Singapore-focused Earth Venture Capital has invested in the US$100 million Series B funding round of US-based Aalo Atomics, a next-generation nuclear energy startup.

Valor Equity Partners is the lead investor.

The funding will be used to construct Aalo’s inaugural nuclear power plant, the Aalo-X, at the Idaho National Laboratory in the US. With criticality targeted for next summer, this project will become the first advanced nuclear facility to commence operations in the US in decades.

This deal comes a year after Earth VC’s investment in Aalo’s US$27 million investment round in 2024.

Also Read: Is Southeast Asia’s data centre boom headed for a PR crisis?

Crucially for the tech sector, this demonstration plant will be paired with an experimental data centre built directly alongside it, showcasing a novel approach to addressing AI’s insatiable energy demands.

The AI-energy nexus and Asia’s urgent need

The implications for Asia, particularly Southeast Asia, are profound. The region is grappling with an escalating energy crisis driven by rapid digital expansion.

Between 2019 and 2023, data centre capacity in Southeast Asia expanded by nearly 30 per cent, consuming an astonishing four to five times more power per square metre than traditional factories. In India, data centre capacity is projected to double by 2026, a direct reflection of surging demand from digital and AI infrastructure.

Across the broader Asia-Pacific region, inventory growth exceeding 20 per cent year-on-year paints a stark picture, with forecast electricity shortfalls of 15-25 GW by 2028. This data underscores an urgent, unmet requirement for clean, reliable baseload energy. Aalo’s modular reactor technology aims to address this pressing issue.

In contrast to conventional gigawatt-scale nuclear installations, Aalo’s reactors are engineered for factory production and rapid, fleet deployment. This design philosophy enables their swift and efficient integration to power critical infrastructure such as data centres, industrial clusters, and utility networks.

Aalo’s strategic roadmap involves scaling from its demonstration plant to deploying thousands of “Aalo Pods.” Each Pod is envisioned to comprise five Aalo-1 reactors and one turbine, designed to power data centres at an unprecedented scale.

The company’s long-term ambition is to slash electricity costs to an exceptionally competitive US$0.03 per kWh, positioning nuclear energy to rival renewables and natural gas in affordability.

Also Read: The AI-energy paradox: Will AI spark a green energy revolution or deepen the global energy crisis?

For Earth VC, Aalo’s Series B achievement resonates deeply with its core mission: to champion ambitious deep tech innovators capable of facilitating the decarbonisation of Emerging Asia and safeguarding prosperity on a habitable planet.

The post Earth VC backs nuclear innovator Aalo Atomics to address Southeast Asia’s data centre power crunch appeared first on e27.

Zhang Fan’s journey with founders–from Maxscend’s slow climb to billions in market capitalisation to mentoring AI-driven startups across Southeast Asia–highlights a simple but powerful truth: technology may start the story, but it’s people, persistence, and practical value that finish it.

As Southeast Asia cements its position as a rising innovation hub, his words resonate as both a challenge and a reminder: tomorrow’s difference makers aren’t just those who shine on stage and quietly keep going when the odds stack high.

Zhang Fan, former founding Managing Partner of Sequoia Capital China and Visionary Partner of the Lee Kuan Yew Global Business Plan Competition (LKYGBPC) by the Singapore Management University, has spent decades backing founders who redefine industries. From early bets on China’s semiconductor pioneers to mentoring student-led startups in Southeast Asia, he has consistently emphasised grit, resilience, and practical value over hype.

Also Read: 60 global startups to compete for US$2M prize at LKYGBPC grand finals

e27 spoke with him to discuss his investment philosophy, lessons from the early days of Maxscend Microelectronics, and why Southeast Asia is brimming with entrepreneurial promise.

You’ve often said you don’t evaluate startups based on hype. Where does that philosophy come from?

It comes from experience. In the early 2000s, I invested in a small startup called Maxscend Microelectronics. The company wanted to build an IC chip that would let mobile phones receive terrestrial TV signals. It felt like a bold, inevitable convergence between broadcasting and mobility at that time.

I didn’t start with spreadsheets; I started with a conversation. These engineers had just returned from Silicon Valley, and their conviction about China’s rising electronics demand was palpable. That belief compelled me to back them.

But the journey was brutal. The company’s first three products failed commercially. They were technically sound, but the market wasn’t ready. Yet what impressed me was how the team refused to give up. By 2010, the team realised the real pain point was in radio frequency (RF) front-end components for smartphones. It quietly pivoted, partnered with TSMC, and launched a low-noise amplifier that global smartphone makers embraced.

Fast-forward: By 2019, Maxscend went public in Shenzhen, and today, it is worth over RMB 41 billion (US$5.7 billion). The lesson? Technology alone doesn’t build enduring companies. It’s about teams that persist, learn, and adapt until they find the right fit.

What do you look for in founders when backing early-stage startups?

I dive into the messy parts: revenue dips, broken funnels, failed go-to-market attempts. I do this not to challenge them but to understand the gravity they’ve fought against.

Resilient founders aren’t afraid of failure; they treat it as feedback. Even now, in the age of AI, where possibilities seem limitless, my compass hasn’t changed. I ask: Does this solve a pain so essential that people will demand it ten years from now?

If the answer is no, then it’s just another flashy demo. That’s why I value practicality over presentation, what I call “practical value.”

Can you share examples of founders who embody this kind of resilience?

Absolutely. Look at Alexandr Wang, who founded Scale AI. He built it to solve a data-labelling challenge he personally encountered. That kind of firsthand problem-solving creates lasting impact.

Closer to home, Lenard Zhuang in Southeast Asia is modernising construction safety using AI-powered video analytics. It’s not glamorous, but it addresses a mission-critical need that saves lives and reduces risk in an overlooked sector.

