Southeast Asian startups are standing at a critical crossroads. While the region’s startup ecosystem continues to raise billions in 2024, a quiet revolution is reshaping how successful companies operate.

The winners of 2025 and 2026 won’t be those with the biggest teams or deepest pockets; they’ll be the startups that master the “One million+ revenue per person” model through intelligent automation.

This isn’t about replacing humans with machines. It’s about amplifying human potential to unprecedented levels, where each team member becomes a revenue-generating powerhouse capable of driving US$1 million or more in annual business results.

The AI acceleration imperative: Why “next few months” means now

The pace of AI development has reached a tipping point that makes 2019’s digital transformation look mild. According to McKinsey’s State of AI report, 71 per cent of organisations are now regularly using generative AI. For Southeast Asian startups, this creates both an opportunity and an existential threat.

The innovators in your fellow startup industry are going to leapfrog over you if you’re too slow and the window for competitive advantage is shrinking rapidly as AI tools become more accessible and powerful.

While some startups struggle with manual content creation, lead qualification and customer research, their AI-powered competitors are simultaneously running 50 personalised outreach campaigns, analysing competitor strategies in real-time, and nurturing hundreds of leads with human-like precision…all with a team half their size.

Pillars of AI-powered revenue generation

Smart Southeast Asian startups are building their competitive moats across 13 critical automation areas. Each represents a multiplier effect that transforms individual team members into revenue-generating machines.

Research and intelligence: The foundation layer

• Research of social media videos: Tools like Brandwatch and Sprout Social now use AI to analyse millions of video interactions, identifying viral patterns and audience sentiment in real-time. Any company can spot trending topics before competitors and understand what resonates with their target market without watching hours of content manually.
• Research of social media copy: Platforms like BuzzSumo and Socialbakers leverage natural language processing to analyse conversation patterns across social platforms.
  This allows startups to identify customer pain points, monitor brand mentions, and discover collaboration opportunities at scale.
• Outbound research: AI-powered tools like Clay and Apollo automate prospect research by aggregating data from multiple sources: LinkedIn, company websites, news articles, and social media, so as to build comprehensive prospect profiles in seconds rather than hours.

Content creation: Scaling your voice

• Scripting videos: AI scriptwriting tools like Jasper and Copy.ai can generate video scripts tailored to specific audiences and platforms. Combined with data from your research layer, you can create compelling narratives that speak directly to your prospects’ needs.
• Content writing: Beyond basic copywriting, advanced AI tools now understand brand voice, audience psychology, and conversion optimisation.
  They can produce blog posts, email sequences, and social media content that maintains consistency while adapting to different platforms and audiences.
• Video generation: Platforms like Synthesia and Pictory enable startups to create professional video content without expensive production teams.
  From product demos to personalised sales videos, AI handles the heavy lifting while they focus on strategy.
• Image generation: Tools like Midjourney and DALL-E 3 democratise visual content creation. Startups can generate custom graphics, social media visuals, and even product mockups that previously required specialised design teams.

Also Read: How to combat burnout and boost your productivity

Distribution and engagement: Maximising reach

• Distribution and engagement of content: AI scheduling tools like Later and Hootsuite Insights optimise posting times, hashtag selection, and cross-platform distribution.
  More importantly, they engage with audiences through intelligent commenting and response systems that maintain authentic brand voice.
• Outbound distribution: Advanced email and LinkedIn automation platforms like Outreach and Lemlist use AI to personalise messages at scale, optimise send times, and adapt messaging based on recipients’ observed behavior patterns.

Sales and conversion: Closing the loop

• Voice agents: AI voice technology has reached human-level quality. Tools like Dialpad Ai and Gong can handle initial prospect calls, qualify leads, and even conduct basic sales conversations, freeing human team members for high-value relationship building.
• Lead nurturing: Sophisticated marketing automation platforms like HubSpot and Marketo can now use predictive AI to determine the optimal nurturing sequence for each lead, automatically adjusting content and timing based on behavioral signals.
• Sales stack: AI-powered CRM systems like Salesforce Einstein and Pipedrive AI provide real-time coaching, predict deal closure probability and automate routine sales tasks, enabling each salesperson to manage and close significantly more opportunities.
• Automated newsletter: Tools like ConvertKit and Mailchimp use AI to optimise subject lines, personalise content, and determine optimal send frequencies for different subscriber segments, turning newsletters into powerful revenue drivers.

The competitive intelligence advantage

Perhaps the most overlooked opportunity lies in AI-powered competitive intelligence. While most startups sporadically check competitor websites, AI-enabled companies are monitoring:

• Pricing changes: Tools like Price2Spy track competitor pricing in real-time
• Content strategy: AI analyses competitor content performance and identifies gaps
• Job postings: Tracking competitor hiring patterns reveals strategic priorities
• Social media sentiment: Understanding how audiences respond to competitor campaigns
• Product updates: Automated monitoring of competitor product launches and features

This intelligence enables rapid strategic pivots and ensures you’re always one step ahead.

Also Read: Myths vs reality: Remote and hybrid managers report high productivity and trust

The Southeast Asian context: Unique opportunities

Southeast Asian startups have distinct advantages in implementing these AI strategies:

• Mobile-first audience: The region’s mobile-centric user base creates massive datasets for AI optimisation, particularly in social media and messaging platforms.
• Diverse markets: Our multi-language and multi-cultural audiences provide rich testing grounds for AI personalisation at scale.
• Government support: Countries like Singapore and Malaysia are actively promoting AI adoption through an array of grants and support programs, reducing strategy and implementation costs.
• Growing digital infrastructure: Improving internet connectivity and digital payment systems create fertile ground for AI-powered customer experiences.

