Tag: T shirts

Bitcoin climbed 5.08 per cent over 24 hours to US$81,106.29, outpacing the broader crypto market’s 4.83 per cent advance to a US$2.72 trillion total capitalisation. The move did not happen in a vacuum. Bitcoin now trades with a 97 per cent correlation to the S&P 500 and an 86 per cent correlation to gold. Those figures reveal a macro-driven repricing rather than a crypto-specific breakout. When digital assets, equities, and bullion rise together, traders are responding to shifting expectations about Federal Reserve policy and global liquidity.

The main catalyst arrived from a dovish turn in rate expectations. Federal Reserve Governor Christopher Waller signalled he would be inclined to support holding interest rates steady if inflation data improved. Before those remarks, markets priced in a 63 per cent chance of a September rate hike. Afterward, that probability fell to around 50 per cent. A pause in US-Iran tensions added further relief by easing oil price fears and removing a recent inflation headwind. The White House also contributed dovish comments, reinforcing the message that the Fed would not tighten aggressively.

With Treasury yields no longer threatening to surge, capital flowed back into equities and high-beta assets like Bitcoin. This repricing pushed Bitcoin into near lockstep with the S&P 500, confirming its current role as a macro-sensitive asset. The immediate trigger for the next move comes from the August Non-Farm Payrolls report due on September 4. Soft jobs data would strengthen the dovish case and could extend gains. A strong report would revive fears of a rate hike and put pressure on the entire complex.

The advance also received a powerful boost from a short squeeze in the derivatives market. Traders liquidated over US$415 million in Bitcoin short positions over 24 hours, with US$164 million of that total vanishing in just four hours. A separate measure showed Bitcoin liquidations hitting US$271 million in 24 hours, with shorts accounting for 95 per cent of the total, a 425 per cent surge.

Forced buying by traders covering bearish bets created a feedback loop that accelerated the price climb. This dynamic explains the speed and intensity of the move. The macro news lit the fuse, but leverage did the heavy lifting afterward. Crowded bearish positions unwound in a cascade, and each wave of forced buying pushed prices higher. That mechanical accelerator is not the same as organic spot demand, and it can reverse quickly if momentum stalls.

Also Read: Bitcoin slipped below US$80,000, so why are traders still betting on US$82,000?

Sentiment indicators reflected the sudden shift. The Fear and Greed Index jumped to 78, signalling greed and overheated conditions. Influential market voices added fuel to the fire by declaring a crypto supercycle, which fed retail fear of missing out. Intense bullish social media narratives reinforced the move. This kind of euphoria often precedes volatility because latecomers chase the upswing after the initial institutional move.

The derivatives market will show whether leveraged speculation is stabilising or building toward another flush, as reflected in funding rates and open interest changes. The squeeze pushed prices higher, but the market now needs sustained spot buying and exchange-traded fund inflows to confirm the move as more than a one-day event.

From a technical standpoint, Bitcoin broke above the US$78,000 resistance with high volume, a bullish signal. The next major hurdle sits in the US$83,000 zone. That level aligns with long-term holder supply and previous rejection points. If Bitcoin holds above US$80,000, especially above the US$81,000 breakout level, the path could extend toward US$83,000 to US$85,000.

A clear break above US$83,000 would open the way toward the Fibonacci extension near US$86,500. If Bitcoin fails to hold above US$78,000, the surge likely came primarily from short covering and could fade. In that scenario, a pullback toward the US$78,000 to US$76,000 support zone would be the next test. The market needs a weekly close above US$81,000 to confirm the breakout and attract longer-term buyers.

Exchange-traded fund flows remain the key institutional signal. On September 2, spot ETFs recorded a net inflow of US$101 million. That is a positive sign, but one day does not establish a trend. If spot ETF inflows re-accelerate after the squeeze, they would provide fundamental support above US$80,000 and reduce the risk of a sharp reversal. Without that follow-through, the rally could lose steam once the forced buying from liquidations ends.

The market has seen this pattern before. Macro news triggers a spike, shorts get squeezed, and then the price drifts without new organic demand. The difference this time is the broader macro backdrop. If the Federal Reserve truly pivots to a dovish stance, the liquidity environment could support a longer rally. If the dovish signals prove temporary, the squeeze alone will not hold Bitcoin above US$80,000.

Also Read: Can Bitcoin defend the critical US$76,500 foundation zone before the September 11 inflation data triggers another massive liquidation cascade?

The near-term outlook is bullish but fragile. Bitcoin’s surge rests on two pillars: a macro reprieve and a violent short squeeze. The macro reprieve is real but depends on upcoming data. The August jobs report on September 4 will be the first test. Inflation data and Fed communications will follow.

Any hawkish surprise could unwind the rate cut expectations that underpinned this move. The second pillar, the short squeeze, has already spent much of its force. More than US$415 million in bearish positions are gone, and the Fear and Greed Index at 78 suggests the easy gains from sentiment reversal are behind us. The next leg higher requires spot buyers to step in with conviction.

From my perspective, Bitcoin’s 5.08 per cent move is a legitimate macro-driven breakout, but it is not yet a confirmed trend change. The 97 per cent correlation with the S&P 500 suggests this is a liquidity trade rather than a crypto-specific narrative. The US$101 million ETF inflow on September 2 is encouraging but insufficient on its own.

The real test is the US$83,000 resistance. If Bitcoin clears that level with sustained volume and ETF inflows accelerate, the path to US$86,500 becomes realistic. If Bitcoin rejects US$83,000 and falls back below US$78,000, the rally will look like a short-covering episode that ran out of fuel.

As I said, the next few days will reveal whether this is the start of a new leg or just a powerful bounce within a range. Watch the August jobs report, watch ETF flows, and watch how Bitcoin behaves around US$80,000 to US$81,000. Those signals will tell you more than any single price print.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post Bitcoin just broke US$81,000: The real reason is not what you think appeared first on e27.

In the world of AI-powered content creation, ChatGPT Canvas is a game-changer. Unlike traditional ChatGPT interactions, where multiple prompts and re-prompts are necessary to fine-tune content, ChatGPT Canvas introduces a new, interactive interface that allows direct text manipulation, structured editing, and enhanced formatting options. This guide explores the benefits, key features, and practical applications of ChatGPT Canvas for content creators, marketers, and business professionals.

The challenges of traditional ChatGPT editing

One of the biggest challenges with using standard ChatGPT or other AI tools for content writing is the difficulty in editing specific sections of generated text. Users often find themselves repeatedly prompting and refining responses to achieve a satisfactory output. This process can be time-consuming and inefficient, particularly when working on long-form content such as articles, blog posts, or marketing copy. Since the AI generates text as a complete block, adjusting a single section often requires rewriting the entire response or editing the entire block part by part manually or via prompts. That may however introduce issues such as with keeping a consistent tone, style, or format throughout the content.

What is ChatGPT Canvas?