When local understanding meets global ambition, you get resilience that’s hard to replicate.

You’ve been actively involved with Southeast Asia’s innovation ecosystem. What excites you about it?

The energy here is palpable. Beyond the sheer scale of opportunity, what excites me most are the founders, especially student founders, who are solving real problems.

Singapore Management University’s Whitepaper on Innovation and Entrepreneurship highlights how universities in this region are becoming hubs for scalable and sustainable talent. I’ve seen it firsthand while reviewing submissions for the Zhang Fan Global AI Initiative Award, part of the LKYGBPC.

These founders aren’t chasing applause. They’re building purposeful solutions, often with sharp market understanding and a clear focus. It’s deeply energising to witness.

What will you look for in the next generation of entrepreneurs when you meet them in Singapore this September?

I’ll apply the same lens I used with Maxscend two decades ago. I want to see if they have discipline, humility, and practical value. Do they listen harder after each failure? Do they adapt instead of giving up? Are they building something that will stand the test of time?

Also Read: Why startup founders should not escape failure

True success doesn’t lie in the spotlight. It’s built in those quiet, determined moments when founders keep pushing forward long after others have stopped believing.

Finally, what advice would you give to young founders just starting out?

Start close to home. The strongest ideas often come from problems you’ve personally faced. Stay resilient; failure isn’t the end, it’s the process. And always focus on value that lasts.

Don’t build for Demo Day. Build for the day when your solution becomes indispensable to your customers, even ten years from now. That’s how you create enduring companies.

The post “Don’t build for Demo Day”: Zhang Fan on the enduring truths of entrepreneurship in the AI era appeared first on e27.

The financial landscape presents a compelling narrative of shifting tides and strategic recalibration as we navigate the final stretch of August. Recent developments emerging from Jackson Hole have fundamentally reshaped market expectations, creating a domino effect across asset classes that demands careful dissection.

Federal Reserve Chair Jerome Powell’s carefully calibrated remarks last Friday did far more than hint at potential policy shifts; they effectively slammed the door on prolonged restrictive monetary policy while opening a wide window for immediate easing.

This pivot represents a significant departure from the Fed’s previous stance and carries profound implications for investors globally. Market participants responded with characteristic speed, pushing major US indices to fresh record highs as the S&P 500 gained 1.52 per cent and the tech-heavy Nasdaq surged 1.88 per cent.

The Dow Jones Industrial Average joined this upward trajectory, climbing 1.89 per cent to touch uncharted territory, a development underscored by the US government’s strategic investment in a major semiconductor manufacturer, which provided additional tailwinds for industrial and technology sectors.

Inflation cools, optimism rises

This renewed optimism stems directly from Powell’s acknowledgement that inflation has sufficiently cooled to warrant policy adjustment. His speech deliberately avoided the cautious hedging that characterised previous communications, instead emphasising the Fed’s readiness to act decisively should inflation continue its descent toward the two per cent target.

The immediate market reaction proved remarkably consistent across fixed income and currency markets. Treasury yields tumbled across the curve with the benchmark 10-year note falling 7.4 basis points to 4.254 per cent, while the two-year note dropped 9.5 basis points to 3.696 per cent. This yield compression reflects investor conviction that the current restrictive policy stance is temporary.

Concurrently, the US Dollar Index retreated 0.92 per cent as capital flowed toward risk assets while gold prices rebounded one per cent on the dual catalysts of dollar weakness and heightened rate cut anticipation. These movements collectively signal a powerful shift in market psychology where the so-called Fed put, the implicit promise of central bank support during market stress, has been reactivated with unusual clarity.

Also Read: Blockchain technology: Revolutionising global payment solutions and cross-border remittance

Earnings season exposes a split reality

The earnings season provides a critical counterpoint to this macro optimism, revealing a more nuanced corporate reality beneath the surface. While the much-discussed Magnificent Seven technology giants delivered robust results exceeding lowered expectations, the broader market tells a different story. Analysis of S&P 500 earnings revisions shows a troubling pattern of downward adjustments for the remaining 493 companies.

This bifurcation creates a dangerous illusion where headline index performance masks underlying weakness in the economic mainstream. Investors now turn their attention to the final wave of quarterly reports from key technology players, including Nvidia, CrowdStrike, Snowflake, and Autodesk, alongside consumer stalwarts Lululemon and Dollar General.

These results will serve as crucial stress tests for both the technology sector’s growth trajectory and consumer resilience amid persistent inflationary pressures. The market eagerly awaits these reports not merely for individual company performance but for what they reveal about broader economic health and corporate pricing power.

Asia’s liquidity pressures and regional sentiment

Asian markets present their own complex dynamics, particularly Hong Kong’s interbank rate market, which has exhibited unusual volatility. The one-month Hong Kong Interbank Offered Rate Hibor has surged dramatically from 1.0 per cent on August 11 to 2.77 per cent as of August 22.

This sharp increase reflects significant tightening in short-term liquidity conditions, likely driven by seasonal funding demands and potential regulatory adjustments. Such movements warrant close monitoring as they can transmit stress through global financial channels.

Also Read: Jackson Hole looms: Can Powell save markets from a global risk meltdown?

Despite these regional headwinds, Asian equity markets opened higher during early trading sessions today, suggesting regional investors remain influenced by the broader risk-on sentiment emanating from Powell’s comments. Yet US equity index futures currently indicate a potential pullback at today’s open, introducing an element of caution that underscores the market’s fragile equilibrium.