Implementation strategy: Start small, scale fast

The key to successful AI implementation isn’t trying to automate everything at once. Smart startups follow a strategic sequence:

• Begin with research: Implement social and competitor research tools to understand your market better
• Automate content creation: Scale your content output to increase brand visibility
• Optimise distribution: Ensure your content reaches the right audiences at optimal times
• Enhance sales processes: Use AI to qualify and nurture leads more effectively
• Integrate systems: Connect all tools for seamless data flow and maximum efficiency

The cost of inaction

The startups that hesitate face a compounding disadvantage. According to a October 2024 Boston Consulting Group study, AI leaders over the past three years have achieved 1.5 times higher revenue growth, 1.6 times greater shareholder returns, and 1.4 times higher returns on invested capital.

In Southeast Asia’s competitive startup landscape, where funding is becoming more selective and customer acquisition costs are rising, the efficiency gains from AI aren’t just advantages anymore. They’re now survival necessities.

Conclusion: The time is now

The “One million+ revenue per person” model isn’t a distant possibility for most, it’s happening today among startups that have embraced AI-powered automation. The tools exist, they’re increasingly affordable, and the competitive advantages are clear.

The question isn’t whether AI will transform how startups operate in Southeast Asia. The question is whether your startup will be among those leading the transformation or struggling to catch up.

The next few months will separate the winners from the also-rans. The choice, and the time to act, is now.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Join us on Instagram, Facebook, X, LinkedIn, and our WA community to stay connected.

Image credit: Canva Pro

The post The US$1M per person revolution: How AI is reshaping Southeast Asia’s startup landscape appeared first on e27.

For decades, the safest career path was simple: join a large corporation, climb the ladder, and let scale and brand reputation shield you from risk.

But AI is rewriting that script. Corporates are automating whole job tiers — customer service, data entry, even junior coding roles. Where once thousands of entry- and mid-level roles were the training ground for future managers, now software eats the repetitive layers.

If corporations no longer absorb as much of the workforce, what happens next?

The future may belong less to sprawling headcounts and more to small, high-output teams — SMEs and startups that can combine AI systems with fractional, remote talent to scale globally without scale in payroll.

The generalist dilemma

Generalists thrive on flexibility. They pivot quickly, learn new tools, and can hold multiple hats. In fast-changing industries, this makes them valuable.

But here’s the risk: shallow generalism is where AI strikes first. Email drafting, basic analytics, social media scheduling — the tasks that junior generalists once owned are now automated.

Survival depends on moving beyond “task doer” into connector, strategist, and problem-solver — the human glue between AI systems and business goals.

Also Read: Value creation: When startups die surrounded by capital

The specialist dilemma

The case for specialisation

Specialists possess deep expertise in a specific domain. In fields such as medicine, engineering, or data science, this depth is essential for solving complex problems and driving innovation. AI is accelerating the demand for specialists who can design, implement, and refine advanced technologies.

Advantages of specialisation:

• Mastery of complex, technical subjects
• Higher demand in niche roles
• Ability to command premium compensation
• Recognition as an expert in the field

However, hyper-specialisation can also create vulnerabilities. As AI automates routine and even some advanced tasks, narrowly focused roles may become obsolete or require constant upskilling.

Specialists stand out with deep expertise. They know a field inside out, and their credibility is built on mastery.

But here’s the trap: if your speciality is codable, you risk obsolescence. Junior developers, paralegals, bookkeepers — entire ladders are being shortened as AI handles entry-level work faster and cheaper.

To stay relevant, specialists must climb “up the stack” into roles AI can’t yet replace: system architects, negotiators, leaders of complex, ambiguous projects. Depth remains an asset, but only if paired with the ability to adapt.

The T-shaped answer: The big shift

The workers — and companies — that thrive will be T-shaped:

• Breadth across domains to stay adaptable.
• Depth in one or two areas to stay differentiated.

Example: A business developer with broad skills in outreach, project management, and digital tools — but deep mastery in regional B2B expansion.

For SMEs, this thinking scales up: build teams that are lean, cross-functional, but anchored by specialists who set direction and operators who adapt.

Also Read: Entering post-unicorn phase, Indonesia signals a structural reset in startup investment

Corporates shrink, SMEs rise

If corporates stop hiring armies of juniors, where will talent go?

• SMEs and startups will increasingly act as the “new training ground.” Leaner, hungrier, and willing to take bets on remote workers, they’ll scoop up talent that corporates no longer absorb.
• Remote and fractional work becomes the bridge. Instead of one full-time job, many professionals will build careers across multiple SMEs, platforms, or projects.
• AI + small teams will compete with big enterprises. A five-person SME with automation and offshore operators can now do what used to require a 50-person department.

The real winners will be SMEs that design teams like portfolios — specialists setting standards, generalists driving execution, and AI filling the gaps.

How to prepare

For individuals:

• Build a core speciality that anchors your credibility.
• Layer generalist skills (AI tools, cross-cultural collaboration, digital literacy) to stay versatile.
• Think globally — your next role may not be in your country, or even in one company.

For SMEs:

• Don’t mimic corporates with bloated teams.
• Use AI as the back office, remote operators as the muscle, and local specialists as the brain.
• Build a system where small teams punch above their weight — faster, leaner, and cheaper than corporates can move.

Conclusion

The future worker isn’t choosing between generalist or specialist — they’re choosing to become both.

The future company isn’t competing on headcount — it’s competing on systems, talent mix, and speed of execution.

As corporates shrink, SMEs have the chance to step up as the engines of growth — building global businesses with teams that look small on paper, but deliver like giants.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Enjoyed this read? Don’t miss out on the next insight. Join our WhatsApp channel for real-time drops.

Image credit: Canva

The post Generalist or specialist? Building future-proof skills in the age of AI appeared first on e27.

The post SpaceX just validated Bitcoin with US$1.4B treasury and Wall Street is taking notice appeared first on e27.

Southeast Asia’s SME economy is often discussed as if it were a single market with a single set of digital needs. It is not. A retailer grappling with refunds in Jakarta does not need the same payments stack as a restaurant managing dine-in, delivery, and click-and-collect in Bangkok, or a services business in Singapore trying to automate recurring billing.