ChatGPT Canvas is an advanced interface that simplifies content generation, editing, and refinement. It allows users to:

• Edit text directly within the AI-generated response
• Adjust content length for different formats
• Modify reading levels to suit various audiences
• Format content using markdown for better readability
• Suggest edits and improvements with built-in recommendations

Also Read: The scarcity mindset is killing creativity, not AI

Key features of ChatGPT Canvas

• Interactive editing

Users can click directly into the generated text to make edits, eliminating the need to regenerate the entire response. This feature is particularly useful when refining specific sections of an article or making minor adjustments to wording and tone. Unlike traditional ChatGPT, where users must copy the generated text, paste it into another document, and manually edit, ChatGPT Canvas allows seamless in-place modifications without disrupting workflow.

• Content length adjustment

ChatGPT Canvas allows users to modify the length of their content effortlessly. Whether expanding a script for a longer video or condensing a post for social media, users can fine-tune the content to meet their needs. This feature eliminates the frustration of starting over due to content being too long or too short, making it ideal for those who need flexibility in their writing.

See for example before and after:

• Reading level adaptation

Different audiences require different levels of complexity in writing. With ChatGPT Canvas, users can adjust the reading level to cater to:

• Middle school students
• General web readers
• Academic or professional audiences

This feature ensures that content is accessible and appropriate for its target readers. By simplifying or elevating the language, users can tailor their messaging to resonate with their intended audience without needing to rewrite the entire content manually.

• Emoji and formatting support

For users who want to enhance engagement, ChatGPT Canvas supports:

• Automatic emoji integration
• Bold and italic text formatting
• Headings and subheadings
• Markdown support for web publishing

These features are particularly useful for social media posts, blog formatting, and digital marketing content. Formatting is essential for readability, and with ChatGPT Canvas, users can quickly transform plain text into a visually appealing format with structured elements.

• Version control and undo feature

One of the standout features of ChatGPT Canvas is version control. Users can:

• Restore previous drafts
• Undo changes without affecting the entire document
• Modify individual sections instead of re-prompting the AI

This makes content revision more efficient and user-friendly. Having the ability to track changes and revert to earlier versions ensures that no valuable content is lost during the editing process.

• AI-generated images

ChatGPT Canvas allows users to insert AI-generated images into their content. While the initial image quality may vary, refining prompts can significantly improve the output. This is useful for enhancing blog posts, articles, and marketing materials.

Pro tips: At times, instead of asking it to generate an image, I ask it to propose a few image and explain why.

Also Read: How creativity, commerce and AI collide in mid-2026 marketing mix

Areas and applications of ChatGPT Canvas

• Content marketing and social media

ChatGPT Canvas is ideal for creating structured marketing copy, including:

• Facebook and Instagram ads
• Email campaign templates
• Social media captions and descriptions
• Product descriptions and landing page copy

The ability to quickly generate multiple versions of ad copy allows marketers to test different messaging approaches efficiently.

• Blog and article writing

For writers and bloggers, ChatGPT Canvas streamlines the content creation process by:

• Generating structured outlines
• Allowing easy edits within the interface
• Improving formatting for web readability
• Supporting markdown for seamless publishing

These features make it an excellent tool for drafting, refining, and finalising long-form content.

• Scriptwriting and video content

YouTubers and video content creators can use ChatGPT Canvas to:

• Generate structured video scripts
• Adjust script length for different formats
• Enhance clarity and readability with formatting tools

This feature simplifies the scriptwriting process and allows for easy collaboration with editors and co-creators.

The takeaway

ChatGPT Canvas is a powerful tool that enhances AI-generated content creation. By allowing direct editing, structured formatting, and content customisation, it significantly improves efficiency and quality. However, to maximise its potential, users should combine AI-generated content with personal insights, real-world examples, and authentic storytelling, seeing it as an aid rather than a replacement and engage in the mass generating of content that lacks depth, originality, and personal touch, ensuring that the final output resonates with audiences and maintains credibility. This ensures the content remains engaging, credible, and unique.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post AI and human creativity: How ChatGPT Canvas bridges the gap appeared first on e27.

Claude dashboards have become a kind of “AI flex” on Instagram recently, especially among founders, marketers, finance people and operators.

The biggest reason is that dashboards suddenly became very easy to make and very easy to share. Claude Code can now turn data or work from a session into an interactive Artifact: charts, tables, filters, KPI cards, timelines, and so on, and publish it as a webpage. Anthropic specifically promotes dashboards as one of the uses for Claude Code Artifacts.

That’s a meaningful shift. A year ago, a founder wanting something like this had two options: pay a developer to build it, or live with a spreadsheet nobody opens. Now the barrier is closer to zero, which is exactly why it’s spreading so fast. It’s not that the underlying analysis got smarter. It’s that the packaging got trivially easy.

And dashboards are perfect social-media content. Compare these two:

“I asked AI to analyse my business,”

versus a screenshot showing:

• REVENUE RM183,420 (US$45,368) up 18.4 per cent
• CONVERSION 3.8 per cent up 0.7 per cent
• TOP PRODUCT: Necklace

charts, graphs, customer segments, forecasts.

The second one looks like someone built a sophisticated internal software system. That’s highly shareable even when the underlying data analysis isn’t particularly complicated.

Dashboard does not equal business intelligence. A beautiful dashboard made from bad data is just a beautiful way of making a bad decision. If the revenue figure is wrong, or the “top product” metric is counting returns as sales, none of that shows up on the screenshot. It just looks confident. That’s the trap: the format signals rigour whether or not the underlying numbers earned it.

Also Read: In Southeast Asia, going global used to mean picking the biggest market, that logic is already dead

For a business, I’d rank the value like this:

• Reliable underlying data
• Correct KPIs
• Useful business questions
• Analysis and decision rules
• Dashboard design

Instagram naturally makes dashboard design look like the important part because that’s what photographs well. Nobody posts a screenshot of “we finally reconciled our SKU-level cost data,” even though that’s usually the unglamorous work that makes everything above it trustworthy.

For something like a jewellery business, for example, I’d much rather have a relatively boring dashboard that tells you sales, gross profit, SKU sell-through, inventory ageing, booth ROI, customer basket size, channel profitability, and flags where money is being lost, than a gorgeous 20-chart dashboard nobody actually uses. One of those tells you to reorder before you run out of your best-selling piece. The other just looks good in a founder’s story highlights.

So the trend itself is useful, but the opportunity isn’t “I should make a Claude dashboard too.” It’s “I can now build a lightweight management system without paying someone to develop one from scratch.” That’s much more interesting, and it’s the part that doesn’t screenshot nearly as well.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post The AI dashboard gold rush: Beyond the pretty charts appeared first on e27.

Here’s a test you can run on your own organisation. Think of the last genuinely important decision your team made. Now ask: did the data shape that decision, or did someone make the call and then reach for the data to justify it?

If you’re honest, it’s usually the second one. And that tells you something most “data-driven” organisations don’t want to hear: they’re not data-driven at all. They’re report-driven. They’ve gotten extraordinarily good at producing and consuming data, and they’ve never actually learned to decide with it.

Seeing is not deciding

We’ve spent a decade conflating two completely different things. One is seeing data, building the dashboard, running the report, watching the metric. The other is deciding with data, letting what you see change what you do, and then committing to that change.

Everything in the modern analytics stack optimises for the first. Prettier dashboards, faster pipelines, more metrics, real-time everything. And it’s all beside the point if the thing the data implies never actually gets done. A dashboard that changes no decision is just expensive decoration.

