MoneyHero Limited (Nasdaq: MNY), the financial aggregation platform operating across Greater Southeast Asia, has reported a nominal net income for the second quarter of 2025 (Q2 2025), a stark reversal from the substantial losses recorded last year.

However, a closer inspection of the financials reveals that this technical profitability was achieved primarily through aggressive cost-cutting and favourable foreign exchange movements, masking a significant 13 per cent year-on-year (YoY) decline in total revenue.

Also Read: Decoding MoneyHero’s Q1: The profit push amid shrinking revenues

The Singapore-based group, which operates platforms like SingSaver and Moneymax, announced a net income of US$0.2 million for Q2 2025, sharply contrasting with the net loss of US$(12.2) million during the same period in 2024. Management celebrated this pivot as evidence that their strategy for “durable, profitable growth” is working.

Yet, this profitability success relies heavily on operational discipline rather than top-line growth. Total revenue dropped to US$18 million in Q2 2025 from US$20.7 million a year earlier. MoneyHero attributes this contraction to a “deliberate moderation of lower-margin credit card volumes” and a strategic shift towards diversifying revenue mix.

The true drivers of profit: Cost cuts and FX gains

While the transition to higher-margin verticals like insurance and wealth is underway (comprising 27 per cent of revenue, up five percentage points YoY), the critical levers driving the US$0.2 million net income were deep expense reductions and external financial factors:

• Operating cost massacre: Total operating costs and expenses (excluding net foreign exchange differences) plummeted by 37 per cent YoY to US$20.6 million. This dramatic reduction was broad-based, including strategic technology cost reductions, simplified operations, and a “comprehensive restructuring” of employee benefit expenses. Employee benefit expenses specifically dropped from US$6.712 million in Q2 2024 to US$3.700 million in Q2 2025.
• Foreign exchange windfall: The reported net income was significantly bolstered by an unrealised foreign exchange gain arising from the weakening of the US dollar against local currencies during the quarter. The net foreign exchange differences swung from a loss of US$(1.848) million in Q2 2024 to a gain of US$2.969 million in Q2 2025. The unrealised foreign exchange gain, net, was US$2.951 million for Q2 2025, compared to a US$1.766 million loss a year prior.

Interim CFO Danny Leung affirmed that the Q2 performance proves the model is “structurally healthier,” translating into stronger profitability thanks to improved unit economics and disciplined reward calibration. The adjusted EBITDA loss also improved by 79 per cent YoY, reaching US$(2.0) million.

Southeast Asia revenue collapses

The focus on profitability has come at a severe cost to key Southeast Asian revenue streams

Data reveals sharp contractions in regional markets:

• The Philippines: Revenue dropped dramatically by 42.2 per cent YoY, falling from US$2.938 million in Q2 2024 to just US$1.697 million in Q2 2025. The Philippines’ contribution to total revenue dropped from 14.2 per cent to 9.4 per cent.
• Taiwan: Revenue was nearly halved, contracting by 47 per cent YoY, falling from US$1.424 million to US$754 thousand.
• Malaysia: Revenue has effectively dried up, registering US$0 in Q2 2025, down from US$28 thousand a year prior. (MoneyHero does, however, retain an equity stake in the operator of RinggitPlus in Malaysia).

The group’s financial stability is now highly reliant on its two key hubs, Hong Kong and Singapore, which together accounted for 86.4 per cent of Q2 2025 total revenue (Hong Kong at 43.3 per cent and Singapore at 43.1 per cent).

Application volume tumbles despite membership rise

While the MoneyHero Group reported a 33 per cent YoY expansion in group members to 8.6 million, the platform’s actual transactional activity declined significantly, underscoring the shift away from high-volume, lower-margin business.

Total applications sourced by MoneyHero dropped by 14.3 per cent, falling from 476,000 in Q2 2024 to 408,000 in Q2 2025. Approved applications saw an even steeper decline, falling by 18 per cent YoY, from 211,000 to 173,000.

Also Read: Profitability gains mask deeper challenges for MoneyHero in Q4 2024

CEO Rohith Murthy noted that these movements were part of a disciplined effort, stating that AI initiatives are already reducing customer acquisition cost per approval and “improving approval quality”. However, the data confirms that fewer customers are successfully completing the application funnel, reflecting the “deliberate moderation of lower-margin credit card volumes”. Credit Cards still generate the vast majority of revenue (60.8 per cent of Q2 2025 revenue).

Unreliable traffic metrics mask true user engagement

A critical caveat buried within the financial report relates to operational data measurement. MoneyHero explicitly stated that due to Google’s mandatory transition from Universal Analytics (UA) to Google Analytics 4 (GA4), which took effect on July 1, 2024, the key metrics of monthly unique users (MAUs), traffic, and clicks are “not comparable” to prior periods.

The company claims that Google has not provided sufficient information to assess this methodology transition’s positive or negative impact. This change means that the reported Q2 2025 MAUs (5.3 million) and total traffic (16.7 million) lack a reliable, verifiable year-on-year benchmark, making it impossible for the media to accurately assess the platform’s organic traffic stability prior to mid-2024.

Looking ahead, MoneyHero is strategically investing in higher-margin segments, including a planned launch of Hong Kong’s Credit Hero Club with TransUnion in Q4, with expected expansion into other markets. The company forecasts that Insurance and Wealth will comprise approximately 30 per cent of Group revenue by the end of 2025.

Management remains confident in achieving positive adjusted EBITDA in the later part of 2025, supported by the structural improvements reflected in their current numbers.

The post MoneyHero swings to profit, but only on cost cuts and FX gains appeared first on e27.

History shows that equities often deliver solid gains in the year following the start of such cycles, with the S&P 500 averaging around 14 per cent returns over 12 months based on past data from various cycles. Yet the initial month after the first cut tends to bring choppiness, as markets adjust to shifting monetary policy.

In this case, equities rallied strongly leading up to the cut, pushing the S&P 500 up 15 per cent in the past six months and a whopping 32 per cent from its yearly lows. I see this preemptive surge as a sign of market optimism, but it also raises flags for potential consolidation ahead.

Investors priced in these cuts long ago, so the real test comes from upcoming events like speeches from 10 FOMC governors and the PCE inflation report due on September 26. If inflation data surprises on the upside, volatility could spike, reminding everyone that even dovish Fed actions carry risks in an economy where growth projections and unemployment trends diverge sharply.

Fund managers’ bold bets on risk assets

Fund managers continue to lean heavily into risk assets, particularly equities, despite nagging worries about persistent inflation and a weakening US dollar. This positioning strikes me as bold, perhaps overly so, given the mixed signals from the broader economy. Many in the industry view artificial intelligence as a deflationary force that could counter current inflationary pressures, an idea that holds water when you consider how AI efficiencies might drive down costs in sectors like manufacturing and services.

Gold’s performance this year underscores these tensions, with prices surging 38 per cent year-to-date amid buoyant rate-cut expectations and geopolitical uncertainties. Research into recent gold price drivers points to factors like central bank buying, trade tensions, and lower interest rates making the metal more appealing as a hedge.

