell you how to clear your inbox, it clears your inbox. It books the meeting. It sends the email. It runs commands on your machine. It talks to other software and gets things done with barely any input from you. That’s the entire ballgame for risk. A chatbot needs a human to type every prompt. Every harm one causes still started with a person asking for it. An agent can plan, decide, and act on its own initiative. It can cause harm nobody asked for. I want to be clear that I’m bullish on this. Hugely. But being bullish and being measured aren’t opposites, and the risk side of this deserves honest airtime. Two examples everyone in the open-source world is talking about. The first is the lobster: OpenClaw. It went viral the moment it dropped. It connects an AI model to your messaging apps and acts on your behalf, books things, browses, runs commands, manages your house. People pulled their old Mac minis out of drawers to run it. Apple caught the wave and nudged the price up. It is not a Southeast Asian product, and we should be honest about that. It went viral hardest in China, which has been well ahead on the open-source movement. Southeast Asia needs to kick into gear as a fast follower, even when we’re not the origin. The second is Hermes, out of a US research lab a few months back. What makes it different is memory. It lives on your own server, runs all the time, and gets better the longer you use it. It remembers what you told it last Tuesday. It writes down how it solved a problem so it never starts from scratch again. By this month it was the most-used agent out there by some measures, hundreds of billions of requests a day, hundreds of thousands of developers piling in within three months. Here’s the part that should make you pause. Three separate security audits this year found malicious code hiding in the add-on skills people share for these agents. Think about what that means. An autonomous thing, running constantly, on your own machine, with access to your messages and files and maybe your ability to spend money, pulling new abilities from a community marketplace that’s already been found to contain things designed to hurt you. That isn’t a future problem. It’s a this-year problem, and it’s happening on hardware people own, in their homes, outside any IT department or compliance check. A friend who’s far sharper than me on this put it well. Permissioning an agent is like onboarding a new intern. You give them enough access to act, but not enough to break things. If humans are entities of action, we have to treat agents as entities of action too, with the same scoping and the same limits. The catch is that getting that right still takes real technical skill, and most of the people downloading the lobster don’t have it. So who’s writing the rules Surely someone’s regulating this. Here’s where it actually stands, and the answer is more interesting than “nobody is.” Three big global players, three different postures. The US is actively deregulating to keep its lead, tearing up the old safety rules and trying to stop its own states from making their own. The posture is get out of the way, though there was an executive order floated recently that would have made new models notify the government before public release, something closer to how the FDA approves a drug. It got paused, not signed. We’ll see. Europe, true to reputation, has the most serious regime, and just this month agreed to delay the hardest parts, the high-risk rules, by over a year. Competitiveness pressure. So even the strictest regulator in the world is loosening its grip right as agents arrive. And China is the strictest in practice and the only one already acting on agents specifically, real enforcement, thousands of non-compliant services shut down. Telling, the country where everyone installed the lobster also told its own government agencies and state banks not to put it on work devices. The adoption champion got nervous about its own craze. Even the deregulating US quietly started building standards for autonomous agents. So nobody actually thinks this is fine. Everyone sees the gap. They’re just moving at wildly different speeds. Southeast Asia is that same story compressed into one region, running at three speeds. Vietnam, maybe not who you’d guess, has the only real binding AI law here, passed late last year, enforced since Q1, risk-based with actual prohibited uses. It tracks, given how much of the region’s developer talent sits there. Singapore did something very Singapore: the world’s first governance framework built specifically for agentic AI, detailed and thoughtful, and deliberately voluntary. No teeth. The bet is give industry sophisticated guidance, remind everyone they’re still liable when their agent screws up, and keep the innovation onshore. They’ve already refreshed it with case studies from the likes of OCBC, Tencent and Workday. A living document, which is the right call given the pace. And then Malaysia, where I’m based, sitting on one of the most aggressive agent rollouts in the region, with its actual rules still in draft. Not here yet. Here’s the whole thing in one line. Everyone, globally and right here at home, is regulating the last war. The last war was chatbots generating bad content, the stuff you can ban after it spreads. We saw it when Indonesia, Malaysia and the Philippines banned Grok over deepfakes, including images of children. Three countries, fast, coordinated, and fully deserved. But that’s the model: react after the harm, fold quickly. And every one of those images still needed a human to type the prompt. The next war is agents taking bad actions on their own, because the black box decided that was the thing to do. That war is already shipping. Through anonymous downloads, onto personal machines, learned at meetups across the region, in a place where exactly one country has even a voluntary framework and the country with the biggest rollout is still drafting. We’re banning the thing that needs a human to ask. We haven’t started on the thing that doesn’t. What I keep coming back to I’ll be honest, I don’t have a clean answer. Part of why I raised this is that it was a quiet news week. But the bigger part is that I can’t stop noticing the trend, and I doubt I’m alone. If you’re a CISO or a CTO or sitting in a compliance function, you’re already living this, because the whole enterprise is integrating more automation and more agents by the month, and the risk side is going to drag a regulatory environment into the room whether we invite it or not. It always does, the moment a technology touches enough of society. So it’s worth thinking now about what that reaction is likely to look like, instead of being surprised by it. But I keep coming back to those meetups. To the rooms full of people building. Because that’s the real story, and it isn’t happening in a lab or a boardroom. It’s happe

    EP 29 - Chatbots to Agents and where Liability Lands
See All (82)

Ratings & Reviews

About

Real, raw, relatable takes on Southeast Asian startups. One investor, the week's news, no script. seaofstartups.substack.com

Information

  • Creator
    Decoding the Pulse of Founders, Capital & Conviction in Southeast Asia.
  • Years Active
    2022 - 2026
  • Episodes
    82
  • Rating
    Clean
  • Copyright
    © Sea of Startups
  • Show Website
    SEA of Startups