The appetite to close this gap is enormous, by the way. Salesforce’s 2026 research found that 93 per cent of business leaders say they’d perform better if they could just ask questions of their data in plain language. Read that as what it actually is: a near-universal admission that the data is there and people still can’t easily turn it into action. The bottleneck was never seeing. It’s the leap from seeing to deciding, and no amount of dashboard polish closes it.

Also Read: The environmental ethics of AI should be a product decision, not a sustainability footnote

Why we hide in reports

Report-driven cultures aren’t stupid. They’re rational responses to incentives, and the incentives are backwards.

Producing a report is safe. Making a decision is exposed. If I build you a dashboard and the business goes sideways, that’s not on me, I delivered the data. If I make the call the dashboard implied and I’m wrong, that’s very much on me. So smart, self-preserving people accumulate reports as a way of looking rigorous while avoiding the risky act of commitment. The dashboard becomes a place to hide.

Then there’s the certainty trap. People tell themselves they can’t decide yet because the data isn’t complete, so they ask for more. But more data rarely produces more clarity; it usually just produces more delay and a false sense that certainty is coming. It isn’t. Most real decisions have to be made on a strong-enough signal, not a perfect one, and a culture that waits for perfect is a culture that decides late, every time.

And underneath both: nobody’s accountable for the decision. We measure whether the report shipped on time. We almost never measure whether a clear signal in that report actually changed what anyone did. We’ve instrumented the production of insight and left the use of insight completely dark.

The fix is human, not technical

Here’s the part that should be encouraging. Because this is a behavioural problem, not a technology one, it’s fixable, and fixable without buying anything.

Start every analysis from the decision, not the question. “What will I do differently depending on the answer?” If nothing, don’t build the report. That one discipline kills half the dashboards nobody uses and forces every remaining insight to have somewhere to go.

Also Read: How AI and blockchain could make commerce decisions more accountable

Give people explicit permission to act on strong-enough signals. Name the difference between reversible and irreversible decisions, you can move fast and loose on the reversible ones, and most decisions are more reversible than people treat them.

And close the loop, relentlessly. What did we predict? What happened? What did we learn? Organisations that revisit their decisions build judgement that compounds. Organisations that don’t repeat the same hesitation forever, decision after decision, learning nothing.

The reframe

So stop measuring your dashboards and start measuring your decisions. Not how much data you have, every competitor has data. Not how many reports you produce, reports are cheap. Measure whether decisions actually change because of what the data showed, whether they get made fast enough to matter, and whether your organisation is getting better at making them over time.

The companies that win the next decade won’t be the ones with the best dashboards. Plenty of people will have great dashboards. They’ll be the ones whose people can stand in front of the data and decide, quickly, under uncertainty, and honest enough to check whether they were right.

Your dashboards are fine. That was never the problem. The problem is that you’re all looking at them, nodding, and then not deciding anything.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post The decision discipline: How to turn insights into action appeared first on e27.

Podcast production has exploded. The number of episodes published annually grew from roughly 3.9 million in 2015 to around 29 million in 2023.

Hours of valuable information are shared every day through this long-form audio. Yet despite how much useful information is buried inside podcasts, there still isn’t a comprehensive way to index them.

We have searchable indexes for news and databases for financial filings and academic papers. So what makes podcasts so much harder?

Having worked on indexing podcast content, I think the difficulty comes down to four problems: transcription, fragmented sources, content quality, and speaker identification.

Podcasts are audio first

Before you can make sense of a podcast, you first have to turn it into text. That sounds straightforward, especially given how much speech-to-text models have improved. But transcription is still far from a solved problem when you care about accuracy.

Proper nouns are a particularly difficult problem. Speakers constantly mention company names, product names, people, tickers, and industry-specific terminology that transcription models may not recognise. A company like Lyft, for example, can easily become “lift,” or Claude can become “cloud.”

This is a huge issue, especially when you are trying to build a searchable index. If the company name itself is transcribed incorrectly, the episode may never appear when someone searches for it. You need another layer that understands the context and corrects these transcription errors. The problem becomes even more challenging with different accents, speaking styles, recording environments, overlapping speakers, and poor audio quality.

The news on the other hand doesn’t have this problem. No form of processing is required as it’s already in the text form.

With podcasts, transcription is an additional computational step before indexing can even begin. At a large scale, that becomes a meaningful cost.

The source landscape is fragmented

The second problem is fragmentation. The barrier to creating a podcast is extremely low. Unlike traditional media, where a relatively small group of publications accounts for much of the trusted coverage, valuable podcast content can come from almost anywhere.

This is partly what makes podcasts interesting. You get perspectives, conversations, and expertise that would never appear in traditional media. But it also makes indexing them much harder. With news, there are a relatively limited number of publications that people consistently trust. You can get by with just indexing the major outlets.

Also Read: Asia’s AI trust gap: strong transparency, weak security and unclear data practices

Podcasts work very differently. A valuable piece of information might come from a huge show, a niche industry podcast, an independent expert, or a founder appearing on a tiny podcast with only a few thousand listeners. That means you cannot simply identify a few hundred trusted sources and call the job done. To build a useful podcast index, the coverage has to be dramatically wider. The long tail is not optional. It is often where the most interesting information lives.

That creates a scale problem that is easy to underestimate until you actually try to build it.

Content provenance and noise

The lack of editorial boundaries creates another problem: noise. AI-generated podcasts have become increasingly common. In our own work with financial podcasts, we have seen roughly 15 per cent of the content we encounter appear to be AI generated. Identifying and filtering this content is becoming surprisingly difficult.

I have worked around voice AI since 2023, and I used to think I had a good ear for identifying synthetic voices. I am far less confident today. As voice models improve, identifying AI voices has become a challenge.

There are other forms of duplication too. Podcasters frequently publish clips from other podcasts (reactions). An indexing system has to understand the difference. Is the person speaking actually a guest on this podcast? Is this a clip from another show? These problems can be solved using LLMs these days.

Not to forget, podcast advertising introduces complications. Podcasts increasingly use dynamically inserted ads, meaning the audio file itself can change depending on when or where it is played. Unlike an article sitting at a fixed URL with pretty much fixed content, podcast content is not always static.

Knowing what was said isn’t enough

The final problem may be the most important: speaker identification. Knowing that a statement was made is useful. Knowing who made it is far more useful.

Imagine someone saying that a particular company has an enormous competitive advantage. The meaning of that statement changes depending on whether the speaker is the company’s CEO, a competitor, an investor, a customer, or an independent industry expert.

The words might be identical but the perception will change accordingly. A useful podcast index needs to understand who is speaking and what their relationship is to the subject.

This is one area where modern AI models come in handy. Given enough context, they can often identify speakers, infer roles and resolve ambiguous references.

Also Read: Why Singapore’s AI finance race is now about data, not models

AI changes what is possible

A few years ago, building a comprehensive podcast index would have been theoretically unfeasible. You would need to transcribe millions of hours of audio, fix errors, identify speakers, and continuously process a huge stream of new episodes.

The only probable way would have been to hire troves of human annotators. LLMs and modern speech models change the economics of that problem. For the first time, it is becoming practical to turn podcasts from an audio format people have to manually consume into a structured data source that machines can understand. And that matters because the information inside podcasts is unusually valuable.