However, fund flows indicate much of this buying stems from speculation rather than genuine hedging, and professional asset managers maintain low exposure to gold and digital assets. In my view, this creates intriguing opportunities for savvy investors to buy on dips, especially as gold hit US$3685.30 per ounce recently. The under-allocation by institutions suggests room for further upside if economic headwinds intensify, but it also warns against chasing the rally without careful consideration.

Also Read: SGX tightens climate reporting rules, expands green products as sustainable finance demand grows

Balancing strategy: Barbell approach and diversification

Strategic advice in this environment boils down to respecting the Fed’s direction without blindly following the herd. The central bank’s dovish stance supports companies with strong earnings growth, yet piling into mega-cap tech stocks at current valuations feels precarious. A barbell approach makes sense here, where you hold core positions in quality names, add selectively during pullbacks, and diversify into global themes and yield-focused plays.

Singapore’s yield stocks stand out, having outperformed the S&P 500 over both five- and ten-year periods, which bolsters the argument for regional diversification beyond US borders. I favour this strategy because it balances growth potential with income stability, particularly in a world where US-centric portfolios risk overexposure to domestic policy shifts.

With the Fed projecting more cuts on October 29 and December 12, lower rates could fuel corporate borrowing and expansion, but diverging economic indicators demand vigilance. Unemployment might tick up if growth slows more than expected, potentially pressuring equities despite the supportive policy backdrop.

Macro shifts and geopolitical influences

Turning to the broader macro picture, global risk sentiment holds firm thanks to the allure of additional rate cuts enhancing corporate earnings outlooks. The week ahead features the high-level General Debate at the 80th UN General Assembly starting Tuesday, with President Trump addressing the opening session and Fed Chair Powell discussing the economic outlook. These events could inject fresh narratives into markets, especially amid positive developments in US-China relations.

President Trump’s recent conversation with Chinese leadership led to a deal on the popular Chinese-owned social media app TikTok, allowing it to continue operations in the US under new controls, which eased some trade anxieties. Wall Street responded enthusiastically, with major indices hitting record highs on Friday, the Dow Jones up 0.37 per cent, S&P 500 up 0.49 per cent, and Nasdaq up 0.72 per cent.

Treasury yields edged higher, the 10-year at 4.127 per cent and the 2-year at 3.572 per cent, reflecting a mix of growth optimism and inflation watchfulness. The dollar index climbed 0.30 per cent to 97.64, while gold rose 1.1 per cent on rate-cut bets. Brent crude dipped 1.1 per cent to US$66.68 per barrel following EU sanctions on Russian oil vessels and buyers, highlighting ongoing energy market fragilities. Asian equities showed mixed results Friday and in early trading today, with US futures pointing to a lower open, suggesting caution amid these crosscurrents.

Also Read: Trust, not just technology: What I learned building AI finance tools for SMEs in Southeast Asia

Bitcoin volatility and regulatory headwinds

Bitcoin’s recent dip adds another layer to the market mosaic, with the cryptocurrency falling 0.93 per cent to US$114,566 over the past 24 hours, lagging the broader crypto market’s 1.91 per cent decline. This underperformance stems from profit-taking after hitting recent highs, technical breakdowns near the US$115,400 resistance level, and regulatory uncertainties.

Drawing from similar patterns in September 2024 analyses, when Bitcoin traded around US$63,000 and faced consolidation phases, the current setup echoes familiar volatility drivers like leverage unwinds and momentum shifts. Over US$176 million in long positions were liquidated as prices tested US$115,000 support, amplifying the drop in a classic liquidity trap where open interest spiked 4.14 per cent to US$937 billion.

Traders piled in near resistance, only to face swift reversals, which I interpret as a healthy correction in an asset that has risen 81 per cent year-to-date. The failure to break above the 23.6 per cent Fibonacci retracement at US$115,400 triggered algorithmic selling, with the MACD histogram still positive at +265 but RSI hovering around 51-53, indicating waning momentum.

Bulls must secure a close above that level to reclaim initiative; otherwise, a breach of US$114,500, the 30-day simple moving average, could cascade toward US$110,000.

Regulatory developments weigh heavily on Bitcoin’s short-term path, presenting a neutral but noisy influence. The US Treasury’s commentary on the GENIUS Act, issued on September 20, emphasises stablecoin regulations requiring full reserves in liquid assets like Treasuries and technological capabilities for freezing assets, aiming to foster innovation while curbing risks. This act, signed into law earlier this year, mandates issuers to comply with federal laws, potentially stabilising the crypto ecosystem but introducing oversight that cools institutional enthusiasm.

Meanwhile, the EU’s MiCA regulation drives exchange consolidation, enforcing consumer protections, market integrity, and restrictions on stablecoin use as exchange mediums, which impacts global flows. Exchanges adapting to MiCA gain credibility and access to unified EU markets, but the compliance shifts have slowed inflows, with ETF volumes dipping and minor outflows from products like GBTC.

In my opinion, these headwinds represent growing pains for the sector; long-term clarity should prove bullish by attracting more institutional capital, yet the immediate uncertainty often halts rallies, as seen in past cycles.

Closing thoughts: Cautious optimism ahead

Overall, the market’s resilience amid the Fed’s pivot impresses me, but I remain cautious about overextended positions in tech and crypto. The historical precedent of positive equity returns post-rate cuts offers encouragement, yet the unique blend of geopolitical events like the UNGA, US-China thawing via the TikTok agreement, and persistent inflation worries calls for measured optimism. Gold’s speculative surge and Bitcoin’s technical wobbles serve as barometers for broader risk appetite, suggesting investors diversify thoughtfully.

The barbell strategy aligns with my view that quality growers paired with yield plays provide a sturdy foundation, especially as Singapore’s outperformance demonstrates the value of global exposure. With PCE data looming and more Fed cuts on the horizon, markets could consolidate, but the underlying dovish support tilts the scales toward gradual upside. Still, chasing crowds in mega-caps or digital assets without waiting for pullbacks risks unnecessary pain; patience often rewards in these environments.

As we head into the final quarter, keeping an eye on unemployment trends and corporate earnings will prove crucial, potentially defining whether this cycle mirrors the average 14 per cent gain or veers into choppier territory. The economy’s divergences remind us that while the Fed guides, real-world data ultimately steers the ship.

—

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 The rate cut rally: Earnings, gold, and Bitcoin in the balance appeared first on e27.

Quantum is no longer confined to research labs; it’s edging into boardrooms, banks, and biopharma pipelines. Few investors grasp this transition as clearly as Alexandra Beckstein, CEO of QAI Ventures, which has chosen Singapore as its Asia-Pacific headquarters.

In this conversation, Beckstein unpacks why Singapore offers the right mix of policy, talent, and industry appetite to become a launchpad for quantum ventures, how lessons from Europe and North America are being adapted here, and why venture building could be the key to turning patents into profitable, globally competitive companies.

Why Singapore? Beyond the government’s strong quantum push, what specific gaps or opportunities did you see that drove QAI Ventures to set up its Asia-Pacific HQ here?