That is one of the clearest takeaways from How Southeast Asia Buys and Pays 2026: Unlocking SMEs’ Potential by IDC and 2C2P. The study argues, implicitly but unmistakably, that generic payment products are increasingly out of sync with how Southeast Asia’s SMEs actually operate. If the region’s next wave of fintech growth is to come from SMEs, it will likely come from more vertical-specific infrastructure, not broader one-size-fits-all checkout tools.

Also Read: Why Southeast Asia’s SMEs are falling out of love with bank-led payments

The numbers back that up. Retail, food and beverage, and services SMEs are all digitising, but under very different operational constraints.

Retail’s problem is not just collecting payments

Retail SMEs in Southeast Asia contribute an average of 13 per cent of total GDP, according to the study, and they sit in the most visibly digitised part of the SME economy. E-commerce platforms, social commerce, marketplaces, and plug-and-play storefront tools have lowered the barrier to entry. But they have also created a new payment mess.

The report shows that 81 per cent of retail SMEs use e-commerce platforms, plugins, or add-ons for online sales. That sounds like a healthy adoption. But the convenience comes with trade-offs. Among retail SMEs, 64 per cent cite limited payment options as a challenge when accepting payments through those platforms, 59 per cent point to security or fraud concerns, and 56 per cent cite high fees or costs.

That is a striking trio of complaints. The very tools that help SMEs get online quickly may also be the ones that constrain conversion and margins once those businesses start scaling.

Then there is the refund problem. Across Southeast Asia, 31 per cent of retail SMEs say returns and refunds affect their payment operations. The pain is especially acute in Indonesia and Malaysia, where the figure rises to 48 per cent and 44 per cent, respectively. In Singapore, it is 37 per cent.

That matters because returns are not a side process in e-commerce. They are part of the customer experience, the cash-flow cycle, and the back-office workload. Yet many SME payment setups are still designed mainly for the initial transaction, not what happens when a sale goes wrong.
Fraud is the other operational tax. Across the region, 55 per cent of retail SMEs say 1 per cent to 5 per cent of transactions turn into fraud or chargeback cases. In Singapore, the proportion reporting that level of incidence rises to 67 per cent. Those are not trivial leakages. They are margin erosion disguised as operating friction.

Retail does not need payment acceptance. It needs payment systems that are built for post-transaction chaos.

F&B lives or dies by channel complexity

Food and beverage SMEs face a completely different set of pressures. The sector contributes around 6 per cent of Southeast Asia’s total GDP and employs an estimated 15 million people, but its digital challenge is less about fraud or refunds than about handling fragmented consumer journeys.

Also Read: Southeast Asia’s digital payments boom has a dirty secret: SMEs still love cash

The report paints a picture of a sector trying to manage dine-in, takeaway, delivery, self-service kiosks, and online order-and-collect flows all at once. That creates an omnichannel payments challenge that many generic providers are poorly equipped to solve.

Across Southeast Asia, 43 per cent of food and beverage SMEs already support online payment and collect at the store through a website or app. At the physical counter, 39 per cent support card payments and 27 per cent support QR code payments. At the table, 31 per cent support card payments and 26 per cent support QR code payments. Even self-service kiosks are generating their own payment mix, with 26 per cent supporting cards and 25 per cent supporting QR payments.

This is not just about offering more ways to pay. It is about reconciling multiple channels, hardware formats, and workflows without adding headcount or operational confusion.

The future demand signals are even more telling. In Malaysia, 56 per cent of food and beverage SMEs want to support online payment and collect at the store in future. In Vietnam and Thailand, that figure is 48 per cent. In Singapore, 47 per cent want better support for card payments at the counter, while 35 per cent want QR code payments at the counter.

The obvious reading is that food and beverage SMEs are not simply digitising payments. They are redesigning the service journey. And when labour shortages and delivery platform competition are already squeezing margins, clunky payment infrastructure becomes a direct drag on resilience.

What this vertical needs is not another generic payment gateway. It needs orchestration across the full stack: in-store, at-table, kiosk, app, and delivery-linked flows.

Services SMEs are still under-automated

If retail is about post-transaction complexity and food and beverage is about channel complexity, services SMEs face automation complexity.

The services segment accounts for around 12 per cent of Southeast Asia’s total GDP, and the report suggests it remains the most under-digitised of the three verticals studied. Many businesses still rely on manual or outdated systems for accounting, governance, customer management, tax, and payments.

Recurring billing is the clearest example. Across the region, only 47 per cent of services SMEs use recurring payment solutions. Usage is higher in Indonesia at 57 per cent, Vietnam at 55 per cent, and Malaysia at 52 per cent, but much lower in Singapore at 38 per cent and Thailand at 39 per cent.

That is surprising, because recurring payments are one of the most obvious ways for service businesses to reduce churn, improve cash-flow predictability, and cut administrative labour. Yet in many cases, SMEs still appear to be processing repeat transactions manually.

The integration data is just as revealing. Across Southeast Asia, only 37 per cent of services SMEs have integrated bookings with their payment technology. The figures fall further for returns at 26 per cent, cancellations at 24 per cent, and refunds at just 14 per cent.

That last number is especially telling. Refunds are often where operational pain becomes most visible to customers, yet they remain the least integrated payment process in the services segment.

The broader issue is that many service businesses are using payments as a bolt-on, not as part of an integrated operational system. That creates inefficiency on both sides of the transaction: more manual work for the business and more friction for the customer.

The next fintech winners will likely look more vertical

The report does not say this outright, but it points strongly in one direction: the horizontal fintech pitch is weakening.

For years, the standard approach was to sell SMEs generic payment acceptance with some localisation layered on top. But as sectors digitise differently, the gaps are becoming more visible. Retail needs better refund and dispute flows. Food and beverage needs channel unification. Services needs recurring billing and process integration.