Executives explain how they think. Investors discuss their theses. Researchers describe work that may never appear in a paper. Founders talk about their companies in more detail than they would in a press release. Industry experts casually reveal insights buried inside hour-long conversations.

Until now, most of that information disappeared into the podcast feed after it was published. We are finally reaching the point where it doesn’t have to. Now that podcasts can be indexed, the more interesting question is what we can build once all of that information becomes searchable.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post Why podcasts are the next big data revolution appeared first on e27.

Restrooms may not be the most glamorous frontier for robotics, but they are among the most revealing. In malls, airports, hospitals and office towers, they sit at the intersection of hygiene expectations, labour shortages and operational cost pressures. For facilities managers, they are also one of the hardest spaces to keep consistently clean.

Singapore-based Hivebotics is betting that this is precisely where robotics can prove its commercial worth.

Also Read: Why robotics is just entering its prime phase

The company has raised US$6 million in a Series A round led by Vertex Ventures Southeast Asia & India, with participation from Fareast Land Development, part of Farglory Group, and Rigel, an Asian manufacturer of smart and eco-friendly restroom products.

The funding will move Hivebotics’s flagship robot, Abluo, from pilot deployments into volume production. It will also support the expansion of the company’s distributor network across Asia, Europe, the Middle East, and North America, in addition to further develop HiveIntelligence, its software layer that plans, monitors, and verifies cleaning jobs.

Founded in 2021 out of the National University of Singapore by Rishab Patwari and Nguyen Tuan Dung, Hivebotics has spent the past year testing Abluo in more than 20 sites, including hospitals, airport terminals, and shopping malls. The robot has logged close to 10,000 operating hours across 12 months of deployment.

Why restrooms are a serious automation problem

Most cleaning robots in commercial buildings focus on floors. That is useful, but limited. Restrooms are far more complicated: they are wet, cramped, uneven environments filled with fixtures, cubicles, pipes, mirrors, and human traffic. Cleaning them involves more than moving in straight lines across open space.

Abluo is designed to clean a commercial restroom end to end, covering toilets, urinals, sinks and floors. The machine uses a mobile base and an articulated arm to reach around fixtures and under rims. Instead of brushes or cloths, which can transfer dirt and bacteria between cubicles, it relies on high-pressure steam, a targeted chemical jet, vacuum extraction and blow-drying.

The company says the system replaces around 30 minutes of manual work with a five-minute human inspection. That framing matters. Hivebotics is not pitching Abluo as a fully invisible robot worker, but as a way to reduce the most repetitive, unpleasant and labour-intensive part of restroom cleaning while keeping people in the loop for checks and exceptions.

Onboard vision allows the robot to identify fixtures. Its AI system then generates a cleaning route in real time. Each job ends with a scored before-and-after check, giving building operators a record of what was cleaned and to what standard.

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

That audit trail could become an important selling point. In facilities management, cleanliness is often judged by complaints, spot checks or subjective inspection. If robots can produce verifiable cleaning records, operators may be able to manage hygiene more like a measurable service level rather than a best-effort routine.

A labour crunch with regional relevance

The market Hivebotics is targeting is large, but the more immediate driver is labour. Soft facilities management, which includes cleaning, security, catering and related building services, was estimated at US$770 billion globally in 2024 and is projected to reach US$1.23 trillion by 2033, according to Grand View Research.

Within that broader market, restroom cleaning is among the most difficult jobs to staff and retain. Kimberly-Clark has reported annual janitorial turnover of 200 to 400 per cent, a figure that reflects how physically demanding and often undesirable the work can be.

That pressure is especially visible in developed Asian markets such as Singapore, Japan, South Korea and parts of the Gulf, where ageing workforces and tighter labour supply are reshaping service industries. In Singapore, cleaning wages have also been rising under structured wage policies, adding pressure on building owners and contractors to improve productivity rather than simply hire more workers.

For Southeast Asia, the picture is more mixed. Labour costs remain lower in some markets, but major cities are dealing with higher hygiene expectations in airports, hospitals, retail centres and transport hubs. The pandemic also made visible something facilities operators already knew: cleanliness is not just a back-office function, but part of public trust.

“Within three years, teams of robots will do most of the repetitive, labour-intensive work of running a facility,” said Hivebotics co-founder and CEO Rishab Patwari. “We started with restrooms because they are the hardest: wet, cramped, and full of moving parts. Solve that, and the rest of the building follows.”

The ambition is broader than toilets. Hivebotics sees restrooms as a proving ground for what the robotics industry increasingly calls “physical AI”, systems that do not merely process information, but perceive and act in messy real-world environments. For years, robots have performed well in controlled settings such as factories and warehouses. Buildings, by contrast, are less predictable.

The competitive field

Hivebotics is entering a robotics market that has grown more crowded over the past decade. Companies such as Avidbots, Tennant, Gaussian Robotics, Pudu Robotics, SoftBank Robotics, and Singapore-based LionsBot have built machines for floor scrubbing, vacuuming and commercial cleaning tasks. Many are already selling into airports, malls, offices and industrial facilities.

The distinction Hivebotics is trying to draw is depth rather than breadth. While floor-cleaning robots are increasingly familiar in commercial buildings, restroom-specific automation remains a tougher category because of the need to manipulate fixtures, apply different cleaning methods and work in confined spaces. If Abluo can perform reliably across diverse layouts, the company may occupy a more specialised niche than broader cleaning robot makers. The trade-off is that niche hardware can be harder to scale: every new building type, fixture design and operating environment introduces complexity.

From pilots to production

The Series A round suggests investors believe Hivebotics has moved beyond the technical demonstration stage. Vertex Ventures Southeast Asia & India, which has backed companies such as Grab, Nium, FirstCry and PatSnap, is positioning the investment around labour scarcity and non-discretionary demand.

“Restroom cleaning is one of the few labour markets where demand is non-discretionary and supply keeps tightening,” said Chan Yip Pang, Executive Director, Investment, at Vertex Ventures Southeast Asia & India. “In Hivebotics’ key markets, the cleaning workforce is ageing and wage floors are rising, so operators need a way to hold hygiene standards without adding headcount they cannot find.”

The next test will be commercial rather than technical. Robots in facilities management must survive long deployment cycles, conservative procurement processes and demanding service expectations. A machine that works in a pilot still has to prove it can be maintained, supported and justified financially across hundreds of sites.

Also Read: Ropedia raises US$22M to build the data layer for robots that understand the real world

Hivebotics has strengthened its commercial bench for that phase. Vincent Sim, formerly head of Kärcher’s Singapore business, joined the company as Chief Sales Officer in 2025, bringing experience from one of the best-known names in cleaning equipment.

For Singapore’s startup ecosystem, Hivebotics also reflects a broader shift. The city-state has long pushed robotics adoption in public services, logistics and built environments, but hardware companies often face a harder fundraising road than software startups. They need capital for manufacturing, field support and inventory before revenue scales.

If Hivebotics can turn restroom cleaning into a repeatable, exportable robotics category, it will show that Southeast Asian hardware startups can compete in global industrial automation from highly specific starting points. The company’s bet is simple: solve the job few people want to do, in the room every building needs to maintain, and the market will listen.

The post Hivebotics nets US$6M to take restroom-cleaning robot Abluo into volume production appeared first on e27.

