For a decade, the Southeast Asian startup script barely changed. Find two or three co-founders, raise a seed round, hire a dozen people, then sprint to the next round before the money runs out. The script assumed a steady supply of early-stage capital. In 2026, that assumption looks naive.
A new report from Gobi Partners, Solo Horizons: Redefining Scale for the Philippine AI Economy, argues the fix may be to need far less capital in the first place. Its protagonist is the AI-native one-person company (OPC): a single founder, few or no employees, and a stack of AI agents doing work that once needed separate engineering, marketing and support hires.
Also Read: One-person company: Why Southeast Asia’s next unicorn may look nothing like Grab
It is a seductive idea, and it lands at a convenient moment for a region where seed money has dried up outside Singapore. Read closely, though, the report also shows how far the Philippines is from turning a clever legal wrapper into a genuine venture pipeline.
The money went to Singapore, and to Series BThe capital map is brutally lopsided. Google, Temasek and Bain & Company put regional digital funding at roughly US$8 billion a year, tilted towards late-stage deals. By Tracxn’s count, Singapore absorbed 94 per cent of disclosed Southeast Asian startup funding in the first nine months of 2026, a concentration that has all but ended the idea of a single regional funding story.
The Philippines sits on the wrong side of that divide. Taguig, the Metro Manila business district, raised under US$130 million. National deal count is down 8.1 per cent year to date, and deal value has fallen 64.5 per cent. Series A value has collapsed by 98 per cent, while Series B jumped 795 per cent to take about three-quarters of all capital deployed.
Sector concentration is just as stark. Fintech incumbents took 78 per cent of 2026 deal value, unsurprising in a market where compliance let fintech scale fast. Disclosed funding for AI-native startups was under one per cent.
GCash operator Mynt’s record listing raised ₱60.9 billion (about US$960 million) in October, proving exits exist for giants. The country’s climb to 63rd in StartupBlink’s global ecosystem index is welcome too. Neither pays a first-time founder’s cloud bill.
Why the maths of starting up has flipped
What has changed is the cost side. Stanford HAI’s 2025 AI Index found the price of running a GPT-3.5-level model fell from US$20 to US$0.07 per million tokens, a 99.6 per cent drop. Its 2026 edition reports AI agents now complete 66 per cent of benchmark computer tasks, against a human baseline of 72 per cent.
Benchmarks are not payroll. Anyone who has watched regional tech leaders keep agents away from production knows the gap between a test score and a client deliverable. Still, Gobi’s central distinction holds: AI lowers the capital needed to start a company, not the cost of venture capital itself.
The report’s comparison puts pre-revenue capital for a traditional seed-stage startup at US$250,000 to US$500,000, against US$5,000 to US$20,000 for an AI-native OPC. A founder who can test a product for a few thousand dollars learns whether customers will pay before asking anyone for a term sheet, though the hidden economics of autonomous agents mean token bills grow with usage too.
Also Read: Southeast Asia’s funding winter is real. Its building winter isn’t
Globally, solo founding is no longer fringe. Carta data analysed by Peter Walker shows solo founders started 36 per cent of new companies on the platform in 2025, double the 18 per cent of 2016, even if only about a fifth of solo-founded ventures historically raise institutional money. The poster child is Base44, the AI app builder its solo founder sold to Wix in June 2025 for about US$80 million upfront plus earn-outs, roughly six months after launch.
Manila already owns the legal wrapper
China has turned one-person firms into industrial policy. Suzhou wants 30 “OPC communities” and 1,000 one-person enterprises by 2028. Shanghai’s Pudong district covers up to RMB300,000 (about US$42,000) of computing costs for early AI startups, while Wuhan and Shenzhen’s Luohu district offer credit lines and partial loss cover to more than 50 solo AI ventures.
The Philippines has something cheaper: a law. The Revised Corporation Code of 2019 created the One Person Corporation, letting an individual incorporate with limited liability, no board and no co-incorporators. OPC filings rose 27 per cent in 2024 to 8,640, or 17 per cent of all 52,304 new corporations.
Here the optimism needs a footnote. Eighty-six per cent of those OPCs are traditional service businesses. The container exists; the AI-native companies to fill it, largely, do not yet.