At QAI Ventures, we take a global approach to investing in quantum tech and advanced computing. Talent, technology, and applications are scarce and globally dispersed, so we continuously scout for emerging hubs that show strong signals of ecosystem maturity.

Singapore combines all the key ingredients: world-class universities, a robust IP regime, strong government commitment through its National Quantum Strategy, and clear industry momentum. A recent example is OCBC’s collaboration with NUS, NTU, and SMU to explore quantum applications in banking – a sector we consider a priority for early adoption.

QAI Ventures has had success in Europe and North America. What lessons from those ecosystems are you directly applying, or avoiding, in Singapore?

I would point out three lessons learnt from our previous work:

• Integration with the ecosystem is key: Success in Switzerland and Canada has shown us that tight integration with academia, startups, corporates, and investors is essential. We are bringing the same approach to Singapore.
• Our global approach in talent scouting and providing access is a core USP: We open our hubs internationally, attracting founders to where industry strengths lie. For example, North American teams join our Basel accelerator to access Europe’s life sciences ecosystem. Singapore offers the same draw for finance and industrials.
• Industry-driven focus pays off: We learned to prioritise verticals with early market fit. That’s why we’re concentrating on Finance, Industrials, and Life Sciences in Singapore, supported by initiatives like our GenQ Hackathon in October, where participants tackle SDG challenges at the intersection of these clusters.

Also Read: DigiCert CEO: Quantum computing’s “ChatGPT moment” is coming

In addition, we see venture building as a critical lever. Accelerator programmes focus on existing startups , but starting a venture building programme allows us to combine IP from Singapore’s research base with our database of over 2,000 quantum patents, so we can form globally competitive ventures here. That’s a new approach we bring to the market for the first time, and we chose Singapore as the first location to do so.

Quantum computing remains largely research-driven. What gives you confidence that quantum ventures in the city-state can move from labs to boardrooms faster than elsewhere?

Singapore has built one of the world’s most coordinated national quantum strategies, backed by long-term funding and public-private partnerships. As a financial and logistics hub, it provides immediate boardroom-level use cases in areas like fraud detection, risk modelling, and secure supply chains. The country’s regulatory clarity and strong IP protection make it easier for ventures to attract global investors and partners. Its compact ecosystem concentrates world-class talent and industry leaders in close proximity, shortening the path from research to pilots.

Combined with Singapore’s tradition of being a living lab for frontier technologies, this makes it uniquely positioned to move quantum ventures from lab to boardroom faster than elsewhere.

Many quantum startups face long commercialisation cycles. How will your venture building programme accelerate research into profitable, globally competitive businesses?

Our unique strength is being an ecosystem builder. We limit each cohort to six founder teams to ensure tailored mentoring, close sparring, and strong connections to corporates, investors, and industry experts.

Quantum startups often face long commercialisation cycles because the leap from lab to market requires both deeptech and industry validation. Our programme shortens that path by pairing researchers with experienced entrepreneurs, industry partners, and early adopters from day one. We focus on market-driven use cases in finance, pharma, logistics, and cybersecurity, ensuring the science is steered toward real customer pain points. With structured support in IP strategy, regulatory readiness, and fundraising, we help teams de-risk their journey and attract global capital faster.

Ultimately, the programme transforms cutting-edge research into profitable, globally competitive businesses by combining scientific excellence with commercial execution.

Which sectors in Singapore (finance, life sciences, industrials) do you see as the first real adopters of QuantumAI?

We see strong potential across all three of them: finance, life sciences, and industrials:

In Singapore, the finance sector is expected to be the first real adopter of QuantumAI. Banks such as HSBC and DBS are already exploring quantum-secure communications and advanced risk modelling, and the potential to apply QuantumAI to fraud detection or faster Monte Carlo simulations for portfolio optimisation makes adoption both commercially attractive and regulatorily relevant.

The life sciences industry is also well-positioned, given Singapore’s ambition to be a global biomedical hub. QuantumAI could accelerate drug discovery by simulating complex molecules far beyond the reach of today’s supercomputers, similar to the recent collaboration between Moderna and IBM on mRNA structure modelling and support precision medicine through the analysis of genomic data.

Finally, industrials and logistics form a natural testbed: the Port of Singapore and Changi Airport already lead globally in digitalisation and are experimenting with quantum-secure networks, as seen in the Port of Rotterdam’s scalable quantum internet pilot.

These examples show how Singapore can move quickly from research to real-world adoption, with finance likely leading, life sciences close behind, and logistics providing large-scale, operational deployments.

Given the competition from other hubs like Tokyo, Sydney, and Hong Kong, what will it take for Singapore to stay ahead?

To stay ahead in a region full of innovation leaders, Singapore must build on its strengths while learning from peers like Tokyo, Sydney, and Hong Kong.

Tokyo brings world-class corporate R&D and long-standing leadership in advanced materials and electronics, while Sydney excels in academic research and deep science talent. Hong Kong continues to serve as a vital financial gateway to China and a bridge to mainland innovation.

Also Read: Quantum computing’s double-edged sword could threaten cybersecurity: Report

Singapore differentiates itself by combining regulatory clarity, strong IP protection, and political stability with dense connectivity between finance, biotech, and logistics industries, all within a compact, highly international ecosystem. Unlike Tokyo, which has deep corporate R&D but slower commercialisation cycles, Singapore’s public-private partnerships allow startups to move from pilot to market much faster. Compared to Sydney’s academic strength, Singapore offers direct access to global banks, biotech multinationals, and world-leading logistics hubs like its port and airport. And while Hong Kong positions itself as a financial hub for China, Singapore has established itself as a trusted global hub with deep international connectivity.

By fostering collaboration across these hubs rather than competition, and by staying true to its tradition as a living lab for frontier technologies like QuantumAI, Singapore can remain the launchpad where global startups and local founders test, scale, and commercialise world-class innovations.

Venture capital in quantum is risky, with some saying it is even speculative. What’s your investment thesis in terms of risk appetite and expected time horizons for returns?

QuantumAI is a frontier field, and risk and opportunity go hand in hand. Our investment thesis is built on de-risking for LPs with a two-sided approach:

With our early-stage fund, we combine capital with our accelerator programme, providing structured mentoring, industry integration, and global exposure to improve market fit and increase survival rates.

With our growth-stage fund, we invest in companies already generating revenue, operating in markets on the cusp of adoption, and with clear exit pathways through M&A or IPO. We believe the next three to four years offer a rare window of opportunity in QuantumAI. Growth-stage investments allow shorter holding periods, faster ROI, and multiple exit options due to strong M&A appetite for IP-rich companies.

At the same time, we don’t just wait for deal flow: through venture building, we create companies from scratch by combining strong IP with entrepreneurial teams. This puts us in a unique position to shape the ecosystem, reduce risk, and capture upside across stages.

The post From labs to boardrooms: QAI Ventures bets on Singapore’s quantum future appeared first on e27.

Indonesia is positioning itself as a major player in the global clean energy transition, with a national strategy that places batteries and electric vehicles (EVs) at its core. The government’s approach combines ambitious adoption targets, downstream industrialisation, and foreign investment. But as Prof. Dr Evvy Kartini, founder of the National Battery Research Institute (NBRI) and professor at the National Research Innovation Agency (BRIN), explains in an email interview with e27, the path forward remains complex.