Up front: the honest answer is, I don’t think anybody is proposing that. Yet.

I don’t know of any plan to put cybersecurity under state ownership, and I am misleading you if I suggest otherwise.

But at a recent industry discussion, an argument surfaced that gets you surprisingly close to that territory. Furthermore, Bill Gates warnings made me think about the issue further.

At the Singapore Press Club event, the question arose as to who is supposed to pay for keeping you safe, and who is answerable when you aren’t.

(The session ran under Chatham House rules, so I’ll share the thinking without naming anyone.)

Private good versus public good

For decades cybersecurity has been treated as a private good. Your company faces a threat, so your company buys protection, out of your own budget. Simple, and until recently, fair enough.

The argument made at this event however, is that this premise is quietly stopping being true. Cybersecurity, it was suggested, is becoming a public good — and we haven’t caught up to what that means.

A public good, is something whose benefits spill well beyond the person who pays for it. And for now, that’s cybersecurity to a tee. When one company hardens its defences, it doesn’t just protect itself — it removes a stepping stone that attackers would have used to reach everyone that company connects to. Your security is increasingly my security, whether or not I ever meet you.

The comparison that made it click was street lighting. No individual shopkeeper pays to install the lamp post outside their door. The city does, because a dark street is one where crime affects for the whole neighbourhood. Note how it’s done (this is important). The government doesn’t run a street-lighting department that builds the lamps itself. It pays a private company to install and maintain them. Privately delivered, publicly funded.

That’s the model the argument points toward for cybersecurity. Not the state taking over. The state paying, while private firms do the work — because the benefit is shared, so the bill should be too.

Also Read: Singapore’s cybersecurity paradox: Leading in digital, lagging in defense

Why private good is breaking

Today, every company is expected to defend itself against threats that are increasingly beyond any single company’s ability.

One line from the discussion put it perfectly. “I don’t build my own air force. I don’t defend my bank against a foreign special forces unit. When the threat is a nation state, I expect the nation to defend me.”

Yet in cyber, we routinely ask a private company or a small business to hold the line against state-sponsored attackers. And with quantum computing on the horizon, the adversary who will eventually be able to break today’s encryption isn’t some criminal gang. Realistically, it’s a state actor. Asking a company to defend itself against is unrealistic, and yet somehow we’ve normalised the notion.

The strain shows most at the bottom of the market. In Singapore, the government has found that around nine in ten businesses surveyed had experienced a cyber incident in the past year, and the costs when it happens are often severe. But the vast majority of companies aren’t large enterprises. They’re small firms, frequently with nobody whose actual job is cybersecurity. They can’t afford enterprise-grade protection, and increasingly they’re the soft entry point attackers use to reach everyone else. The people who need protection most can afford it least — and their exposure is now everyone’s exposure.

Who pays?

If cybersecurity really is becoming a public good, two questions follow.

The first is: who pays? If the benefit is shared, is it right that each company still shoulders the full cost alone? Singapore already nudges in the collective direction, requiring baseline certification in sensitive sectors like healthcare, using government procurement to demand minimum standards, funding schemes that help smaller firms get covered. None of that is nationalisation. But all of it is the state accepting that it has a stake in security it doesn’t directly own.

The second question came from the floor at the event, and it hung: if cybersecurity is a public good, who is independently accountable when preventable failures expose citizens’ data — the hospital records, the national digital identity, the bank accounts? And what enforceable standards protect public trust before the next breach, rather than after it?

That question didn’t get a clean answer. Perhaps an answer doesn’t exist yet. The gap between the benefits shared, costs private, and accountability is unclear. This is the space into which public policy tends to eventually move.

Also Read: The demand for SMB cybersecurity is inevitable, the supply was never built correctly

What this means for now

Leaders don’t need to wait for the policy debate to resolve to act on what it’s telling you.

If your organisation’s security affects the people and businesses around you then framing it purely as your own private cost could already be an out of date notion. Expect that framing to change: more sector requirements, more security conditions written into contracts, more pressure to prove you meet a standard before you win the work, not after you lose the data.

The organisations that will navigate this passage well are the ones that refrain from treating cybersecurity as a grudging line item and treat it as part of the trust they offer everyone they deal with.

That’s what this shift is really about. When what you’re protecting is no longer just your own information, but the confidence of an entire network that depends on you, security stops being an IT question and becomes a matter of reputation.

So — is Singapore about to nationalise cybersecurity? No. But it is, like everywhere else, edging toward treating it as something we all have a stake in and, eventually, all help pay for.

Recognise it. Position yourself as trustworthy custodians rather than reluctant spenders. You will be the ones still standing when accountability catches up with ambition.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. You can also share your perspective by submitting an article, video, podcast, or infographic.

The views expressed in this article are those of the author and do not necessarily reflect the official policy or position of e27.

Join us on WhatsApp, Instagram, Facebook, X, and LinkedIn to stay connected.

The post Should cybersecurity be nationalised? appeared first on e27.

For much of the past decade, Southeast Asia’s startup story was told through speed. That narrative met its limit in 2025, when regional venture funding fell to a seven-year low, according to a new Southeast Asia Startup Funding Report from DealStreetAsia and Kickstart Ventures.

Investors are no longer rewarding growth stories on faith; they are asking harder questions about controls, board oversight, cash discipline and regulatory exposure, and the clearest evidence sits in climatetech and agritech, where fraud cases have made adaptation-focused deals far harder to close than mitigation ones.

Founders who once treated governance as paperwork before a funding round are now building it as infrastructure. At Singapore’s Eezee, co-founder Logan Tan enforces strict separation of duties on every payment; at Transcelestial Technologies, CEO Rohit Jha maintains open board reporting because enterprise and government clients assess reliability as closely as product performance. Corporate-linked capital is gaining weight for the same reason: strategic investors such as Ayala Corporation and Globe Telecom offer regulatory knowledge and distribution that cash alone cannot buy.

The takeaway for founders: in a fragmented, geopolitically exposed region, trust has become the product being sold to investors, customers and regulators alike, not a box to tick before due diligence.

Read the full story

---

Regional

Hivebotics raises US$6M to scale restroom-cleaning robot Abluo: Singapore’s Hivebotics raised a Series A led by Vertex Ventures Southeast Asia & India to move its restroom-cleaning robot Abluo from pilots into volume production, betting labour scarcity across Asia turns automation into a durable category.

Carsome posts record US$8.3M EBITDA in tenth profitable quarter: Malaysia’s used-car platform Carsome grew Q2 EBITDA 38 per cent year-on-year to US$8.3M, its tenth straight profitable quarter, as retail and financing sales outpaced wholesale volume across Malaysia and Indonesia.

Fintech, DeFi and applied AI define SEA’s 2025 venture discipline: Southeast Asia’s 2025 funding pattern split sharply by sector: fintech held its floor at US$1.3B, DeFi models matured into mainstream infrastructure, and applied AI overtook foundation-model bets, per DealStreetAsia and Kickstart Ventures.

Late-stage deals revive in SEA, early-stage founders still squeezed: Late-stage funding more than doubled to US$2.23B in H2 2025, minting four new unicorns, while seed valuations fell to a median US$2M as investors demand proof before backing early bets.