The talent argument is stronger. The IT-BPM sector employs 1.89 million people, according to IBPAP, many already running workflows for foreign clients. Gobi calculates that if just one per cent launched service-focused OPCs, the country would gain about 19,000 firms, more than the 14,000 startups Tracxn tracks nationwide.
That maths carries an uncomfortable irony. The agents that let a call-centre team lead become a founder are the same ones most likely to automate her colleagues’ jobs. For many BPO workers, an OPC may be less an entrepreneurial leap than a lifeboat, much as the one-person AI business is already coming for the agency model.
A pipeline that still needs a Series A at the end
Gobi concedes the trade-offs. Agent-run firms create few payroll jobs, and their survival through a long downturn is untested. Thin teams also mean thin moats: a product one person built on off-the-shelf models can be cloned by another person with the same models.
So the report frames the OPC as a validation pipeline for venture, not an end state, and proposes four levers. First, startup incentives under the Innovative Startup Act (Republic Act No. 11337), the pending House Bill 6481 and CREATE MORE, including its 20 per cent corporate income tax rate, should cover OPCs from the day they file.
Second, the National Development Company’s Startup Venture Fund, which put ₱16.7 million (about US$290,000) into two startups in 2024, should become a private-led fund-of-funds modelled on Israel’s Yozma. Third, regulators should endorse a standard convertible note along the lines of Japan’s J-KISS to settle the legal grey zone around SAFEs.
Also Read: The AI wrapper reckoning has reached SEA’s funding tables
Fourth, the country should borrow from Estonia’s e-Residency, which supports more than 39,000 remotely run companies and earned €125 million (about US$145 million) in 2025 against €10 million in costs, and from India’s 2021 OPC reforms, with easy conversion into multi-shareholder companies.
The sensible reading is narrower than the hype. OPCs can slash the cost of proving an idea, but they cannot conjure the Series A cheques that vanished this year. If thousands of solo founders validate demand and then hit the same 98 per cent wall, the Philippines will simply have built a more efficient waiting room. The policy work is making sure there is a door at the other end.
The post One founder, many agents: The Philippines’s answer to a vanishing Series A appeared first on e27.
For a decade, the Southeast Asian startup script barely changed. Find two or three co-founders, raise a seed round, hire a dozen people, then sprint to the next round before the money runs out. The script assumed a steady supply of early-stage capital. In 2026, that assumption looks naive.
A new report from Gobi Partners, Solo Horizons: Redefining Scale for the Philippine AI Economy, argues the fix may be to need far less capital in the first place. Its protagonist is the AI-native one-person company (OPC): a single founder, few or no employees, and a stack of AI agents doing work that once needed separate engineering, marketing and support hires.
Also Read: One-person company: Why Southeast Asia’s next unicorn may look nothing like Grab
It is a seductive idea, and it lands at a convenient moment for a region where seed money has dried up outside Singapore. Read closely, though, the report also shows how far the Philippines is from turning a clever legal wrapper into a genuine venture pipeline.
The money went to Singapore, and to Series BThe capital map is brutally lopsided. Google, Temasek and Bain & Company put regional digital funding at roughly US$8 billion a year, tilted towards late-stage deals. By Tracxn’s count, Singapore absorbed 94 per cent of disclosed Southeast Asian startup funding in the first nine months of 2026, a concentration that has all but ended the idea of a single regional funding story.
The Philippines sits on the wrong side of that divide. Taguig, the Metro Manila business district, raised under US$130 million. National deal count is down 8.1 per cent year to date, and deal value has fallen 64.5 per cent. Series A value has collapsed by 98 per cent, while Series B jumped 795 per cent to take about three-quarters of all capital deployed.
Sector concentration is just as stark. Fintech incumbents took 78 per cent of 2026 deal value, unsurprising in a market where compliance let fintech scale fast. Disclosed funding for AI-native startups was under one per cent.
GCash operator Mynt’s record listing raised ₱60.9 billion (about US$960 million) in October, proving exits exist for giants. The country’s climb to 63rd in StartupBlink’s global ecosystem index is welcome too. Neither pays a first-time founder’s cloud bill.
Why the maths of starting up has flipped
What has changed is the cost side. Stanford HAI’s 2025 AI Index found the price of running a GPT-3.5-level model fell from US$20 to US$0.07 per million tokens, a 99.6 per cent drop. Its 2026 edition reports AI agents now complete 66 per cent of benchmark computer tasks, against a human baseline of 72 per cent.