Crypto’s reaction: from Bitcoin to Ethereum

The cryptocurrency sector experienced particularly dramatic fluctuations following Powell’s speech, creating a fascinating case study in market psychology and whale behaviour. Bitcoin initially surged above US$67 000 following the dovish Fed commentary as traders anticipated lower interest rates would boost risk asset valuations.

However, this rally proved short-lived with the digital asset subsequently declining approximately two per cent. Blockchain analytics firms identified significant movement by large holders shifting substantial Bitcoin positions into Ethereum, a trend that accelerated over the weekend.

Lookonchain data revealed one prominent wallet recently converted part of its 100,784 Bitcoin holdings to acquire 62,914 Ethereum tokens while simultaneously establishing a large derivatives position. This strategic rotation by major players suggests a fundamental reassessment of digital asset allocation priorities, where Ethereum increasingly appears as the preferred vehicle for institutional exposure to the crypto ecosystem.

Ethereum’s technical indicators present both opportunity and warning signs that demand careful interpretation. The cryptocurrency’s 30 day Market Value to Realised Value MVRV ratio has reached 15 per cent a threshold historically associated with profit taking and potential corrections.

Analytics firm Santiment explicitly warns that this constitutes a danger zone that could trigger selling pressure if Ethereum fails to break US$5,000 in the near term. Yet this short-term caution contrasts with the more favourable long-term MVRV ratio of 58.5 per cent, indicating substantial unrealised gains for patient holders.

Additional bullish signals include the declining supply of Ethereum held on exchanges, which suggests growing investor confidence and reduced immediate selling pressure. Combined with rising staking participation and expanding decentralised finance DeFi activity, these fundamentals position Ethereum as the structural cornerstone of the crypto economy rather than merely a speculative alternative.

Strategic imperatives for investors

For investors navigating this complex environment, several strategic imperatives emerge clearly.

Also Read: The intersection of tech and climate change: 5 key forces that will redefine the global market

First, the renewed viability of the Fed puts creates a tactical opportunity to accumulate quality assets during periods of volatility. Well-capitalised investors should view market pullbacks as entry points for fundamentally strong companies, particularly those demonstrating pricing power and resilient cash flows.

Second, the rotation from Bitcoin to Ethereum observed among major holders warrants serious consideration as it reflects a maturation of institutional crypto strategies. Dollar cost averaging into Ethereum provides a prudent approach to managing volatility while maintaining exposure to the asset’s long-term potential.

Third, investors should actively hedge existing cryptocurrency positions using options or futures contracts to protect against potential corrections, especially given the current MVRV warning signals.

Fourth, attention must remain fixed on Ethereum’s technological roadmap, where continued protocol upgrades like further implementation of EIP 4844 will drive sustainable value creation beyond mere speculation.

The road ahead: Volatility and value

The coming weeks will test the durability of this optimistic market posture as investors confront key data points, including the August CPI inflation report, consumer sentiment figures, and potential developments on trade policy. Historical precedent suggests September often brings increased market volatility; the current environment differs significantly from past cycles due to the Fed’s explicit commitment to policy normalisation.

While technical indicators show investor positioning has become somewhat extended, introducing near-term correction risks, the fundamental backdrop of potential rate cuts, combined with resilient corporate earnings, supports continued market advancement. The critical distinction this time involves the quality of the underlying assets driving the market.

Unlike previous cycles, where broad-based speculation fuelled gains, the current environment rewards careful stock selection focused on companies with demonstrable earnings power and sustainable competitive advantages.

This nuanced market landscape demands intellectual rigour and disciplined analysis from investors. The days of indiscriminate buying are over, replaced by an era requiring a granular understanding of both macroeconomic currents and individual company fundamentals.

Powell’s Jackson Hole speech has reset market expectations in profound ways, creating both opportunity and risk that will define investment outcomes for the remainder of 2024. Investors who combine patience with strategic precision while avoiding emotional reactions to short-term volatility will best position themselves to navigate the complex months ahead.

The market’s message is unambiguous: lower rates are coming, but their arrival does not guarantee universal gains. Success will belong to those who recognise that the Fed’s policy shift merely creates favourable conditions; the real work of identifying enduring value remains squarely the investor’s responsibility.

As we move toward September’s pivotal Federal Reserve meeting, the financial world watches with bated breath, knowing that the decisions made in the coming weeks will reverberate through markets for years to come.

—

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 courtesy: Canva Pro

The post Powell’s pivot: How Jackson Hole reshaped markets and what comes next appeared first on e27.

What does it take for a 10-person team to generate US$200M in annual recurring revenue? Or for a startup to reach US$250M ARR in under two years, without a massive fundraising round?

For Marc Manara, head of startups at OpenAI, the answer lies in a disciplined blend of speed, focus, and the intelligent use of AI infrastructure. Speaking at OpenAI x SCB 10X in Bangkok, Manara pulled back the curtain on how the company works with founders worldwide, and why its platform is becoming a launchpad for the next generation of market leaders.

From workflows to full-stack agents

Manara’s definition of an AI “agent” was intentionally functional: a workflow to guide behaviour, tools to expand capability, and guardrails to ensure appropriate, ethical outcomes. In his telling, agents are not novelties; they are the new operational backbone.

The real shift, he argued, will be from isolated AI functions to full-stack solutions capable of handling multi-step processes, integrating into existing workflows, and adapting across industries.

The 2025 investment lens

OpenAI’s priorities for the year ahead underscore this trajectory:

• Models and customisation
• An agents platform
• Multimodality

Multimodality, where text, audio, and images flow seamlessly through one system, is already reshaping product design. Manara framed it as a “first-class capability,” not an experimental feature.