Also Read: SEA’s SMEs are global in ambition but stuck at checkout

That does not mean horizontal providers disappear. It means the winning products will increasingly be those that feel vertical even if the underlying rails remain horizontal.

For startups, this is less a warning than an opportunity. Southeast Asia’s SME market is huge, but it is not uniform. Founders who keep treating it as a single broad payments category may find adoption and retention harder than expected.

The smarter bet is that the next meaningful gains will come from solving the exact operational pain of a specific segment, then building outward from there.

In Southeast Asia, generic fintech is starting to hit the limits of generic thinking.

The post One size fits none: Why SEA’s SMEs need vertical payment stacks appeared first on e27.

At Echelon Philippines, Maansi Vohra of Monk’s Hill Ventures sat down with Jessica de Mesa-Lim, Founder and CEO of Kindred Health, for an insightful fireside chat on building an omnichannel product tailored to the Philippine market — with a sharp focus on women’s healthcare.

De Mesa-Lim walked through her approach to product development, which begins with deep consumer research, including focus group discussions designed to surface real pain points and unmet needs. From there, the work becomes about crafting solutions that genuinely meet women where they are.

A central theme throughout the conversation was consumer education — de Mesa-Lim believes that when women are informed and empowered about their own health, they become the strongest advocates for their own care. That shift in awareness, she argues, is ultimately what drives adoption and long-term engagement with Kindred Health’s services.

The post Echelon Philippines – From niche to necessity: What Kindred Health teaches us about building tech for Filipino users appeared first on e27.

Singapore’s Doozy Robotics has announced an aggressive global expansion into the United States, the Gulf Cooperation Council, and Asia as it prepares for a Series A.

The startup, which builds humanoid and autonomous-vehicle fleets controlled by a central orchestration layer called Eywa-OS, is pivoting hard into a subscription-first model and pitching itself as a solution to chronic labour shortages in manufacturing and logistics.

What Doozy is selling

Doozy isn’t offering single-purpose machines. Its stack combines an industrial super humanoid, autonomous mobile robots (AMRs), and autonomous forklifts, all governed by Eywa-OS. The software converts high-level production goals into task assignments, dynamically reallocating resources across the floor and reacting to disruptions in real time. The industrial humanoid is slated to launch in Q3 2026, with deployments set to follow.

Also Read: The humanoid robot economy is no longer science fiction

The package is intentionally vertical: hardware, fleet orchestration, and a robot-as-a-service model rolled into one. Customers subscribe to a multi-agent workforce every month and can scale capacity up or down as demand changes, shifting automation from capital expenditure to an operational cost.

The labour thesis

Doozy frames its push around a long-running demographic squeeze and tightening labour markets. The company points to projections that labour constraints in the US could shave more than US$1 trillion off GDP by 2030 and notes that nearly half of American workers are over 45 while Gen Z accounts for roughly eight per cent of the workforce. In that context, Doozy argues facilities will need agentic automation that operates like an intelligent factory manager.

“The global labour shortage is a structural shift, not a temporary imbalance,” said co-founder and CEO Suresh Chandrasekar. “We are building the Physical AI workforce that will power the next era of manufacturing.”

Commercial traction: impressive, but caveated

For a seed-stage company, Doozy’s commercial claims are eye-catching. It says it has a qualified global pipeline exceeding US$200 million, a US$144 million memorandum of understanding (MOU) with a major industrial conglomerate, and a large pilot underway with a US pharmaceutical leader. Doozy lists paying customers across two continents, naming Daimler, Carrier and VitaQuest among its engagements.

Cocoon Capital, an early investor, has publicly backed the company’s traction. “They have cracked some of the most persistent bottlenecks in industrial robotics, such as seamlessly navigating uneven floors and disorganised spaces, and reducing the operational footprint by 50 per cent compared to traditional forklifts,” said Michael Blakey, managing partner at Cocoon. He also called Doozy’s pricing a driver of “inevitable” mass-market adoption.

Also Read: Rise of the machines: 20 robotics startups shaping Southeast Asia’s future

Those metrics deserve scrutiny. A qualified pipeline is not revenue; an MOU signals intent rather than a binding order; and pilots often fail to convert. Investors and potential customers will watch conversion rates from pilot to production deployments, how subscription pricing compares to traditional capex models over time, and the real-world economics of operating humanoid fleets.

Engineering and regulatory hurdles

Humanoids promise dexterity and adaptability where wheeled robots fall short, but they add engineering complexity. Challenges include robust manipulation, battery life, component wear, safety, and maintaining uptime in gritty, industrial environments. Doozy claims its systems can handle uneven floors and cluttered spaces, a key advantage in older warehouses that were not designed for automation.

Safety and regulation are another hurdle. Standards and certification requirements vary across the US, GCC and Asia; integrating humanoids alongside human workers will demand a high bar for safety assurance.
Doozy’s global push will test whether its technology can meet disparate regulatory requirements without eroding margins or slowing deployments.
The economics of subscription robotics

Turning automation into an operational expense is compelling for capital-constrained manufacturers. Doozy’s multi-agent subscription model promises elasticity: scale robots up during peaks, scale down when demand falls. The model also bundles maintenance and software upgrades, theoretically lowering friction for adoption.

However, lifetime customer value and unit economics are critical. Humanoids may have higher maintenance and component costs than simpler AMRs. Supply-chain pressures on actuators, sensors and specialised parts could affect uptime and service costs. The viability of the subscription model hinges on predictable reliability and compelling total cost of ownership compared with manual labour or incumbent automation.

Competition and market timing

Doozy enters a crowded field. Traditional robotics giants, AMR specialists and a growing cohort of humanoid and legged-robot startups are all racing for customers. Doozy’s differentiators are vertical integration and Eywa-OS, which could simplify deployments and enable cross-agent coordination. But the company must prove these advantages at scale and across multiple industries.