Benchmarks are not payroll. Anyone who has watched regional tech leaders keep agents away from production knows the gap between a test score and a client deliverable. Still, Gobi’s central distinction holds: AI lowers the capital needed to start a company, not the cost of venture capital itself.
The report’s comparison puts pre-revenue capital for a traditional seed-stage startup at US$250,000 to US$500,000, against US$5,000 to US$20,000 for an AI-native OPC. A founder who can test a product for a few thousand dollars learns whether customers will pay before asking anyone for a term sheet, though the hidden economics of autonomous agents mean token bills grow with usage too.
Also Read: Southeast Asia’s funding winter is real. Its building winter isn’t
Globally, solo founding is no longer fringe. Carta data analysed by Peter Walker shows solo founders started 36 per cent of new companies on the platform in 2025, double the 18 per cent of 2016, even if only about a fifth of solo-founded ventures historically raise institutional money. The poster child is Base44, the AI app builder its solo founder sold to Wix in June 2025 for about US$80 million upfront plus earn-outs, roughly six months after launch.
Manila already owns the legal wrapper
China has turned one-person firms into industrial policy. Suzhou wants 30 “OPC communities” and 1,000 one-person enterprises by 2028. Shanghai’s Pudong district covers up to RMB300,000 (about US$42,000) of computing costs for early AI startups, while Wuhan and Shenzhen’s Luohu district offer credit lines and partial loss cover to more than 50 solo AI ventures.
The Philippines has something cheaper: a law. The Revised Corporation Code of 2019 created the One Person Corporation, letting an individual incorporate with limited liability, no board and no co-incorporators. OPC filings rose 27 per cent in 2024 to 8,640, or 17 per cent of all 52,304 new corporations.
Here the optimism needs a footnote. Eighty-six per cent of those OPCs are traditional service businesses. The container exists; the AI-native companies to fill it, largely, do not yet.
The talent argument is stronger. The IT-BPM sector employs 1.89 million people, according to IBPAP, many already running workflows for foreign clients. Gobi calculates that if just one per cent launched service-focused OPCs, the country would gain about 19,000 firms, more than the 14,000 startups Tracxn tracks nationwide.
That maths carries an uncomfortable irony. The agents that let a call-centre team lead become a founder are the same ones most likely to automate her colleagues’ jobs. For many BPO workers, an OPC may be less an entrepreneurial leap than a lifeboat, much as the one-person AI business is already coming for the agency model.
A pipeline that still needs a Series A at the end
Gobi concedes the trade-offs. Agent-run firms create few payroll jobs, and their survival through a long downturn is untested. Thin teams also mean thin moats: a product one person built on off-the-shelf models can be cloned by another person with the same models.
So the report frames the OPC as a validation pipeline for venture, not an end state, and proposes four levers. First, startup incentives under the Innovative Startup Act (Republic Act No. 11337), the pending House Bill 6481 and CREATE MORE, including its 20 per cent corporate income tax rate, should cover OPCs from the day they file.
Second, the National Development Company’s Startup Venture Fund, which put ₱16.7 million (about US$290,000) into two startups in 2024, should become a private-led fund-of-funds modelled on Israel’s Yozma. Third, regulators should endorse a standard convertible note along the lines of Japan’s J-KISS to settle the legal grey zone around SAFEs.
Also Read: The AI wrapper reckoning has reached SEA’s funding tables
Fourth, the country should borrow from Estonia’s e-Residency, which supports more than 39,000 remotely run companies and earned €125 million (about US$145 million) in 2025 against €10 million in costs, and from India’s 2021 OPC reforms, with easy conversion into multi-shareholder companies.
The sensible reading is narrower than the hype. OPCs can slash the cost of proving an idea, but they cannot conjure the Series A cheques that vanished this year. If thousands of solo founders validate demand and then hit the same 98 per cent wall, the Philippines will simply have built a more efficient waiting room. The policy work is making sure there is a door at the other end.
The post One founder, many agents: The Philippines’s answer to a vanishing Series A appeared first on e27.
投资人评论
聊判断，不堆黑话登录且账号审核通过后可评论。 立即登录