Indonesia’s efforts to promote EV adoption date back to a 2019 presidential regulation. The regulation set out bold 2030 targets: 14 million electric two-wheelers and 4 million electric four-wheelers. These commitments align with the country’s pledge to the Paris Agreement.

Yet, progress has been slower than anticipated. “Though the number of EV products sold increased twice this year, [it is] still below one million,” Prof. Dr Kartini says. She points to persistent barriers, including high costs compared with internal combustion cars and limited infrastructure. “The infrastructure of charging stations or swap stations, as well as the standardisation and safety, are still a concern in Indonesia.”

Government incentives, such as tax exemptions, have been introduced to spur demand, but the gap between targets and reality highlights the scale of the challenge.

Downstreaming nickel for strategic advantage

Indonesia’s new administration has placed industrial downstreaming at the centre of its energy and economic strategy. Under its “Asta Cita” or eight-vision framework, the fifth vision is dedicated to expanding value-added industries. Nickel, which Indonesia holds the world’s largest reserves of, is at the heart of this plan.

Also Read: Tiger New Energy raises US$3.5M to accelerate deployment of its battery swapping network

“With this vision, we expect that the ‘nickel’, for example… could be processed to become not only Mixed Hydroxide Precipitate (MHP) but also the battery precursors and cathode materials in Indonesia,” Prof. Dr Kartini explains.

With vast nickel reserves, Indonesia is investing in EVs and batteries to boost its role in Asia’s clean energy future.  The government has already opened investment in upstream and midstream industries, and a gigawatt-hour (GWh) battery factory project broke ground in June. It is led by a consortium of state-owned miner ANTAM, the Indonesia Battery Corporation (IBC), and China’s Contemporary Amperex Technology Limited (CATL/CBL).

The professor, who recently spoke at Thailand’s 3rd ASEAN Battery Technology Conference 2025, argues that this will create a more efficient supply chain. “The existence of a battery factory with supply chains from Indonesia will reduce the cost of production,” she says.

Despite these opportunities, Prof. Dr Kartini cautions against potential mismatches between policy plans and industry realities. She cites the ongoing debate between lithium iron phosphate (LFP) and nickel manganese cobalt (NMC) battery chemistries as a possible stumbling block.

“The issue of using LFP instead of NMC for the EV may become the burden for this program,” she says.

The government risks undermining its nickel-centred industrial policy without precise regulation or incentives favouring nickel-based EVs.

Innovation and sustainability

Prof. Dr Kartini sees technological innovation and sustainability as essential for the industry’s long-term viability. “The innovation in the battery industry is finding a material that can provide higher energy density and a larger capacity. [It must also be] lightweight but with high power density and fast charging. It should be friendly to the user and the environment and safe to use,” she says.

Also Read: Transitioning to new energy? Here’re 5 prominent solutions for your business

She also stresses the importance of circular economy practices, particularly battery recycling. “The circular economy, such as the reuse and recycling of spent lithium-ion batteries, must be carried out; do not wait until it burdens the environment and society at the end,” she warns.

Indonesia’s trajectory has attracted significant foreign investment across the battery value chain. Prof Dr Kartini points to progress across all stages: upstream mining, midstream battery manufacturing, and downstream applications in EVs and renewable energy. “Indonesia has already reached a milestone and will become the largest battery producer through the consortium of ANTAM-IBC-CBL,” she says.

With its domestic market and regional links, Indonesia is well-positioned to play a central role in Asia’s EV sector. She predicts that four-wheeler EVs, particularly affordable imports from China, Korea, Vietnam, and Japan, will dominate soon.

However, the success of this plan depends on whether regulatory clarity, infrastructure investment, and technological innovation can keep pace with targets. Realising Indonesia’s vision will require alignment between government, industry, and innovation: “Otherwise, it will be very difficult in the end to penetrate the market.”

—

Image Credit: ABTC 2025

The post Facing EV hurdles, Indonesia looks to nickel and battery supply chains for answers appeared first on e27.

In a sunlit conference room in San Francisco, a group of developers huddle over a table scattered with phones, tablets, and laptops. The air is electric with possibility, but tinged with frustration. One device keeps lagging, another refuses to load the demo. The problem isn’t the code. It’s the connection.

“We’ve optimised everything,” one engineer mutters, thumbing through logs. “But when the signal drops, the whole thing falls apart.”

Welcome to the fragile reality of AI on mobile.

Despite astonishing leaps in generative AI (image generation, natural language interfaces, and real-time summarisation) the entire experience often hinges on a single, outdated assumption: that the user is always connected to the cloud. As AI continues to embed itself deeper into our daily lives, from customer support to healthcare diagnostics to personal assistants, that assumption is becoming less and less viable.

The next frontier of AI isn’t in bigger models or smarter prompts. It’s in bringing intelligence closer to the user: physically. And that means rethinking where data lives, how it moves, and what happens when the cloud is simply out of reach.

Cloud can’t keep up

To understand the shift, start with how most AI systems currently operate. When you open a chatbot on your phone or ask your smart assistant a question, the request typically travels across the internet to a cloud server where the large language model (LLM) processes the input, consults relevant databases, and sends a response back.

This setup works beautifully when bandwidth is strong, latency is low, and privacy isn’t a concern.

But increasingly, those conditions don’t hold.

Users are demanding smarter apps that work offline, in real-time, without sacrificing responsiveness or control. They expect AI to assist them on subways, in rural areas, on factory floors, and in high-security environments where external data connections are limited or forbidden. And they expect it to do so while preserving data privacy and ensuring low power consumption on resource-constrained devices.

The cloud, once the crown jewel of modern computing, is beginning to look like a bottleneck.

Also Read: From village to cloud: Why public-private partnerships hold the key to inclusive tech in SEA

On-device AI is ready—almost

Technically, we’re closer than ever to running real AI locally.

Thanks to advances in quantisation, pruning, and specialised silicon—such as Apple’s Neural Engine, Google’s Tensor SoC, and Qualcomm’s AI chips—it’s now possible to run compressed versions of LLMs directly on mobile hardware. Models like Phi-2, LLaMA, and Gemma have been distilled to sizes small enough to live on-device and still perform impressively on many natural language tasks.

But here’s the kicker: AI doesn’t exist in a vacuum. Models need data. They need context. And they need it fast.

Without local access to relevant, real-time data (user preferences, recent activity, stored documents, or business-critical information) the model may be running on your device, but it’s still blind. That’s the paradox. We’ve brought the brain to the edge, but not the memory.

The data problem at the edge

Imagine a travel assistant that knows your itinerary, hotel reservations, and preferences. You’re in a taxi in Rome, with no Wi-Fi, and you ask for a dinner recommendation near your hotel. The model runs on your phone, but the hotel address? That’s stored in the cloud. Without it, the AI draws a blank.

Or take a field medic using an AI tool to interpret patient data in a disaster zone. The model is on the device, but if the patient records are locked in a remote database, the application fails at the moment it’s needed most.