Also Read: China’s humanoid robot leader AGIBOT sets sights on Southeast Asia

The planned Series A will be a bellwether for investor appetite in capital-intensive, humanoid-first automation. If Doozy converts a meaningful share of its pipeline into long-term contracts and demonstrates repeatable deployments, it could secure the capital needed to scale hardware production, service infrastructure and international operations. If not, the path forward will be tougher.

The post Doozy’s humanoid fleet goes global; US and GCC in sights appeared first on e27.

Touchstone Partners, a Vietnam-focused climate-tech venture firm, has made a partial exit from Stride, a distributed solar platform for households and micro, small and medium-sized enterprises (MSMEs) in Vietnam, following the startup’s successful Series B raise.

Stride closed a US$15 million Series B co-led by Lightrock and TRIREC via the Accelerate7 platform, with existing backers Clime Capital’s SEACEF II and UOB Venture Management’s Asia Impact Investment Fund II also reinvesting. The financing has prompted at least one early investor to monetise part of its holding, reflecting growing interest in commercial-scale distributed solar in Southeast Asia.

A quick backstory

Touchstone and Clime Capital co-led Stride’s US$2 million seed round in 2023, among the earliest institutional bets on Vietnam’s residential distributed solar market. Since then, Touchstone, which specialises in climate-tech and impact-driven businesses, says it has provided strategic support, fundraising help and introductions to policy and corporate stakeholders while working alongside Stride’s founding team.

Also Read: Stride lands Series A funding to power rooftop solar expansion in Vietnam

Stride’s footprint has expanded substantially: the company now claims operations across north, central and south Vietnam, supported by more than 75 verified engineering, procurement and construction (EPC) subcontractors. Touchstone says Stride has nearly tripled its project portfolio since the seed round and has become Vietnam’s largest platform for residential solar installations.

Why the partial exit matters

Partial exits from early-stage venture funds are noteworthy for several reasons. For limited partners, they provide realised returns and proof that the fund can return capital; for the fund, they free up deployment capacity to back new opportunities. For the market, such exits signal maturation: capital is being recycled from early backers into new deals within the same ecosystem.

“Since we have invested, the company has grown 7.25x in valuation,” Tu Ngo, General Partner at Touchstone Partners, said in a statement. She framed the exit as validation of Touchstone’s thesis that locally built climate solutions can generate both commercial returns and emissions impact, a claim consistent with rising investor appetite for Southeast Asia climate plays.

The figures behind the headline

Stride’s Series B of US$15 million is a meaningful vote of confidence from international and regional investors. Lightrock, a global investor that has backed climate and health tech, and TRIREC, an investment vehicle that uses the Accelerate7 platform,co-led the round. The participation of Clime Capital and UOB Venture Management, both of which reinvested, underscores continued support from regionally focused impact investors.

Touchstone’s partial sell-down comes after a 2023 seed investment of US$2 million that helped Stride scale its operations. Touchstone did not disclose the exact quantum of capital realised in the partial exit, nor the post-money valuation implied by the Series B, standard practice when exits involve negotiated secondary purchases.

What this says about Vietnam’s distributed solar market

Vietnam sits at an inflexion point in distributed renewable energy. Rising energy demand, frequent grid constraints, and a policy environment increasingly focused on decarbonisation have created fertile ground for rooftop and small-scale solar models. However, deployment has long been held back by financing, standards, and fragmented supply chains.

Stride’s model, a marketplace-style platform that aggregates EPC partners and focuses on households and MSMEs, tackles two of those constraints: professionalising installation supply and simplifying customer acquisition. Growing to a network of more than 75 vetted EPC subcontractors suggests the company has solved at least some of the operational challenges that often plague fragmented markets.

Also Read: Lightrock bets US$500M on energy access play across SEA’s unelectrified millions

Yet the market is not without risks. Household and MSME adoption depends on affordable financing, stable incentives, and consumer trust in installation quality and after-sales service. The participation of a range of investors in the Series B indicates confidence in Stride’s ability to manage these risks. Still, competition from utilities, specialised solar providers and other platforms will intensify as capital flows into the space.

Investor strategy and sectoral signal

Touchstone’s move also reflects a broader pattern among early-stage climate funds in Southeast Asia: backing local founders through product-market fit and early scaling, then recycling capital as winners consolidate. By partially exiting a portfolio company after a successful institutional round, the firm increases its dry powder to chase adjacent opportunities in mobility, energy efficiency, clean production and other climate-tech verticals.

Touchstone’s portfolio includes companies such as Selex Motors and Enfarm, positioning the fund across mobility and agri-tech as it seeks to capture multiple vectors of the green transition in Vietnam. The firm also runs the Net Zero Challenge with Temasek Foundation, an initiative intended to bridge funding gaps for green-economy startups, further signalling its role as a catalytic investor in the local climate ecosystem.

What remains unanswered

Important operational and financial details around Stride were not disclosed in the announcement. Key unknowns include:

• The precise amount Touchstone realised from the partial sale.
• Stride’s revenues, profitability metrics or unit economics on residential and MSME installs.
• The company’s roadmap for growth outside Vietnam, if any, and capital allocation plans following the Series B.

These gaps matter for assessing whether Stride’s growth is primarily driven by demand or by favourable capital markets. They also affect how one reads the exit: validation of a durable business model versus a timely secondary sale enabled by strong investor appetite for climate-linked growth stories.

Broader implications for the region

Southeast Asia has become a target for climate-first capital, with distributed solar an obvious entry point. Solar complements other decarbonisation strategies and can be deployed rapidly compared with large grid-scale projects, making it attractive to both impact and growth investors.

The participation of global and regional players in Stride’s round underscores two trends. First, foreign capital is comfortable backing Vietnam’s climate startups when founders demonstrate scalable operations and clear go-to-market strategies. Second, re-investment by existing regional funds signals belief in the company’s execution, not merely a de-risking by new entrants.