Lean teams, outsized returns

Two examples dominated the discussion:

• Cursor: 20 employees, US$250M ARR in just 21 months
• Midjourney: 10 employees, US$200M ARR in two years

Both exemplify what Manara calls seedstrapping: building significant traction and revenue before pursuing large-scale funding. The model favours rapid iteration, tight feedback loops, and a relentless focus on product-market fit.

Also Read: AI gold rush: How OpenAI’s Singapore expansion could reshape the startup ecosystem

Beyond the API

OpenAI’s engagement with startups goes far beyond providing API access. Programmes include:

• Enhanced concierge support: solution architects, account escalation, and dedicated office hours
• Exclusive resources: API credits, invite-only technical sessions, and “build hours” with OpenAI experts
• Direct influence: alpha and beta access to shape the product roadmap

This is underpinned by OpenAI’s internal loop: Research → Apply → Deploy → Repeat, which Manara urged founders to replicate.

For those paying close attention

Not everything shared in Bangkok was on the slides. Manara pointed to two resources that, while technically public, are rarely promoted and often overlooked: one on the emerging frontier of text-to-speech, and another on a discreet pathway to privileged access within OpenAI’s startup ecosystem.

I will share both with readers who follow me here on e27. Think of it as a private briefing for those who are actively paying attention.

A global play, from Bangkok

The keynote underscored a bigger truth: global-scale AI infrastructure is no longer a Silicon Valley monopoly. As OpenAI expands its reach into Southeast Asia, the advantage will go to founders who can not only access these capabilities but operationalise them faster than their peers.

Want the two “golden nuggets” I mentioned? Follow me here on e27 and I’ll send them to you directly. Sometimes, knowing where to look is the real advantage.

—

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 courtesy of the author.

The post From Bangkok to billions: Inside OpenAI’s startup growth playbook appeared first on e27.

e27 has been nurturing a supportive ecosystem for entrepreneurs since its inception. Our Contributor Programme offers a platform for sharing unique insights. As part of our ‘Contributor Spotlight’ series, we shine a spotlight on an outstanding contributor and dive into the vastness of their knowledge and expertise.

This episode features Jaslyin Qiyu, Founder and Managing Director of Mad About Marketing Consulting. With over 20 years of B2B and B2C marketing experience across the Asia Pacific, she has led regional teams at global MNCs, including Citibank, EY, JLL, Kantar, Credit Suisse, and State Street.

At Mad About Marketing, Qiyu focuses on brand building, client experience, MarTech, and performance marketing, while championing flexible, fractional talent models that empower women to pursue both career and personal goals. She also serves on advisory and industry boards, including CX Networks, University College Dublin, Trigger-UNDP, and RSVP Singapore.

In the sections below, she reflects on her journey, the lessons he’s learned, and what keeps her going.

How I got here

The turning point for me came when I was advising a startup on their wealth tech platform’s customer journey and value proposition. Their genuine trust in my inputs, coupled with their dedication despite having limited resources, made me realise I could create more impact outside of traditional corporate structures.

That experience crystallised my frustration with the industry’s one-size-fits-all approach and inspired me to start Mad About Marketing Consulting, to democratise access to sophisticated marketing capabilities.

If I had to explain my work to a kid

I help companies show people how cool their products are, kind of like when you see a toy ad on TV that makes you say, “I want that!” I teach them how to tell stories so customers understand why their stuff is fun, useful, or special. It’s like being a matchmaker, helping the right people find the things they’ll really love.

Also Read: Leading through transformation: How CMOs and CEOs must evolve in the AI era

Lessons learned along the way

I used to think I had to feel “completely ready” before saying yes to new opportunities, so I turned down many roles because I thought I wasn’t qualified enough. Over time, I realised that real growth comes from solving real problems for real people, not from collecting more credentials.

My question shifted from “Am I qualified enough?” to “How can I create value while learning?” That change in mindset transformed my approach, from being cautious and over-preparing to stepping forward with confidence and contributing right away.

What more people should notice

There’s a dangerous misconception that generative AI can replace experienced marketing expertise. Too many startups think they can bypass hiring seasoned marketers by using AI tools to generate campaigns, content, and strategies without understanding fundamental customer psychology or market dynamics.

The result is often superficial marketing that lacks depth and misses the mark. The real opportunity lies in combining AI’s capabilities with experienced marketing leadership, where strategy, customer insight, and execution excellence create impact that AI alone cannot deliver.

Why I write

I have always been passionate about writing. In fact my childhood ambition was to be a writer! I also believe knowledge should be free and accessible to all, which is why AI powered research and info crawling has taken off so rapidly.

My writing often starts with real client challenges I’ve encountered and evolves into actionable insights others can use. At its core, my flow comes from a genuine belief that sharing knowledge creates positive ripple effects across the business community.

Also Read: How tech startups can attract Gen Z and millennials seeking flexibility and purpose

My advice for aspiring thought leaders

Here are a few principles I follow whenever I share my thoughts and experiences:

• Start with solving real problems rather than trying to sound impressive. Authenticity resonates more than jargon.
• Always explain complex ideas simply, because clarity demonstrates true understanding better than complexity.
• Share your failures and learning moments alongside successes. Vulnerability creates deeper connections and more valuable insights for your audience.

What drives my curiosity

All things spiritual and the universal realm, including what exists outside of what we can see with the human eye and mind.

Influences that shaped me

Growing up would be my parents and in terms of books, it’s mix of animal books by James Herriot and philosophical ones that prompted me to become naturally curious about the human mind and perception. One of the earliest books I read was Sophie’s World that got me really inspired to learn and read more about philosophy.

Take a look at Qiyu’s articles here for more insights and perspectives on her expertise.