The problem isn’t just about fetching data, either. It’s more about ensuring that the data is accurate, synchronised, and secure, across millions of devices, many of which may be offline for hours or days at a time. Data has to move intelligently between the cloud and the edge, updating incrementally, resolving conflicts, and maintaining integrity without draining battery life or compromising privacy.

Solving this data mobility challenge is the linchpin to unlocking AI’s full potential on mobile.

Also Read: Singapore hit by 6.4M cyberattacks in 2024 as AI supercharges threats

Rethinking infrastructure for AI at the edge

What’s needed now is a new kind of infrastructure, one that treats data like a living, breathing entity that must exist both on the device and in the cloud, capable of thriving in both connected and disconnected environments.

This means:

• Offline-first design: AI apps must function even without a network. No fallback mode. Full functionality, even in airplane mode.
• Bidirectional sync: Changes made on-device should sync back seamlessly once a connection returns, without data loss or duplication.
• Latency-free access: Models must be able to query data at memory speed, not over-the-air latency, especially during inference.
• Security and privacy at the core: Sensitive information should never leave the device without explicit consent and encryption.
• Scalability across fleets: Whether it’s 100 or 10 million devices, the system must keep them in sync, without micromanagement.

In short, the backend of AI needs to evolve, away from centralised architectures and toward distributed, intelligent data systems purpose-built for the edge.

A future that’s closer than it looks

We are entering a phase where the intelligence is not just in the cloud, not just in the model, but in the choreography between device, data, and environment.

AI that truly feels human, i.e. contextual, responsive, and always available, will not be achieved through brute compute or bigger models. It will come from systems designed to operate at the speed of thought, wherever that thought occurs, and regardless of whether a signal bar is present.

The future of AI on mobile will not be replacing the cloud, but rather liberating the user from it.

And that future is already starting to take shape. Quietly. Locally. One intelligent, offline interaction at a time.

—

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 Why the future of AI on mobile may not be in the cloud appeared first on e27.

For startups, visibility has always been a survival skill. You need to win investor trust, attract talent, and convince customers to bet on you instead of an established player. In today’s environment, that challenge has a new dimension. It is no longer just about reaching people; it is about connecting with them. It is about ensuring your brand is part of the knowledge base that powers Generative AI tools, such as ChatGPT, Gemini, Copilot, and DeepSeek.

A recent Forrester survey found that 89 per cent of B2B buyers already use GenAI in their decision-making process. This technology is no longer optional. It is integrated into every stage of the buyer journey, from scanning the market to comparing vendors to validating ROI.

For startups trying to punch above their weight, this shift can either level the playing field or make you invisible. The question is: how do you ensure that when AI platforms answer, your brand is part of the story?

The media hierarchy that AI listens to

Unlike Google, AI models do not crawl the web in real time. They rely on training data and high-credibility sources. That creates a new hierarchy of influence that every startup founder should be aware of.

• Mainstream media: Outlets like BBC or The New York Times set global credibility. One strong placement here can echo in countless AI-generated answers.
• Industry publications: Coverage in places like TechCrunch or e27 defines trends and categories. For startups, this is often the most attainable and highest-leverage tier. A TechCrunch profile or an e27 founder story is more likely to be referenced in AI explanations of “emerging fintechs in Asia” than a single blog post on your site.
• Branded thought leadership: Whitepapers and founder essays, if picked up by respected platforms, become frameworks AI repeats when offering strategic advice.
• Academic and policy reports: Data-driven research remains gold. If your startup contributes to or is cited in these, you gain durable authority.
• Forums and Q&A: Communities such as Reddit or Quora shape how AI models learn conversational tone. A viral founder AMA can have more downstream influence than you think.

For startups, the takeaway is clear. Visibility is not about chasing one channel. It is about showing up across the spectrum so you influence both the authoritative and conversational layers of AI.

Also Read: The agritech challenge in Indonesia: Can AI and mobile apps enhance productivity?

Why content structure matters more than ever

Startups often have fewer resources, which means your content must work harder. AI models prefer structured, example-rich material that they can reuse. Three formats stand out.

• How-to guides: Step-by-step advice on how to solve a problem. For startups, this could be “How to set up cross-border payments in Southeast Asia.” AI picks these up for “how do I” queries, giving you authority by default.
• Frameworks and lists: Clear models, even if simple, travel well. If your startup coins a framework such as “3 ways SMEs can digitise their logistics,” AI is more likely to replicate it in answers.
• Case studies: Concrete stories with metrics. “In 90 days, our pilot customer cut costs by 25 per cent” is more valuable than aspirational messaging. It teaches AI to ground advice in evidence.

The rule for founders: if you want your insights to spread, package them in ways that machines can easily copy and humans can quickly grasp.

From SEO to GEO: Generated exposure optimisation

Startups are used to thinking about SEO. Now there is a new layer: Generated Exposure Optimisation, or GEO. This is the discipline of making sure AI platforms see, trust and repeat your story.

That means:

• Strategic placements: Secure founder bylines and expert commentary in outlets that AI training data prioritises.
• Thought leadership: Share unique insights that AI can adopt as reference points.
• Mentions in reports: Contribute data or commentary to analyst or ecosystem reports to expand your footprint.
• Answer optimisation: Position your startup in trusted sources so you appear in AI-curated responses.
• Amplification: Use your owned channels to reinforce earned placements, ensuring humans and machines keep encountering your message.

We have seen this with startups we work with. One e27 story amplified across podcasts and LinkedIn ended up cited in AI responses months later. That is reach and credibility no paid ad could replicate.

Also Read: The story of an ‘accidental entrepreneur’

The amplified value of earned media

Startups cannot outspend incumbents on ads. Earned media is the smarter path. In the AI era, its value compounds.

• Multiplier effect: One profile or byline can spawn thousands of algorithmic mentions.
• Trust transfer: AI inherits the credibility of the sources it quotes. A mention in e27 carries more weight than your own blog.
• Longevity: Paid campaigns expire. A respected article or interview can live inside AI knowledge bases for years.

For founders, this means investing in PR early is not vanity. It is infrastructure.

Where to start

Audit your assets. Do you already have:

• Practical how-to guides you can publish externally?
• Frameworks you can brand and share?
• Case studies with measurable proof points?

If not, build them. Then target one flagship placement this quarter. For startups, even a single strong article can be amplified by AI into enduring visibility.

Final thought

PR is your megaphone. In the AI era, it does more than win human eyeballs. It teaches the algorithms that will advise your customers, investors and partners. For startups, this is not a nice-to-have. It is how you make sure your brand is heard in the conversations that shape your future.

—

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 PR is your megaphone: Why startups must master visibility in the AI era appeared first on e27.

The World Economic Forum’s prediction of a US$3.5 trillion decentralised physical infrastructure network, or DePIN, by 2028 is almost certainly giving investors in each and every DePIN project a warm, fuzzy feeling of self-validation. I have a hard time arguing with a trillion-dollar market cap…but most of these investors are focused on the trees and not the forest, marvelling at individual projects like Helium’s wireless network or Render’s GPU marketplace without addressing the underlying bottleneck: an entire multi-trillion-dollar physical infrastructure sector trying to bootstrap itself onto decentralised networks that are fragmented islands unto themselves.