What to watch next

• How Stride deploys the new capital: whether it focuses on deeper penetration in Vietnam, moves upmarket to commercial and industrial clients, or invests in financing solutions for customers.
• The competitive response: whether utilities and incumbent solar installers consolidate or partner with platforms like Stride.
• Follow-on exits: whether Touchstone and other early backers exit further as Stride matures toward eventual larger liquidity events.

Also Read: Clime Capital, Touchstone Partners inject US$2M into Vietnamese cleantech startup Stride

For now, Touchstone’s partial exit is a modest but useful data point: investors are willing to both fund and monetise distributed solar bets in Vietnam, suggesting the sector is moving from pilot phase toward commercial scale. That, in turn, could accelerate a wave of follow-on funding and operational consolidation across Southeast Asia’s rooftop solar market.

The post Touchstone partially exits solar startup Stride after US$15M Series B appeared first on e27.

Singapore’s AI national strategy has received its most significant tune-up since the launch of NAIS 2.0 in December 2023. Describing the update as a “double-click rather than a system reboot,” Minister for Digital Development and Information Josephine Teo announced 10 refreshed priorities at the ATxSummit on May 20, underscoring the government’s intent to accelerate rather than reinvent its AI agenda.

The refreshed priorities follow the establishment of the National AI Council in February 2026, chaired by Prime Minister Lawrence Wong, which was set up to provide strategic direction for Singapore’s AI agenda. The update is organised around three focus areas: deepening sectoral and public sector transformation, mainstreaming AI adoption across the broader economy, and cementing Singapore’s position as a regional AI hub.

A central pillar of the plan is the launch of National AI Missions in advanced manufacturing, financial services, connectivity, and healthcare — four industries that are critical to the country’s economic backbone. Together, those four sectors contributed roughly 40 per cent of Singapore’s GDP in 2025. The strategy envisages AI being embedded more deeply across government agencies as well, with the aim of accelerating public sector transformation and improving citizen services.

Workforce development features prominently. The National AI Impact Programme targets 10,000 SMEs for meaningful AI adoption, while the Champions of AI programme offers more targeted support for enterprises seeking to go further. Broad-based capability-building and what the strategy terms “AI bilingual talent” — professionals fluent in both domain expertise and AI application — are positioned as foundational requirements for the transformation ahead.

Also Read: Singapore lands OpenAI’s first lab outside the US with US$225M commitment

On the infrastructure front, Singapore will expand local research compute capacity from 2026 through the National Supercomputing Centre’s ASPIRE 2B supercomputer, as part of a planned national advanced compute, AI, and scientific computing platform. A Digital Infrastructure Act is also expected to be tabled in Parliament to set baseline sustainability standards for data centres.

Internationally, Singapore is leaning into its role as a convener. Minister Teo was candid about the country’s constraints, noting that Singapore’s domestic market alone may not warrant the level of attention it receives, but that its value lies in the global networks it is connected to and its track record in trusted technology adoption.

Business analysis: Opportunity tempered by execution risk

From a business perspective, the updated strategy presents a compelling and largely coherent value proposition. The targeting of four high-GDP sectors with dedicated AI Missions gives multinationals and local enterprises alike a clearer signal of where government support, regulatory clarity, and talent pipelines will converge.

The commitment of more than S$1 billion to public AI research and talent development from 2025 to 2030 — announced earlier this year — provides meaningful financial scaffolding for the private sector to build upon. NVIDIA’s new Research Lab in Singapore and the Punggol Digital District are early markers that global tech players are already responding to Singapore’s hub ambitions.

Also Read: Why you should be hiring humans when others are hiring AI agents

Yet the strategy is not without its risks. The sheer breadth of the 10 refreshed priorities — spanning compute, data governance, workforce capabilities, international partnerships, and sectoral transformation — raises legitimate questions about execution capacity and coordination across agencies.

Smaller enterprises, in particular, may find the SME-focused programmes insufficient to keep pace with the pace of AI disruption, especially as global competitors ramp up investment at scale. Singapore’s acknowledged constraint of a small domestic market also means that the hub ambitions are ultimately contingent on sustained foreign investment and geopolitical stability — factors that lie well beyond any national strategy document. For businesses watching closely, the direction is right; the proof will lie in delivery.

The post Singapore’s AI National Strategy gets a sharp refresh with business ambitions front and centre appeared first on e27.

Quantum themes are now everywhere—from multiverse plots to Schrödinger’s cat, the famous paradox of a cat that is both alive and dead until observed. This surge in storytelling isn’t just cultural curiosity. Experts see it as a sign that imagination is beginning to meet scientific feasibility.

When Sci-Fi becomes signal

The trajectory echoes that of artificial intelligence. AI appeared in fiction for years, but only reshaped industries after deep learning breakthroughs in 2012. Quantum computing may be approaching a similar tipping point.

Until recently, quantum ideas lived mostly in theory. But researchers are now building real systems—stabilising qubits and testing early quantum algorithms.

As science catches up to storylines, analysts note that public imagination often signals where real progress is emerging. For those watching closely, quantum’s rise in fiction could be more than a coincidence—it could be the earliest sign of real-world transformation.

A new kind of computation

Quantum computers differ fundamentally from classical machines. While traditional systems handle tasks step by step, quantum computers explore many possibilities at once—thanks to quantum phenomena like superposition and entanglement.

One expert likens it to a scene from Avengers, where Doctor Strange scans millions of futures simultaneously to find the best path forward. That’s essentially how quantum systems approach complex problems—by evaluating countless outcomes in parallel.

This makes them especially suited for challenges that overwhelm classical systems, such as:

• Cracking next-generation encryption
• Optimising vast logistics networks
• Simulating molecular interactions in drug or material discovery

Quantum isn’t here to replace everyday computing. But for specific high-complexity problems, it represents a fundamentally new—and more powerful—computational model.