—

Are you ready to join a vibrant community of entrepreneurs and industry experts? Do you have insights, experiences, and knowledge to share?

Join the e27 Contributor Programme and become a valuable voice in our ecosystem.

The post How Jaslyin Qiyu is redefining marketing leadership with flexible talent models and real impact appeared first on e27.

I was trained as a general practitioner. My early career began in rural clinics, working face-to-face with patients who often arrived late into illness, and sometimes too late for care. Some walked for hours just to reach the nearest facility. Many couldn’t afford the medicine. A few never came at all, out of fear, stigma, or the belief that nothing would change anyway.

Practising in those conditions changed how I understood health systems. The pathways that led people to care or kept them away from it entirely are so much deeper than just treatments or interventions. Distance, cost, shame, and bureaucracy are all systemic and technical obstacles, not human obstacles (and definitely not something your physicians can get rid of with a snap of their fingers). These obstacles existed long before diagnosis or treatment.

That experience stayed with me. It led me to ask harder questions about how health should work. It also led me into innovation, initially in ways that felt small. I started testing different approaches. I looked into patient vital sign self-check-ins, mobile consults, early screening tools, and how digital workflows could reduce drop-offs in care, even insurance products based on real cost and claim data (within our own beta population). But it didn’t take long to realise that the moment you try something new in healthcare, you run into friction.

That friction isn’t always direct from users (whether it’s physicians, health admins, or patients themselves, but rather it’s policy. Sometimes it’s culture, sometimes it’s just inertia. It becomes difficult to move fast when every step is bound by systems that were designed to minimise risk. That instinct makes sense in a clinical context. But it creates resistance when we’re trying to redesign the system itself.

This point in building user feedback and commencing trials with healthcare facilities or healthcare payors that have a sandbox for innovation is a lifesaver (for an early stage company like mine). We started integrating with other services to strengthen their reach, building mutual channel partnership. We saw how technology, when placed carefully, could expand care without increasing pressure on already overburdened systems. We focused on design that removed barriers for both patients and providers.

Now, I work more deeply with AI in healthcare. I see the same patterns re-emerging. We talk about scribing, supply chain coordination, clinical decision support, Software as a Medical Device (SaMD), and even risk modelling for population health. Each use case offers clear advantages. Yet the resistance often comes before the discussion starts.

Also Read: Decoding digital preferences: A glimpse into the future of health tech ecosystem in SEA

People worry about safety, scope, ownership, ethical review, and clinical validity. These concerns matter. But what I’ve observed is that this resistance isn’t stronger than any pushback we’ve seen before (new drugs, supplements, wearables, even robotic surgery; once faced the same level of pushback and some even scrutiny). Every medical innovation in history has gone through it, whether it was antiseptics, laparoscopic surgery, or digital health records. Change is often uncomfortable. But it is never new.

So what’s the real challenge?

Instead of calling it a blocker, I think we need to shift the frame. The misconception is that value is the main driver for innovation. Only after innovating you understand that it actually is about understanding regulation, workforce, education, procurement, reimbursement, and behaviour. Medical innovation becomes normalised when the whole ecosystem is ready to hold it and is aligned across multiple levels of influence (not a single breakthrough overnight).

I’ve seen AI pilots fail, especially because the workflows couldn’t adapt to the real-time day to day operations our healthcare workers face, not because the models. I’ve seen great tools ignored because they didn’t match how clinicians document cases. I’ve seen hospitals decline adoption because IT budgets weren’t structured to handle long-term updates or retraining. These are signals that we need better integration strategy and regulatory pathways (like any other new drug in the market).

Healthcare is complex because it should be. We are dealing with lives. We are dealing with trust. But complexity shouldn’t stop us from building. It should shape how we build.

Also Read: What telemedicine and Health Tech holds across SEA amidst COVID-19

In Southeast Asia, the opportunities are real. We have gaps that technology can help close. The transformation should starts with people who understand the gaps and are willing to build bridges. It starts with small, focused systems that can grow and scale. It starts with conversations that go beyond hype and address what readiness actually looks like. Once we understand that, product building now becomes problem solving deliveries on a deep level.

My path began in rural clinics. I now build for broader systems. The problems have changed shape, but the mission remains the same. Make care more reachable. Make care more trusted. Make care feel possible.

If we want AI in healthcare to succeed, we need to stop waiting for the perfect pilot. We need to understand what adoption truly takes. We need to stop labelling every pause as resistance, and start seeing it as part of a wider transformation journey. Every advancement in medicine required coordination. This one is no different.

—

The post A doctor’s journey through rural practice, healthcare economics, innovation, and ecosystem appeared first on e27.

A significant shift is underway in the corporate finance world, as Chief Financial Officers (CFOs) abandon cautious approaches to Artificial Intelligence (AI) in favour of aggressive, strategic investments aimed at long-term revenue growth.

New research from Salesforce reveals that AI, particularly ‘agentic AI‘ — digital labour capable of autonomous task execution — is not merely a tool for cost reduction but a fundamental driver of business transformation. This global trend holds crucial implications for tech startups and established enterprises across Southeast Asia looking to optimise operations and foster innovation.

The paradigm shift in AI strategy

Just five years ago, a striking 70 per cent of global CFOs adhered to a conservative AI strategy, a figure that dropped to 34 per cent two years ago.

Today, that number has plummeted to a mere 4 per cent, indicating a widespread recognition among financial leaders that AI is now a crucial tool for enhancing efficiency, optimising operations, and driving critical long-term growth. In fact, a third of CFOs have now officially adopted an aggressive approach to AI.