DePIN’s existing US$27 billion market cap is less than one per cent of the expected US$3.5 trillion opportunity by 2028. And the US$11 billion total value locked across cross-chain bridges today is being exploited every few months with quarterly major hacks, over US$2.5 billion stolen since 2021. We are collectively trying to build the next interstate highway system using nothing more than rope bridges.

The fact is that DePIN, as a sector, is held back by the lack of fundamental infrastructure needed to function at scale. Right now, each individual project is like a country, building its own border controls to admit people from other networks. Consider how these projects actually work: A single transaction might involve sensor data from IoT devices on one blockchain, compute resources on another, data storage on a third, and then finally a payment settlement layer on a fourth chain. At the moment, every bridge is a choke point that requires expensive, centralised relayers, repeated O(N2) security checks, and adds latency and fees that make large-scale use cases economically infeasible.

Core Scientific pivoting from Bitcoin mining to AI infrastructure is an excellent example. A subsidiary, Core Scientific Cloud, is already generating 80 per cent gross margins by repurposing their miners for AI compute! But only because it’s a fully integrated system, 100 per cent centralised with no real dependencies on other chains. If they had to bridge that compute power with storage networks or IoT sensor data streams on separate chains, those margins evaporate.

Also Read: How to launch collaborations that grow communities: A guide for Web3 founders

Recent legislation such as the GENIUS Act aims to provide clearer regulatory frameworks for stablecoins and digital commodities. Jurisdictions that establish regulatory clarity may gain an advantage over regions still working through fragmented approaches. But before we can build that cross-chain DePIN infrastructure, three things need to happen:

First, we need to build cross-chain communication natively, not through wrapped assets or centralised relayers. The Inter-Blockchain Communication (IBC) protocol has seen over US$30 billion in volume across over 12 chains per year with zero exploits as of today’s date. Interoperability is possible on a massive scale if we build it correctly from the ground up.

Second, we need liquidity layers that unify value transfer instead of fragmenting capital across chains. The user capital efficiency of current DePIN networks is orders of magnitude worse than the traditional web because every project forces you to hold a separate wallet with its own native token just to participate. A farmer using DePIN-powered blockchain coordinated irrigation should not need to hold five different wallets and five different tokens to receive payment.

Third, standardised physical-world attestation data. Every DePIN network will have validators attesting to some piece of physical world truthiness, whether it’s Helium validators attesting to wireless coverage, or Filecoin nodes validating data storage proofs. All of that data must be readable and interoperable cross-chain, or we risk building an entire digital infrastructure industry version of the Tower of Babel.

The use cases for DePIN go far beyond speculative opportunity. McKinsey estimates that value unlock through tokenisation of real world assets could reach US$2 trillion by 2030. But DePIN is the last missing puzzle piece to actually bridging the gap between digital tokens and real world utility, but only if we build the infrastructure to support it.

Projects based in jurisdictions with clearer regulatory frameworks are better positioned to capture this value, and recent market performance reflects investor appetite for certainty. Global capital is already flowing into DePIN initiatives in regions such as the United States, where legal clarity and technical capacity offer stronger foundations for scaling.

The counterintuitive part is that DePIN doesn’t need to anoint one or two big winners. Helium doesn’t have to completely beat traditional telecom. Render doesn’t have to bankrupt AWS. Each DePIN network is competing against centralised infrastructure, not other DePIN networks. The real revolution will happen when those networks all interconnect and coordinate in a completely new way to provide emergent value that the legacy systems can never replicate.

Also Read: How AI and blockchain collaborate for a transparent Web3 future

Think of a supply chain where IoT sensors track physical shipments, decentralised compute optimises the route, distributed storage acts as the immutable ledger, and smart contracts auto-execute payments — all coordinated across different blockchains, with no centralised choke points or single points of failure. It’s all technologically possible right now. The missing part is the infrastructure that connects it all.

This is an infrastructure moment. Just as the interstate highway system transformed commerce in the 20th century, a new generation of cross-chain physical-digital infrastructure is now emerging to define the next era of convergence. Recent legislative efforts, such as the CLARITY and GENIUS Acts, illustrate how regulatory frameworks can lay the foundation for growth. What is needed next is the vision to move beyond siloed DePIN projects and begin constructing the connective tissue that can bind them into a unified network.

The reason DePIN is a US$3.5 trillion opportunity isn’t because someone is going to pick three or four winner projects. It’s because each of those thousands of projects will win when we build the infrastructure to connect them all. Every nation and company that understands that is going to own a piece of the future of physical infrastructure. The future will be shaped by the regions that build scalable, interoperable infrastructure first. 

The trillion-dollar question is not if DePIN will revolutionise physical infrastructure. The question is which regions will succeed in building the cross-chain highways to capture that value. With regulatory clarity beginning to emerge in multiple markets, the race has officially begun.

—

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 DePIN’s US$3.5T opportunity: Turning fragmented projects into unified infrastructure appeared first on e27.

As platforms like Meta race toward fully automated ad creation — where AI builds entire campaigns from a single product image and budget — the advertising industry is entering a new phase of hyper-personalisation. This isn’t just about performance; it’s about resonance.

Emerging AI advertising technologies, from neural networks to interest and habit targeting, are evolving rapidly to help brands deliver more culturally relevant, emotionally resonant campaigns. These tools promise more than just efficiency: by analysing nuanced user behaviours and preferences, AI can tailor ads that truly connect across languages, cultures, and contexts — minimising miscommunication and boosting engagement.

Yet, there’s an inherent tension. As AI-generated content becomes more common, critics warn of a growing sameness: templated ads, generic visuals, and a lack of emotional spark — all signs of creativity being flattened by automation. The risk? A sea of blandness where no brand stands out.

But ironically, AI may also hold the key to solving this. Rather than replacing human creativity, AI’s strength lies in precision targeting — allowing brands to craft bespoke experiences for micro-audiences, and freeing up creative teams to focus on strategy, storytelling, and emotional nuance. When used right, AI doesn’t dull the message — it sharpens the delivery.

The cross-cultural advertising challenge

Too often, “localised” ads are just translated versions of global campaigns. While technically correct, they often miss the cultural mark — either failing to resonate or, worse, turning off the audience completely.

In today’s globalised digital economy, cultural context matters more than ever. According to Google and Bain’s e-Conomy SEA report, 72 per cent of Southeast Asian consumers expect brands to personalise communication based on culture, not just demographics. A well-placed emoji or influencer-style callout might work in Vietnam or Thailand, but that same message could fall flat or feel inappropriate to a CIS audience.

Meanwhile, demand from Russian-speaking consumers is rising fast. In the first half of 2024, Yango Ads data reveals there were 570 million tourism-related search queries in Russian-speaking markets. 23 per cent of “Travelling to Asia” queries focused on Thailand, with destinations like Phuket seeing surging interest. Yet many Southeast Asian hotels, retailers, and restaurants still rely on translated materials rather than culturally adapted campaigns.