For SEA startups, distress may surface before the cash runs out: AlixPartners flags four early warning signs for Asian companies — costlier capital, EBITDA-cash mismatches, missed milestones and management churn — as regional insolvencies rose 39 per cent in 2025.

The AI wrapper reckoning has reached SEA’s funding tables: Strip out Kling AI’s US$2.8B round and SEA’s Native AI funding falls to US$1.3B, with deal count down sharply — evidence investors are punishing thin, undifferentiated AI interfaces across the region.

Hashed’s ShardLab invests in StoreHub to build merchant rewards tools: Malaysia’s StoreHub, which processes US$3.5B in annual transactions across 20,000 merchants, will build programmable payment and loyalty products with Hashed’s ShardLab, betting distribution beats pilot-stage blockchain rewards experiments.

DSGCP and Saket Gore buy bback to build a wider recovery brand: DSG Consumer Partners and ex-Himalaya Wellness executive Saket Gore acquired Singapore’s hangover-recovery brand bback from Evo Commerce, betting recovery can stretch from alcohol into travel, fitness and everyday fatigue.

Laters.com raises US$1.5M to expand flexible flight payments: The Singapore-founded travel agency, rebranded from Fly Fairly, raised a seed round led by XBO Ventures to widen buy-now-pay-later and crypto payment options, betting checkout flexibility beats fare discounts.

SEA’s blockchain sector has raised US$6.2B across 1,323 firms: Tracxn data shows Singapore holds 82.5 per cent of Southeast Asia’s cumulative blockchain funding, with crypto financial services the largest 2026 segment and acquisitions far outnumbering IPOs as the region’s main exit route.

Indonesia risks missing the AI boom without a manufacturing pivot: AMRO warns Indonesia could become a lucrative consumer market rather than a producer unless it shifts into tech manufacturing, noting high-tech goods make up just 8.7 per cent of its exports versus 60 per cent in Singapore.

The EU called ChatGPT a search engine. SEA should take note: Brussels’ decision to regulate ChatGPT like a search engine under the Digital Services Act will likely be copied across ASEAN, as regional regulators have imported the EU’s GDPR and AI Act templates before.

Malaysia fines, Singapore funds: two paths to the same digital wave: Malaysia’s e-invoicing mandate fines non-compliant SMEs while Singapore subsidises adoption — opposite tools producing the same outcome, with vendors selling compliance rather than growth now capturing the region’s quieter second digital wave.

500 Global winds down Southeast Asia operations: The US-based VC firm is exiting its SEA presence, marking a significant retreat from a region it helped seed for over a decade, with implications for early-stage funding pipelines across markets.

TaniHub founder’s corruption conviction upheld by court: An Indonesian court upheld the three-year prison sentence handed to TaniHub’s co-founder in a corruption case, dealing a further blow to the once-prominent agritech startup’s legacy.

---

International

Delivery Hero board backs Uber’s US$15B takeover bid: The tie-up, which follows Grab’s US$600M purchase of Foodpanda’s Taiwan business, would double Uber’s delivery footprint and intensify consolidation pressure on regional players competing with DoorDash and Just Eat Takeaway.

Uber is cutting 3,300 jobs, or 10 per cent of its staff: CEO Dara Khosrowshahi’s restructuring will shrink management layers by a fifth and end remote work for most employees, as Uber consolidates its engineering, delivery and robotaxi divisions.

Unacademy sells to upGrad for US$206M, 94 per cent below peak: The all-stock deal values India’s once-hot edtech firm far below its 2021 peak, underscoring how sharply investor appetite for pandemic-era hypergrowth stories has reversed across South Asian markets.

Adobe acquires Indian marketing-workflow startup Rilo: The India-founded startup, which built AI-driven competitor intelligence and campaign tools, will shut down post-acquisition as Adobe folds its six-person team and technology into its enterprise marketing suite.

India’s Jio opens its cloud-PC service to turn old computers AI-ready: Reliance Jio is betting consumers will pay roughly US$11 for two months of cloud computing rather than replace ageing hardware, challenging India’s traditionally strong preference for owning a physical PC.

Medtronic to invest US$700M in Hong Kong’s Cornerstone Robotics: The deal gives Medtronic distribution rights to Cornerstone’s Sentire surgical robot, approved in China, Europe and Singapore, as the medtech group widens its robotic-surgery portfolio beyond its own Hugo platform.

Japan’s NETSTARS, Singapore’s imToken to explore stablecoin store payments: The pair signed a non-binding agreement to bring imToken’s wallet onto NETSTARS’ Stablecoin Pay system, following a Lawson stores trial, aiming to extend stablecoin use from trading into everyday retail.

Amazon’s Zoox extends its robotaxi service to Las Vegas airport: The Amazon-owned firm becomes the only robotaxi operator serving Harry Reid International, picking up near baggage claim, as Tesla, Uber and Waymo prepare to launch competing services in the county.

Larry Page’s flying-car company Pivotal loses its CEO: Ken Karklin departs after four years leading the eVTOL maker; aviation executive Mike Ross takes over on an interim basis as Pivotal prepares to bring its fourth-generation Helix aircraft to market.

OpenAI faces 30 more lawsuits tied to a Canadian school shooting: New complaints allege OpenAI’s leadership, not just its safety team, decided against alerting police to a user’s violent conversations before a shooting in Tumbler Ridge, escalating claims beyond earlier negligence filings.

South Korea’s president warns rate rise is unavoidable in 2026: President Lee’s remarks signal a tightening monetary policy stance that could dampen startup valuations and venture activity across Northeast Asia and ripple into SEA investor sentiment.

---

Semiconductor

NASA-linked, MIT-trained founders’ nSWX raises US$2M for chip packaging: Kuala Lumpur’s nanoSkunkWorkX raised a seed round led by Tin Men Capital for graphene-copper interfaces that improve heat and current flow inside AI chip packages without replacing existing manufacturing lines.

Enflame targets US$908M in Shanghai IPO amid record demand: The Chinese AI chipmaker’s Shanghai listing drew 6,109x oversubscription in online demand, underscoring surging investor appetite for domestic chip alternatives as US export restrictions tighten.

Bluehill leads US$11M seed round in Indian semiconductor startup: The raise signals growing venture interest in South Asia’s chip sector, as investors look beyond established hubs to back semiconductor design talent across the broader Indo-Pacific region.

---

Cybersecurity

OpenAI’s GPT-6 Astra becomes its first Critical-tier cyber model: OpenAI’s new frontier model can reportedly find unknown exploits without step-by-step human direction, prompting stricter jailbreak defences and a US$1B subsidised-access programme for water, power and community-bank defenders.

Should cybersecurity be nationalised?: A Singapore Press Club discussion argued cybersecurity is quietly becoming a public good, like street lighting — privately delivered but collectively funded — even as no government is proposing outright state ownership.

---

AI

Nvidia agrees to acquire Hugging Face for US$12.93B: The deal pulls the open-model hub, used by more than 18 million developers, deeper into Nvidia’s orbit; the chipmaker has pledged to keep the platform hardware-neutral despite the obvious incentive not to.

Meta discounts its new AI model 95 per cent for user data: Meta’s Muse Spark pricing charges a fraction of standard rates to developers who let it train on their prompts and outputs, formalising what Claude Code once obtained by default retention.