Also Read: Forget the rest: ChatGPT alone drives more traffic than 10,500 AI tools combined

This rapid transformation stems from a fundamental rethinking of technology investment Return on Investment (ROI). Robin Washington, President and Chief Operating and Financial Officer at Salesforce, commented, “The introduction of digital labour isn’t just a technical upgrade; it represents a decisive and strategic shift for CFOs. With AI agents, we’re not merely transforming business models; we’re fundamentally reshaping the entire scope of the CFO function. This demands a new mindset as we expand beyond financial stewards to also become architects of agentic enterprise value.”

The rise of agentic AI and redefined ROI

A significant 61 per cent of CFOs report that AI agents are changing how they evaluate ROI, moving beyond traditional metrics to encompass a broader range of business outcomes. On average, CFOs are now dedicating a substantial 25 per cent of their current total AI budget to AI agents. This commitment underlines a belief among 61 per cent of CFOs that AI agents and digital labour are, and will continue to be, critical for competing in the current economic environment.

Furthermore, 64 per cent of CFOs state that AI agents are changing their perspective on how their business spends money, with over a third (35 per cent) acknowledging that AI necessitates a riskier mindset around technology investments.

Beyond cost-cutting: Revenue and strategic value

While traditional technology investments often focused on immediate, measurable results, the perception of AI’s value extends far beyond short-term cost-cutting. Today, CFOs recognise AI’s returns may accrue over the long term through ongoing processes and new business models.

Seventy-four per cent of CFOs believe that AI agents will not only cut costs but also drive revenue. CFOs implementing AI agents anticipate these agents will increase company revenue by almost 20 per cent.

AI agents are uniquely suited to improve long-term business outcomes such as revenue generation, productivity gains, and improved decision-making. Significantly, 72 per cent of CFOs say AI agents will transform their business model, and 55 per cent believe AI agents will take on more strategic work than routine tasks.

New metrics for success

The introduction of AI agents has expanded the top factors CFOs consider when evaluating AI ROI.

These now include:

• Cost savings, risk and compliance improvements, and revenue growth (tied as the number one factor).
• Productivity or efficiency improvements (ranked second).
• Improved decision-making (ranked third).

One CFO survey respondent noted, “Traditional technology investments mainly focus on immediate financial returns that can be easily visible, but AI benefits are a mix of long- and short-term duration. KPIs are focused based on business outcomes.”

Additionally, AI is viewed as a valuable way to ensure ROI through better financial control, with one CFO stating, “AI provides real-time budget tracking, which improves forecasting accuracy and helps protect ROI from overspending through better financial control”. For CFOs, redefining ROI demands a mindset shift from valuing short-term to long-term success.

Concerns and the path forward

Despite the aggressive adoption, CFOs still face significant concerns regarding their AI strategy. The two primary worries are security or privacy threats (66 per cent) and the long time to ROI (56 per cent). Concerns also include the ethical risks associated with AI, which could affect reputational cost and ROI, and the ongoing investment required for retraining, monitoring, and improving AI models, making ROI more fluid compared to fixed-function tools.

Also Read: AI for the real world: SEA’s cost-efficient playbook is winning investors over

This global shift underscores the growing imperative for businesses, including those in the vibrant Southeast Asian tech startup ecosystem, to strategically integrate AI into their core operations. As financial leaders redefine value beyond immediate returns, embracing agentic AI is becoming a critical competitive differentiator for long-term growth and innovation in the digital labour era.

The post Salesforce study 2025: Only 4 per cent of CFOs still play it safe on AI appeared first on e27.

Despite the rapid advancement of artificial intelligence (AI) in customer experience (CX), consumers are far from ready to abandon human interaction.

A new Verizon report highlights a strong preference for the human touch, presenting a clear mandate for brands to adopt a hybrid approach to CX rather than seeking fully automated solutions.

Humans still reign supreme in customer satisfaction

The report’s findings are unequivocal: 88 per cent of consumers are satisfied with online interactions involving human agents, compared to only 60 per cent for interactions driven solely by AI. This significant gap underscores a fundamental truth: while AI can handle routine queries efficiently, complex or emotionally charged situations necessitate human empathy and problem-solving.

Also Read: Verizon report: Businesses hail AI in CX, but customers still prefer humans

Consumers are “broadly relaxed” about AI for tasks like purchase transactions and product inquiries, but “fewer are comfortable when AI handles their complaints”.

The most prominent frustration consumers cite in automated interactions, by a large margin, is the inability to speak or chat with a live agent when needed, affecting 47 per cent of respondents. Brands concur, noting a similar proportion of customer complaints regarding this lack of human access. Stacy Sherman highlights that even when human agents are eventually involved, “information about you/the customer is lost (must be repeated) at different stages of the interaction,” causing further “friction with customers”.

A hybrid future for CX investments

Acknowledging this preference, companies are not solely betting on AI. The report indicates that a substantial 44 per cent of brands intend to split their future CX investments roughly equally between AI-driven and human-driven improvements. This signals a recognition that a balanced approach, integrating the strengths of both AI and human agents, is the most effective path forward. Only 29 per cent foresee CX operations being mostly or fully AI-driven.

“Some challenges require more than just solutions—they require the empathy and care that only people can provide,” states Morlon Bell-Izzard from Exelon.

Upskilling the human workforce

For this hybrid model to succeed, customer-facing staff require specific upskilling to work effectively alongside AI. Executives are prioritising training in three key areas:

• Handling customer complaints about chatbots.
• Understanding AI prompts during interactions.
• Handling complaints about data privacy issues.