The consequence: ad dollars that fail to convert because the message feels wrong.

Also Read: Storytelling: A humane way to advertise your startup

AI’s new role from translator to cultural interpreter

Newer neural network–based systems can generate dozens of ad variations in multiple formats, adapting tone and imagery to different audiences.

By leveraging large language models (LLMs), AI tools can quickly spin up many creative variations across formats. The idea is to better match audience behaviour and context, though the quality still depends on human oversight.

And for campaigns targeting Russian-speaking audiences, native language support is baked in — helping APAC brands communicate with emotional fluency, not just functional grammar.

The potential is significant, but the real value depends on how well these tools are applied:

• Early studies suggest that AI-optimised campaigns often outperform traditional approaches on both efficiency and conversions, though results vary by sector and execution.
• Compared to campaigns without any AI optimisation, they deliver 17 per cent more conversions on average.
• The system even analyses visual content — identifying which image elements attract the most attention and automatically enhancing those creatives, all while preserving brand identity.

This marks a broader trend in AdTech: creative is no longer just designed, it’s trained.

Context-aware targeting: not just who, but how they think

Tone is only half the equation; precision targeting is the other. Beyond demographics, effective campaigns are increasingly shaped by long-term interests (for example, wellness travel or boutique hotels) and short-term behaviours (like last-minute bookings or halal dining searches).

For example, a Phuket hotel can target users who’ve recently searched for eco-stays or who show a pattern of browsing spa retreats. An F&B brand could target Russian-speaking tourists actively seeking Japanese cuisine or healthy dining options.

The engine behind this? First-party data. Yango Ads data reveals that 33 per cent of search queries about travelling to Asia are about Thailand in Q1 2025 — more than any other country.

With this kind of behavioural insight, brands can avoid broad-stroke messaging and instead build micro-targeted creative designed for intent-rich audiences.

Also Read: Why building user communities is far better than paid advertising

The commercial payoff for APAC brands

Southeast Asia is quickly becoming a global nexus for cross-cultural consumer flows, from outbound Chinese travellers to inbound Russian tourists, Indian remote workers, and more. But most regional brands don’t have large in-house localisation teams or endless creative bandwidth.

This is where AI becomes an equaliser. It allows smaller players — boutique resorts, family-run F&B chains, or regional e-commerce sellers — to scale creative adaptation without scaling headcount.

But creative targeting isn’t the only underused growth lever. Even seemingly passive user actions, like taking a screenshot, can signal high intent. Retail and travel apps that recognise when a user captures content (like a hotel listing or restaurant menu) can turn that moment into action: prompting a share, follow-up, or even a referral.

Even passive actions, like taking a screenshot, can signal strong intent — a reminder that engagement doesn’t always look like a click. They show what users care about, not just what they tap on.

This kind of screenshot-driven engagement, when paired with AI-personalised ad creatives and behaviour-based targeting, creates a loop of contextual relevance that drives higher conversions and loyalty, especially in mobile-first, socially driven markets like Southeast Asia.

That said, AI isn’t perfect. Generated content, particularly in non-English languages, still benefits from human review. Local context and linguistic nuance can’t always be assumed — even by the smartest models.

Still, the strategic opportunity is clear: while generic automation creates a risk of sameness, smart automation enables uniqueness. When used right, AI doesn’t just make campaigns more efficient, it makes them more emotionally precise.

The next frontier

The future of AI in advertising isn’t about removing the human, it’s about empowering it. Emerging tools for creative optimisation, behavioural targeting, and even intent tracking free up marketers to focus less on churn and more on meaning: what their message feels like, how it lands, and whether it resonates across cultures.

And that might be the most human outcome of all.

—

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 How to avoid cultural misfires: The rise of emotionally intelligent ads appeared first on e27.

As artificial intelligence (AI) continues to transform industries, businesses must find ways to leverage its potential to stay competitive, innovate, and improve efficiency. However, integrating AI into an organisational strategy requires thoughtful planning, a clear vision, and a phased approach.

This article outlines a step-by-step guide on how organisations can successfully incorporate AI into their strategic framework to drive long-term growth and success.

Define clear objectives and goals

The first step in integrating AI is to define clear objectives that align with the organisation’s overall business goals. Without a clear understanding of why you’re implementing AI, it’s easy for efforts to become fragmented or misaligned. Consider the following questions:

• What specific problem is AI solving?
• How can AI help improve existing processes?
• What measurable outcomes do you hope to achieve, such as reducing costs, improving customer experience, or increasing productivity?

By setting clear, achievable goals, you ensure that AI implementation is focused and measurable, helping to build momentum and demonstrating value early on.

Assess your current capabilities and infrastructure

AI integration requires a robust technological foundation. Assess your organisation’s current data infrastructure, software tools, and workforce capabilities. Consider the following:

• Do you have the necessary data? AI thrives on data, so having quality, well-organised data is essential.
• Is your IT infrastructure capable of supporting AI tools, such as cloud services, machine learning platforms, or processing power?
• Does your team have the skills required to develop and manage AI systems, or will you need to invest in training or hiring new talent?

Understanding your current capabilities will help you identify gaps and prioritise investments to ensure your organisation is ready for AI integration.

Choose the right AI tools and technologies

AI is not a one-size-fits-all solution, and selecting the right tools and technologies is critical for success. Depending on your objectives, you might explore various AI solutions such as:

• Machine Learning (ML): For predictive analytics, recommendations, or optimising business processes.
• Natural Language Processing (NLP): To enhance customer service with chatbots, sentiment analysis, or automated document processing.
• Robotic Process Automation (RPA): To automate repetitive tasks and free up employee time for more value-added work.
• Computer vision: To interpret and process visual data, useful in industries such as manufacturing, healthcare, or security.

Selecting the appropriate AI tools requires careful consideration of your organisation’s specific needs, industry, and available resources.

Foster a data-driven culture

AI relies heavily on data, so fostering a data-driven culture is essential for success. This involves not only collecting and storing data but ensuring that it is clean, structured, and easily accessible. Encourage departments across the organisation to make data-driven decisions, and promote collaboration between data scientists, business leaders, and domain experts.

Also Read: How is AI transforming the future of cancer diagnosis

To build a data-driven culture, invest in data literacy programs for your team and provide tools that make data analysis easier. When employees across the organisation embrace data and AI insights, the benefits of AI integration will multiply.

Start small with pilot projects

Rather than implementing AI across the entire organisation all at once, start with smaller, manageable pilot projects. A focused pilot allows you to test AI’s effectiveness on a smaller scale before committing to larger, organisation-wide changes. Pilot projects can:

• Demonstrate AI’s potential value and feasibility.
• Help identify any challenges or obstacles that need to be addressed.
• Provide insights that will inform broader AI integration efforts.

For example, you might start by using AI for customer service automation or inventory management before expanding its use to other areas of the business.

Invest in employee training and change management

Integrating AI can be a significant cultural shift for many organisations, as it may change how employees perform their tasks or even the roles they occupy. To ensure a smooth transition, invest in training programs to equip your team with the necessary skills to work with AI tools. Offer training in areas such as:

• Understanding AI concepts and their applications.
• Using AI-powered tools effectively in daily tasks.
• Analysing AI-generated insights and making data-driven decisions.