Where quantum will hit first

Quantum computing is expected to make its earliest impact in industries where computational complexity is high and financial upside is significant.

Key early applications include:

• Pharmaceuticals: simulating molecular interactions to accelerate drug discovery
• Advanced materials: designing new compounds or batteries at the atomic level
• Finance: optimising asset portfolios, particularly in ETFs and derivatives

Take ETF construction, for example. Selecting the ideal combination of dozens of assets involves combinatorial optimisation—a task that becomes exponentially harder as the number of variables increases. While AI tools help, classical systems struggle beyond a point. Quantum computers, by evaluating multiple combinations simultaneously, offer a clear advantage.

Also Read: Navigating Asia’s business boom: The quantum leadership advantage

In the short term, industries that combine high complexity with high value potential are best positioned to adopt quantum solutions—because the benefits justify the infrastructure investment.

Early wins in the quantum race

While today’s quantum computers remain in the early stages—most with only a few dozen usable qubits—they are already beginning to show practical value in select domains. Many available qubits are still dedicated to error correction, reflecting the sensitivity of current hardware.

Yet despite these limits, meaningful use cases are emerging.

Notable early applications include:

• Drug discovery: simulating molecular behaviour at quantum levels
• Advanced materials: modelling atomic interactions for next-gen compounds
• Finance: improving asset rebalancing strategies in complex portfolios
• Logistics: optimising large-scale routing problems that scale exponentially

These are all areas where classical systems struggle as complexity increases.
Even with today’s constraints, quantum systems are starting to outperform traditional methods in narrow but high-impact scenarios.

The technology may still be maturing, but its real-world value is no longer theoretical—it’s beginning to take shape.

Why quantum-AI hybrids matter

Quantum–AI hybrid computing is drawing growing attention as a practical way to extract early value from quantum systems. Rather than replacing classical computing, the hybrid model assigns different parts of a task to the most suitable processor.

• Classical computers or AI handle large-scale, repetitive calculations
• Quantum systems tackle tasks involving simulation, optimisation, or quantum-specific modelling
• Cloud platforms or machine learning layers integrate and interpret the combined outputs

This division of labour leverages the strengths of each architecture, delivering faster, more efficient results than either could alone. Experts see hybrid models as the most viable short-term strategy—not only technically, but also commercially and operationally—to scale quantum’s impact without waiting for perfect hardware. 

Rethinking the internet for a quantum era

If quantum computers reach commercial viability, today’s internet security architecture will require a complete overhaul. Most current encryption methods—used in banking, e-commerce, communications, and authentication—are vulnerable to quantum algorithms capable of breaking them. This wouldn’t just call for software updates; it would demand a structural redesign of global digital infrastructure.

Experts describe it as a foundational shift, not a technical patch. That said, the transition is expected to unfold gradually over the next decade, giving rise to quantum-resistant cryptographic standards and long-term planning by governments and enterprises.

For infrastructure providers and investors, the key is timing: anticipating when and how to adapt before disruption becomes inevitable.

Korea’s quantum edge: Beyond hardware

While Korea has produced notable quantum researchers, including one of IonQ’s co-founders, full-scale hardware development remains concentrated in global hubs like the US, where companies such as IBM, Google, and IonQ lead in capital and infrastructure.

Instead, Korea is gaining ground in quantum-resilient infrastructure, particularly in quantum-safe cybersecurity. A standout example is SK Telecom, which acquired Swiss-based ID Quantique—a global leader in quantum key distribution (QKD)—and later entered a strategic partnership with IonQ.

Also Read: Horizon Quantum CEO on the Singapore advantage in starting a quantum computing company

This positions Korea to lead in quantum-proof security systems, a field likely to reach commercial scale well before universal quantum computing becomes mainstream.

Experts draw parallels to the early AI wave (circa 2014–2015), when Korea didn’t build foundational frameworks but found success through application-level innovation. Similarly, Korea’s future in quantum may lie in industry-specific algorithms, secure infrastructure, and applied software—not hardware.

How big tech is positioning for quantum leadership

Major tech companies are taking two main paths toward quantum computing: in-house development and strategic partnerships or acquisitions.

• IBM and Google have pursued full-stack integration from the outset—developing quantum hardware, software tools, and embedding quantum capabilities directly into their cloud platforms. They remain the most vertically integrated players in the field.
• Microsoft and Amazon initially focused on enabling quantum access through their cloud ecosystems, partnering with startups to provide tools via Azure and AWS. But as the commercial potential grows, both are moving toward greater internal control.

Recent shifts include:

• Microsoft’s launch of proprietary tools through initiatives like MyOrionow
• Reports of Amazon collaborating with or acquiring startups such as OxenT to build its own quantum stack

This signals a broader trend: big tech is transitioning from collaboration to ownership, aiming to secure key positions as quantum computing moves from theory to viable markets.

Quantum investing: Echoes of the deep learning era

Some early-stage investors see clear parallels between today’s quantum computing landscape and the deep learning boom of the mid-2010s.

In 2012, deep learning began to show real promise. By 2013–2014, major tech firms were investing heavily. During that wave, investors backed AI startups that later went public after a 7-year growth cycle.

Quantum computing now appears to be following a similar arc:

• Foundational research is maturing
• Big tech is entering aggressively
• Use cases are emerging in finance, pharma, and cybersecurity

Many investors haven’t placed direct bets yet, but they’re watching closely.
Angel investors may need a 10-year horizon, while VCs could see returns within three to five years years.

The consensus among early movers? This is the beginning of a new wave—and the time to position is now.

The quantum era won’t arrive overnight—but once it’s here, it will move fast. Those who act early won’t just adapt to the future—they’ll help shape it.

Special thanks to Dr. Jihoon Jeong, General Partner of Asia2G Capital, for his valuable contributions to this article.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Join us on Instagram, Facebook, X, LinkedIn, and our WA community to stay connected.