The report also stresses the importance of addressing the “emotional and psychological barriers” employees might have about AI, ensuring transparency about how AI will enhance, not replace, their roles. As Abhii Parakh, Head of Customer Experience at Prudential Financial, observed, employees initially sceptical of AI became excited “once they saw how beneficial AI is for them”. Companies can use AI itself as a “powerful simulation tool” for training, allowing employees to practice interactions and build confidence.

Also Read: AI personalisation isn’t working; more consumers say it hurts CX than improves it

In essence, while AI will continue to automate and optimise parts of the CX journey, the human element remains irreplaceable for delivering truly empathetic and comprehensive customer service, making a seamless human-AI collaboration paramount.

The post 88% of consumers favour human agents; AI alone fails to deliver CX satisfaction appeared first on e27.

Customer expectations now grow faster than most operating models can adapt. Cisco’s 2025 Global Contact‑Centre Survey predicts that by 2028, 68 per cent of customer‑service interactions will be handled by agentic AI systems. Meanwhile, a 2024 Dixa poll shows 96 per cent of consumers still consider empathy critical to brand relationships.

Technology must scale, but humanity cannot be engineered out. The solution is not another point bot; it is a platform-first architecture that orchestrates data, workflows, autonomous agents and human experts inside a single, governed fabric.

The disconnect: Why AI isn’t delivering on the hype

Boards approve nine‑digit AI budgets, yet many projects remain stalled in pilot purgatory. Forrester’s 2024 Digital Business Strategy Survey reports that only 56 per cent of leaders have an enterprise‑wide view of technology. McKinsey’s 2025 research finds 95 per cent of AI initiatives never scale beyond pilots. Fragmented systems starve models of context‑rich data, while governance is treated as an afterthought.

Consider a telecom provider that launched separate chatbots for billing, network faults and promotions. Each bot answered its narrow script, but none shared a common data layer. Customers bounced between channels, escalation volumes spiked, and Net Promoter Score barely moved. Scattered tools, even “smart” ones, cannot substitute for a unifying platform.

What makes agentic AI different?

Traditional chatbots follow decision trees; agentic AI is goal‑driven, contextual and action‑oriented. Picture an AI agent that detects a billing anomaly, issues the refund, updates the CRM, emails an apology and alerts a human only if the amount crosses a threshold. Autonomy at that level creates three non‑negotiables:

• Context hunger: Curated, lineage‑tracked data streams
• Governance demand: Transparent audit trails and policy controls
• Interoperability: The freedom to swap models without re‑wiring applications

Only a platform layer that abstracts data, policy and workflow can meet all three at production scale.

Also Read: Agentic AI, urban mobility & smart tourism: 2025’s travel investment hotspots

Human + AI: Better together

Automation excels at speed and pattern recognition; humans excel at judgment, negotiation and relationship‑building. The goal is augmentation, not substitution. A global med‑tech firm recently introduced an agentic‑AI layer that triages tickets and surfaces knowledge‑base articles.

Human agents now focus on complex clinical queries, pushing first‑contact resolution and CSAT to record highs. When machines handle the routine, people deliver empathy exactly what 96 percent of customers want.

Platforms: Architecture for orchestrated intelligence

In Forrester’s 2024 survey, 70 per cent of digital leaders said technology and business executives now collaborate closely on change initiatives a shift directly linked to unified platform strategies. Such platforms provide:

• Composable services: API‑driven micro‑components that let teams plug in new AI models within days, not quarters
• Unified data fabric: Clean, trusted streams feeding both agents and analysts
• Governance by design: Access controls and audit logging embedded where functionality lives

This is strategic infrastructure, not middleware. A robust platform turns isolated bots into a coordinated workforce that learns, adapts and stays auditable.

APAC spotlight: DBS Bank

A 2024 Harvard Business School case study details how Singapore‑based DBS Bank built an internal data‑and‑model hub plus a PURE (Purposeful, Unsurprising, Respectful, Explainable) responsible‑AI framework. With that foundation, DBS scaled 300‑plus AI use cases across lending, fraud and service, boosting self‑service adoption and helping lower false‑positive fraud alerts. The case exemplifies platform‑first thinking in one of EdgeVerve’s key regions.

Also Read: 88% of consumers favour human agents; AI alone fails to deliver CX satisfaction

Scaling through strategic orchestration

A platform mind‑set reframes AI from a stand‑alone tool to a capability woven through every business process. High‑performing organisations:

• Swap models without disruption: Treating language or vision models as hot‑swappable modules under existing guardrails.
• Propagate success: Templating connectors so a winning use case in one region can be cloned elsewhere with minimal recoding.
• Monitor holistically: Combining experimentation metrics and production KPIs in a single observability stack.
• Automate compliance: Making centrally defined policies inherit automatically to every new workflow.

One Asia‑Pacific conglomerate recently merged a dozen AI pilots onto a single platform. Release cycles for new virtual‑assistant features shrank from months to weeks, and CSAT climbed double digits proving orchestration, not model count, drives value.

CXO playbook: three principles for the next wave

• Think platform‑first: Invest in data fabric, API gateways and governance layers before chasing the next generative model.
• Design for empathy + autonomy: Map journeys where human touch is irreplaceable and bake those checkpoints into orchestration.
• Embed governance early: Treat explainability, lineage and compliance as design inputs, not retrospective audits.

The road to harmony

Enterprises that orchestrate human and agentic intelligence on a strong platform spine will transform customer experience from reactive support to proactive value creation. Cisco’s survey notes 81 per cent of CX leaders believe vendors that master agentic AI will carve out enduring competitive advantage.

The stakes and the opportunities couldn’t be clearer. From hype to harmony, the future belongs to those who blend scalable technology with human empathy into one coherent, governed platform

—

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 courtesy: Canva Pro

The post From hype to harmony: Why agentic AI needs a platform-first mind-set to redefine CX appeared first on e27.