In addition to training, prioritise change management to help employees adapt to new workflows and technologies. Communicate the benefits of AI integration clearly and show how it can enhance their work, rather than replace it.

Monitor, evaluate, and optimise

Once AI solutions are deployed, ongoing monitoring and evaluation are essential to ensure their effectiveness. Establish key performance indicators (KPIs) to track the success of AI initiatives, such as cost savings, productivity improvements, or customer satisfaction scores.

AI systems also require continuous optimisation to adapt to changing conditions. Regularly review the AI models to ensure they remain accurate and relevant. This might involve updating algorithms, retraining models with new data, or incorporating feedback from users.

Ensure ethical AI use and data privacy

As AI plays an increasingly central role in decision-making, it is critical to prioritise ethical considerations. Implement policies and frameworks that ensure AI systems are transparent, fair, and unbiased. Consider the following ethical principles:

• Bias mitigation: Regularly audit AI models for biases that may result in unfair or discriminatory outcomes.
• Transparency: Make AI decisions explainable and understandable to stakeholders, especially when they impact customers or employees.
• Data privacy: Comply with data protection regulations, such as GDPR or CCPA, and ensure that AI tools handle personal data responsibly and securely.

By focusing on ethical AI practices, you help build trust with customers, employees, and other stakeholders while minimising risks associated with AI implementation.

Also Read: Unlocking a sustainable future: A new model for green building management

Scale gradually and continuously innovate

Once initial AI projects have been successfully implemented and refined, consider expanding the use of AI across other areas of the business. Scaling should be done thoughtfully, with continuous innovation and adaptation to new technological developments and business needs.

AI is a rapidly evolving field, so staying up-to-date on new advancements and opportunities is crucial. Encourage experimentation and innovation within your organisation to unlock new AI use cases that drive value.

Conclusion

Successfully integrating AI into your organisational strategy is a multifaceted endeavour that requires a clear vision, careful planning, and a willingness to evolve. By defining clear objectives, assessing current capabilities, investing in the right technologies, and fostering a data-driven culture, your organisation will be well-positioned to harness the power of AI.

Remember, the integration of AI is a continuous journey of learning, optimisation, and adaptation. With the right approach, AI can become a powerful driver of growth and innovation for your organisation.

—

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 December 2, 2024

The post Strategies for effectively integrating AI into your organisation appeared first on e27.

The financial landscape of Southeast Asia (SEA) is on the cusp of a dramatic shift. By 2030, a staggering 79 per cent of the fintech narrative will be written by Millennials and Gen Z, making this a significant redefinition of what the future generation will expect from their financial partners. These generations aren’t just digital natives, but they are also value-driven, experience-focused, and deeply discerning.

Therefore, for fintechs to stay relevant, it’s no longer enough to just digitise the traditional products, but reimagine finance from a perspective that is relatable through empathy, clarity, and a deep understanding of how these users think, feel, and act.

Why innovation alone isn’t enough to close the widening empathy gap

Much of fintech innovation to date has centred on making complex financial products more accessible through technology. But despite this progress, the core challenge of an empathy gap that outlines what users say they want, and how they actually behave.

This is most evident in recent research that revealed Gen Z and Millennials seeking a “trifecta” of “money, meaning, and well-being”, or the equivalent of financial tools that support their goals, not just bank accounts. Yet one in four Gen Zs aren’t saving, investing, or insuring enough, suggesting a disconnect between their aspirations and their current financial actions.

The reality is that many offerings today still focus on features over feelings. So how do we move beyond purely transactional relationships and truly resonate with the unique needs of these generations, all while navigating an increasingly crowded market

Also Read: The fintech ‘Wild West’ in Southeast Asia is over and maybe that’s a good thing

Guided simplicity that is designed for confidence, not complexity

To bridge this empathy gap, fintechs must transform their approach from mere digitisation to profound user-centricity. It’s no longer enough to just offer educational content and hope users will piece everything together. The responsibility now lies with fintechs to embed financial literacy and confidence directly into the user experience itself.

This means embedding real-time, personalised guidance into the solution, rather than relying on passive content or dense dashboards. Smart designs that gently nudge users towards positive outcomes like showing the potential benefits of saving a little more, or illustrating trade-offs in an actionable way, have proven to be non-disruptive and less overwhelming. 

Visual clarity, bite-sized insights, and contextual suggestions can shift a user from confusion to confidence. When done right, users report not just better understanding but reduced stress and more consistency in their pursuit of their financial goals. 

Meeting the emotional need by moving from utility to experience

Today, we’re also in the ‘experience economy’ whereby Gen Z’s prioritise solutions that simplify their lives, and Millennials expect excellence in customer experience. Both expect digital products to be highly responsive to their personal context and goals. 

For fintechs, this translates into a need for agile, user-centric development that incorporates emotional design and continuous feedback. The challenge isn’t just solving current pain points, but also anticipating future issues, before it happens. Every part of the journey, from onboarding to notifications, should quietly reinforce confidence. Whether it’s a message about a successful transaction or a reminder to save, each touchpoint is an opportunity to build trust and signal progress.

Doing so would align to an emerging insight of younger users wanting to feel in control of their finances, not just track it. This means using everyday language,, and ensuring that interfaces don’t overwhelm with jargon or data overload. 

Also Read: How the global growth of fintech defies age and gender

Solidifying the relationship through trust, transparency, and alignment

Ultimately beyond the user experience and education, establishing, building and sustaining trust with users is key for longevity. For today’s users, trust is built on transparency and shared values — does it serve my long-term goals? Does it align with how I see the world? Is it honest about what it offers and genuine about wanting to help me? 

For fintechs, this means rising to the challenge of designing products that prioritise ethical clarity, not just in how fees, terms and products are explained, but in how the product itself supports the user’s broader well-being. That means using clear, everyday language, being upfront about trade-offs, and removing hidden catches. When a product says, “This is a safe place for your spare cash,” and then proves it with behaviour, it shifts the tone from persuasion to partnership. That’s when users feel informed, respected, and genuinely understood—not just marketed to.

More importantly, trust is built through consistency. When products behave the way users expect and reflect a commitment to their financial success, not just conversion metrics, is when you’ve truly convinced them to stay with you for the long term. 

So what does the future need?

Looking ahead, fintechs that lead with empathy will lead the market. In a landscape where the finish line constantly moves, the true differentiator is not just functionality, but emotional resonance. 

This requires more than feature innovation or hype. It requires a consistent culture of listening, learning and co-creation with the very users we aim to serve. It’s about recognising that financial decisions are deeply personal and shaped by context, mindset, and emotion. 

As Millennials and Gen Z continue to reshape the fintech ecosystem, success will belong to those who don’t just build for them, but build with them. In doing so, these shared experiences will not only support their wallets, but truly resonate with them and their aspirations.  

This article has not been reviewed by the Monetary Authority of Singapore.

—

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 The empathy gap? How fintech can truly speak to Gen Z and Millennials appeared first on e27.