Image credit: Canva Pro

The post Quantum’s inflection point: Why the smart money is watching now appeared first on e27.

AI adoption across Southeast Asia is accelerating, and with the edge AI market expected to reach US$66.47 billion by 2030 (21.7 per cent CAGR), organisations are moving quickly from pilots to embedding automation in core operations. Agentic AI is expanding what can be automated, prompting enterprises to reassess their technology foundations amid growing regional expansion and rising demands on infrastructure design, visibility and control.

For manufacturers, diversification often replaces a mature ecosystem with a fragmented, cross-border supply chain. Navigating differing regulations and dispersed suppliers deepens reliance on public cloud and hyperscaler AI services, turning infrastructure choices into strategic decisions that determine performance, reliability and resilience.

At the same time, enterprises are shifting from traditional data centres to hybrid control planes. Intelligent edge devices now span factories, clinics and shop floors, unlocking new automation gains but also multiplying attack surfaces. This underscores the need to secure AI workloads consistently across distributed environments and build scalable, resilient architectures.

The AI-cybersecurity arms race

The barrier to entry for sophisticated cyberattacks has collapsed. Agentic AI allows attackers to automate reconnaissance, identify vulnerabilities and scale targeted attacks without deep expertise. Threats that once required manual skill can now be executed by prompting an AI agent to map weaknesses and exploit them efficiently.

Meanwhile, supply chain diversification has fractured defensive perimeters. Thousands of devices and sensors across dispersed facilities now form a sprawling attack surface. As business speed increases, attacks move at the same pace. Manipulative social engineering, deepfake voice impersonation and automated phishing campaigns overwhelm human analysts and exploit the weakest links.

Also Read: How an AI cybersecurity company harnesses the power of AI for optimal business performance

This raises a critical question: how can organisations detect and neutralise AI-enabled threats quickly enough to prevent meaningful damage?

The quantum computing effect

Quantum computing introduces another layer of urgency. As organisations expand their digital borders into fragmented environments, attackers gain conditions to automate and accelerate intrusions. One threat has become especially concerning: “harvest now, decrypt later”. Attackers can steal encrypted data today, store it and wait until quantum systems can break the underlying cryptography. Health records, intellectual property and long-term customer data could become liabilities once decrypted.

This makes the migration window critical. Organisations have limited time to upgrade cryptographic systems before quantum technologies render them vulnerable. Upgrading at scale – discovering dependencies, securing keys and deploying post-quantum algorithms across many systems takes years. If migration takes five years and quantum capability arrives in ten, the clock is already ticking.

Visibility complicates matters. Many enterprises lack a full inventory of keys, certificates or hard-coded encryption calls. Manual audits are slow and incomplete. AI-powered code scanners can accelerate discovery, map quantum-susceptible components and guide modernisation. AI can also detect subtle data exfiltration patterns and deploy countermeasures such as injecting fake data to neutralise stolen datasets.

Compliance will tighten across Asia

Regulators are tightening supply chain mandates and raising expectations for cybersecurity maturity. Japan’s Ministry of Economy, Trade and Industry (METI), will introduce its Cybersecurity Measures Evaluation System for Strengthening Supply Chains in 2026. Similarly, South Korea is strengthening cybersecurity oversight and Hong Kong’s Protection for Critical Infrastructures (Computer Systems) Bill, effective 2026, imposes stronger obligations on organisations to modernise defences.

Compliance is no longer a checkbox exercise — it is a strategic imperative tied to operational resilience and competitive readiness.

Data-heavy industries, look out

Healthcare illustrates the stakes. When sensitive data flows across cloud systems, hospitals and connected devices, even a minor breach can trigger cascading disruption. Similar vulnerabilities appear in manufacturing, logistics, finance and retail, where interconnected digital ecosystems amplify the impact of AI-driven threats.

Also Read: Unchecked shadow AI poses a major cybersecurity risk for 2026: Exabeam

A realistic scenario: an attacker scrapes public data to profile a medical professional, generating a cloned voice and calling the IT help desk to reset authentication credentials. Once inside, the attacker can move laterally and quietly. AI-powered defences are essential because they detect behavioural anomalies — unfamiliar browser fingerprints, impossible travel events or unusual directory access — rather than relying on malware signatures.

How enterprises can stay ahead

• Correlate telemetries at scale: Organisations can improve detection accuracy by correlating telemetry across networks, devices and applications. This uncovers hidden anomalies designed to evade traditional tools. Proactive red-teaming of AI models uncovers vulnerabilities such as data poisoning or manipulation. Explainable AI techniques support forensic analysis by showing why alerts were generated.
• Enforce data provenance and sanitisation: Security begins at the data layer. Organisations should validate data at every ingestion point and prevent modified or corrupted inputs from entering critical systems. Immutable ledgers or blockchain mechanisms ensure trusted provenance and integrity for high-assurance pipelines.
• Address the human element and “shadow AI”: Cybersecurity awareness must extend to all staff. Shadow AI – unvetted tools in daily workflows – poses a growing risk. Core hygiene practices such as least privilege access, multi-factor authentication and granular role-based controls remain essential. Training helps staff recognise modern risks, including seemingly harmless third-party AI tools that could execute tasks autonomously on corporate networks.

The winners in this new era will be those who treat AI security as a strategic advantage, not an afterthought. Building resilience at machine speed requires more than technology—it demands a mindset shift towards dynamic, multi-layered defence. In Southeast Asia’s AI-driven economy, confidence will belong to enterprises that synchronise innovation with security, turning risk into a competitive edge.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic.

Enjoyed this read? Don’t miss out on the next insight. Join our WhatsApp channel for real-time drops.

Image credit: Canva

The post AI adoption in Southeast Asia: Balancing automation gains with the rising threat of cyberattacks appeared first on e27.