While AI is increasingly becoming a common practice in organisations, the ability of organisations to develop better growth prospects and more efficient strategies and customised consumer experiences also emerges clearly. AI brings unseen possibilities, such as automating decisions and delivering distinctively targeted advertising campaigns.

However, this technological advancement has not come easy without a set of problems. For instance, AI has alarmed privacy issues, the main reason being that systems demand personal data as inputs, and this acts as a violation of customer trust between firms and customers. Apart from that, the possibility of involuntary biases in the algorithms themselves, whereby an AI system unwittingly designs bias into the algorithm, resulting in “unfair” results, could go a long way in straining brand image and consumer trust.

Regardless, there is no doubt that AI has profoundly transformed the rapidly growing marketing technology (martech) industry, especially in the face of these aforementioned ethical issues. While AI optimises automation, increases customer satisfaction and improves analytical capabilities, organisations have to find the right balance between innovation and accountability. In this article, we will explore what these issues are and how they can be effectively addressed.

Essential ethical considerations to address

As AI continues to penetrate different industries, it has never been more urgent to raise awareness and ensure that AI is properly implemented. According to Forbes, more than 51 per cent of company leaders believe that AI transparency and ethics are critical to their operations, and 41 per cent of top executives have halted the deployment of AI technologies due to a potential ethical concern.

Transparency in the context of AI means that the functioning of artificial intelligence should be perceptible and understandable, while the process of making decisions should conform to ethical norms and general human values. One can find an example of transparency in the case of many companies employing AI to understand the behaviour of customers, targeting their advertisements and the overall marketing management.

To support the increase in transparency, some organisations have started giving customers more information about the usage of their data. AI transparency is also important where the risks are especially high that the consequences of AI decisions will impact lives or have large social outcomes, such as in healthcare and finance.

Another ethical consideration that has to be discussed is the issue of bias and discrimination in AI. AI comes with many advantages but is not without its controversies, particularly on issues of bias and discrimination. This is due to the fact that most AI models are trained from large datasets that could mirror some of the bias in the society hence the biased results.

Also Read: Blockchain and AI copyright: A revolution in digital rights management

Bias in AI can stem from various sources such as: 

• Bias in training data: If the training data contains inherent biases, the AI system will likely reproduce these biases in its decision-making processes. For instance, in a study, scientists tasked AI with developing a facial recognition system designed to classify individuals into three categories based on their characteristics: doctors, criminals, and homemakers. However, the AI demonstrated bias in its decision-making, frequently labelling women as homemakers, Black men as criminals, and Latino men as janitors, and selecting women of all ethnicities less often as doctors.
• Algorithmic bias: Beyond the data, poorly designed algorithms can amplify existing biases or create new ones. In 2018, Amazon’s AI recruitment algorithm was designed to assess candidates based on their fit for different roles. However, due to the underrepresentation of women in technical positions, the system developed a bias, favouring male applicants as it learned that men were historically preferred for these roles.
• Cognitive bias: Personal experiences and perspectives may lead developers to prioritise certain data over others, potentially skewing the AI’s outputs. For example, favouring data from a particular demographic or geographic region might result in an AI system that does not accurately reflect a global or diverse population. 

Strategies for mitigating bias and promoting fairness in AI

In 2024, Malaysia presented a PDP Bill that outlines significant changes in the Personal Data Protection Act (PDPA), including the definition of the terms, added responsibilities for data controllers, and increasing fines for non-compliance. The government regards these changes as great progress in enhancing data protection in the country and as a part of the continuous shift toward stricter privacy rules. This presents a good chance for companies to enhance their protection of data and bring them to par with global standards.

To start, there are various measures that companies can take to make the process ethical and responsible. One of the key strategies is to prioritise transparency, where businesses must provide clear insights into how AI algorithms operate. For instance, developing an explainable AI (XAI) plays a vital role in this process, as it offers techniques to help users understand and trust the decisions made by AI. By incorporating simplified visuals or user-friendly software interfaces, employees can grasp the underlying processes without relying on AI systems blindly. 

In addition to transparency, maintaining robust data security is critical. Research shows that 44 per cent of security decision-makers say their companies incorporate security and privacy measures from the outset when developing services, products, or applications.

Moreover, 87 per cent of consumers state that they won’t engage in business with a company if they have concerns about its security practices. This underscores the importance of continuous data monitoring, with dedicated personnel responsible for safeguarding information and preventing leaks. 

Also Read: How AI and automation can shape the future of farms

Companies should also ensure that their AI solutions comply with industry regulations and legal standards, as organisations that prioritise ethical AI are more likely to gain consumer confidence and create reliable AI systems. Furthermore, creating the role of human supervision as an AI control factor–where the AI makes suggestions that are then passed on to human experts to make the final decision is another positive since it assures that the systems are running fairly and effectively.

Implementing all these measures is critical in developing AI systems that are not only ethical but also efficient. At OpenMinds, we believe that we have a responsibility to lead by example, and we understand the importance of integrating ethical considerations into any AI development process.

Conclusion

In conclusion, reducing bias and encouraging fairness in the AI system is not only a technical issue but also an ethical issue. The strategies outlined are essential steps towards building trustworthy and ethical AI systems. We believe that these ethical considerations are especially important. As we continue to innovate in the martech industry, we aim to contribute to a future where AI benefits everyone, regardless of their background and identity.

—

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, and LinkedIn to stay connected.

Image credit: Canva Pro.

This article was first published on August 19, 2024.

The post Navigating the AI maze in Malaysia’s martech: Striking a balance between efficiency and ethics appeared first on e27.