OpenAI’s new reasoning technique alarms AI safety researchers: Astra’s “opaque recurrence” processes queries in loops rather than legible steps, prompting warnings from Redwood Research and other safety figures that chain-of-thought monitoring could erode as labs race to adopt it.

Google adds conversational AI voice features to Gmail, Docs and Keep: Gmail Live, Docs Live and Keep Live let subscribers query inboxes and dictate documents in natural language, extending Google’s push to embed Gemini-powered voice tools across its productivity suite.

---

Thought Leadership

Why most AI-driven reorgs are solving the wrong problem: Klarna’s rehiring reversal shows the pattern: 60 per cent of firms cut headcount anticipating AI, but only 2 per cent had AI actually doing the displaced work, a Harvard Business Review survey found.

Who’s building AI for the way Southeast Asia actually speaks?: Nearly 70 per cent of the region’s AI prompts now arrive in native languages, but national models remain uneven — Indonesia and Thailand invested heavily while Vietnam’s PhoGPT team was absorbed by Qualcomm.

You spent fifteen years building guanxi, then nobody picked up: Guanxi, nemawashi and Korea’s approval hierarchy are distinct systems, not interchangeable “Asia relationship-building” — foreign operators who mistake proximity for obligation often discover their network was never as deep as assumed.

The yellow flag problem: most risk functions fail at culture first: Institutions where the CRO reports to the CFO, not the CEO, train risk officers to soften red flags into amber ones — a cultural failure, the author argues, that precedes every technical one.

Vietnam’s new growth engine is built on constraint: Manufacturing captured 65 per cent of Vietnam’s H1 2026 registered FDI, with Samsung, LG Innotek and Viettel betting on semiconductors — but power reliability and technical talent now gate how far the country can climb.

The creator economy is distribution, not marketing: Indonesia’s 2023 TikTok Shop shutdown proved the point: platforms, not brands, own the storefront in creator-led commerce, and businesses budgeting for it as advertising rather than infrastructure are miscounting their real margins.

The Podular future: why AI demands new organisational architecture: Solo operators now match team-scale output, the author argues, but carry key-person fragility that small, three-to-seven-person “pods” of sovereign operators could resolve without recreating corporate bureaucracy.

I stopped hiring. I’m not sure it’s a strategy yet: A Jakarta founder who has run two years without a full-time hire warns that lean, AI-native teams are really a bet on subsidised model pricing — one that could unwind if frontier labs raise prices.

Enabling 22-year-olds to build judgement in the post-AI world: The author proposes a “Forward Deployed Learner” model, pairing students with real company problems years before graduation, arguing AI can accelerate the path to an expert conversation but not replace it.

The MMM barrier didn’t disappear. It moved: AI agents have made marketing-mix modelling trivial to run but not to trust, the author warns, since a 2019 Facebook study found observational methods misjudged true ad lift by a factor of three.

AI agents are outpacing companies’ ability to govern them: Over half of organisations have seen AI agents exceed their intended permissions, a Cloud Security Alliance survey found, as the author argues accountability must be designed into agent workflows before deployment, not after.

What building logistics tech in Sweden taught me about SEA: A Bangladesh-based CTO building for a Swedish logistics platform argues that engineering discipline travels between markets, but implementation — payments, compliance, onboarding — must always stay local.

Why vertical AI will define medicine’s next century: Healthcare is shifting from a product to a platform industry, the author argues, with AI models trained on deep clinical datasets becoming the defensible infrastructure layer beneath fluorescence-guided surgery and biosensor monitoring.

Staying secure in the AI era: the habits we need to rethink: The biggest AI-era security risk may be comfort rather than sophistication, the author writes, urging workers to share only the minimum information needed and verify AI output before trusting it.

The travel eSIM market is moving beyond price per GB: Revenue per gigabyte has fallen 13 per cent since 2023 even as user numbers head toward 215 million by 2028, pushing providers to compete on setup reliability and support rather than data alone.

Why podcasts are the next big data revolution: Roughly 15 per cent of financial podcast content the author’s team encounters now appears AI-generated, one of four problems — alongside transcription, fragmentation and speaker identification — blocking a searchable podcast index.

The AI dashboard gold rush: beyond the pretty charts: Claude-built dashboards have become an “AI flex” on social media, the author warns, but a beautiful dashboard built on bad data is just a persuasive way to make a bad decision.

The decision discipline: how to turn insights into action: Ninety-three per cent of leaders say they would perform better with plain-language access to their data, Salesforce found, yet most organisations remain report-driven rather than decision-driven, the author argues.

AI and human creativity: how ChatGPT Canvas bridges the gap: The interactive editing interface lets writers adjust tone, length and reading level in place rather than re-prompting from scratch, though the author cautions AI output still needs authentic, original input to stay credible.

Bitcoin just broke US$81,000: the real reason is not what you think: A short squeeze, not fresh spot demand, drove Bitcoin’s 5 per cent surge past US$81,000, the columnist writes, with September’s non-farm payrolls report now the key test of whether the rally holds.

Bitcoin slipped below US$80,000, so why bet on US$82,000?: Kalshi traders are pricing a September rebound on seasonal precedent even as Ethereum absorbed a US$367M liquidation cascade, the columnist notes, with the Fed’s September 16 decision the next catalyst.

Can Bitcoin defend the critical US$76,500 zone before September 11?: Rising Brent crude and Treasury yields pressured both Bitcoin and Ethereum, the columnist writes, even as US spot ETFs logged a US$216.7M daily inflow that he reads as a structural demand floor.

 

The post Ecosystem Roundup: Governance is Southeast Asia’s new venture currency appeared first on e27.

In Southeast Asia’s tougher VC market, governance is no longer boring
For much of the past decade, Southeast Asia’s startup story was told through speed: faster user growth, faster market entry, faster fundraising, faster expansion. In 2025, that narrative met its limits.

The region’s venture capital market has entered a more selective phase, according to the Southeast Asia Startup Funding Report for 2025 published by DealStreetAsia and Kickstart Ventures. With startup fundraising falling to a seven-year low, investors are no longer rewarding growth stories on faith. They are asking harder questions about controls, compliance, board oversight, cash discipline, and regulatory exposure.

Also Read: The AI wrapper reckoning has reached SEA’s funding tables

In other words, governance, which was once treated by many founders as an administrative burden to be tidied up before a funding round, has become a competitive advantage.

That shift matters deeply in Southeast Asia, where founders do not operate in a single, harmonised market. They build across economies with different licensing regimes, tax rules, labour laws, data policies, foreign ownership restrictions, and political realities. In a funding winter shaped not only by interest rates but also by geopolitics, the startups that can prove they are trustworthy may find themselves at the front of the queue for scarce capital.

Trust becomes the bottleneck

The clearest evidence of this reset can be seen in climate and agricultural technology.

On the surface, climatetech appeared resilient in 2025. It accounted for 15.4 per cent of total deal volume in Southeast Asia, up from 13 per cent in 2024. But beneath that headline, capital moved unevenly. Mitigation-focused sectors, such as renewable energy and waste management, raised US$563 million across 60 deals. Climate adaptation, by contrast, fell sharply to just 16 deals and US$43 million.

The pain was concentrated in agritech, where deal volume dropped 57 per cent and deal value plunged 79 per cent. The problem was not that Southeast Asia suddenly stopped needing agricultural innovation. Quite the opposite: the region remains highly exposed to food security pressures, extreme weather, and the productivity gaps of smallholder farming.

The issue was trust.

The report notes that investor caution in adaptation was intensified by governance concerns after several high-profile fraud cases in agriculture. For limited partners, those episodes reinforced a simple lesson: even sectors with strong long-term demand can become difficult to back if transparency is weak.

“LPs are now demanding stronger accountability, greater transparency, and more rigorous startup governance standards before reallocating capital to funds,” said Minette Navarrete, President and Managing Partner of Kickstart Ventures.

That demand is flowing down the chain. Venture funds are under more pressure to show discipline to their own backers. Startups, in turn, are being asked to prove that their numbers, contracts, reporting lines and internal controls can withstand scrutiny.

Also Read: For Southeast Asian startups, distress may show up before the cash runs out

In a looser market, gaps in governance could be patched later. In this market, they can stop a deal from happening at all.

The founders treating governance as infrastructure

Some founders in the region are already treating governance not as a defensive exercise, but as core operating infrastructure.

At Eezee, a Singapore-based e-procurement marketplace, compliance is built into the company’s day-to-day model. Procurement is a sensitive corporate function: buyers need clear records of who approved what, when, at what price and under which terms. Eezee therefore tracks transactions end-to-end, using real-time dashboards and audit trails across four countries.

Co-founder and CEO Logan Tan describes the company’s approach through clear separation of duties, careful hiring and direct reporting to the board. “We have a strict separation of duties — what I call ‘you can’t eat the food you cook,’” he said, referring to the need for multiple approvals on payments and transactions depending on their value.

Tan said Eezee reports its financial, operational and business metrics to the board “without sugarcoating”. That may sound basic, but in a region where many startups scaled quickly across borders before their internal systems matured, it is not always the norm.

The same principle applies in more technically complex and regulated sectors.

Transcelestial Technologies, which develops laser communications systems, works in areas that intersect with telecommunications, space and defence. For such companies, governance is tied directly to customer confidence. Enterprise and government clients do not only evaluate product performance; they also assess reliability, security, oversight and continuity.

Rohit Jha, Co-founder and CEO of Transcelestial, said the company maintains broad oversight from its board, leadership team and employees, with open sharing of wins, losses and operational challenges. In sectors where a single misstep can damage trust with regulators or customers, transparency is not a cultural nicety. It is a risk-control mechanism.

Geopolitics enters the investment memo

Governance is also becoming more important because Southeast Asian startups are operating in a more complicated geopolitical environment.

Inflation and interest rates still matter, but investors are increasingly focused on structural risks: US-China rivalry, supply chain protectionism, cybersecurity threats, data localisation rules and fragmented regulation across the region. These factors affect where startups can expand, which customers they can serve, how they source components and whether they can move data or capital across borders.

This is particularly relevant for Southeast Asia because the region is economically connected but politically and legally diverse. A fintech licence in one market does not guarantee an easy path into another. A supply chain that works in Vietnam may face different constraints in Indonesia or the Philippines. A data product that scales in Singapore may need significant changes before entering markets with stricter localisation rules.

Also Read: Late-stage deals revive in Southeast Asia, but early-stage founders remain under pressure

As a result, investors are no longer underwriting only total addressable market and revenue growth. They are also assessing whether management teams can navigate regulation, protect customer trust and adapt to sudden policy shifts.

Navarrete described experienced leadership as the “steady hand on the tiller” in such an environment. That phrase captures the mood of the current cycle. The market is not closed to ambitious startups, but it is less forgiving of improvisation.

Why corporate capital is gaining weight

This also explains the rising importance of corporate-linked venture ecosystems.

In a difficult fundraising market, strategic investors can offer more than money. They can provide distribution, procurement credibility, regulatory knowledge and access to established operating platforms. For startups trying to sell into heavily regulated sectors — banking, telecoms, energy, healthcare or infrastructure — that support can be as valuable as capital itself.

Ayala Corporation, one of the Philippines’ oldest conglomerates, illustrates this model. Its venture arm, Kickstart Ventures, gives the broader group exposure to emerging technologies across areas such as fintech, telecoms, renewable energy and enterprise software. For Ayala President and CEO Cezar Consing, good governance, execution and portfolio selection are not optional extras but table stakes.

Globe Telecom follows a related path through internal venture building via 917Ventures and global strategic investing through Kickstart. Globe President and CEO Carl Cruz said the company looks for technologies that can improve its network, customer experience and internal efficiency.

For startups, such partnerships can act as a form of institutional validation. Transcelestial’s backing from investors including Japan’s NTT Finance and Australia’s Paspalis Capital, for instance, gives it not only funding but also credibility in markets where local relationships and trust matter.

The next phase of Southeast Asian VC

The region’s funding reset is painful, but it is also forcing a healthier conversation about what durable companies look like.

The last cycle rewarded speed. The current one rewards proof. Founders need to show clean reporting, responsible capital use, realistic expansion plans and boards that ask difficult questions. Funds need to show LPs that they can spot not only market opportunity but also operational and governance risk.

Also Read: Fintech, DeFi and applied AI define Southeast Asia’s new venture discipline

That does not mean Southeast Asia’s startup ecosystem has become less ambitious. It means ambition now needs stronger foundations.

For founders, governance is no longer a box to tick before due diligence. It is part of the product they are selling to investors, customers, regulators and partners. In a fragmented and geopolitically exposed region, trust may be the most valuable currency left.

The post In Southeast Asia’s VC reset, governance becomes the new growth story appeared first on e27.

OpenAI has launched GPT-6 Astra, its newest flagship model, pitching it as a step-change in artificial intelligence systems that can not only answer questions, but also operate software, browse the web, write code, analyse data, and complete multi-step professional tasks with limited human intervention.

The model is being rolled out today to a limited set of organisations, before becoming available over the coming days to ChatGPT Plus, Pro, Business, and Enterprise users. It will also be accessible through the OpenAI API and AWS, a distribution path that matters for startups and larger companies in Southeast Asia already building AI into customer support, internal operations, software development, financial services, and logistics workflows.

Also Read: Thailand’s AI startup push gets OpenAI backing through new public-private accelerator

OpenAI said Astra is its “most intelligent and aligned” model to date, built on advances in pre-training, reinforcement learning, and alignment. Stripped of the technical phrasing, the company is arguing that Astra is better at learning from large-scale data, improving through feedback, and following user intent safely.

Greg Brockman, President of OpenAI, framed the launch in unusually sweeping terms. “If we fast forward a couple years, and we look back and say when was it really that AGI was created, I think it’s going to be about this time, and I think it might be about this model,” he said.

That is a big claim, and one the broader industry will scrutinise closely. Artificial general intelligence, or AGI, has no universally accepted definition. But in practical terms, Astra’s significance lies in whether it can make AI agents more useful in everyday work — particularly in areas where previous systems have been impressive in demos but brittle in production.

From chatbots to computer operators

Astra’s headline capability is computer use. OpenAI said the model can carry out multi-step workflows, produce polished documents, spreadsheets, and presentations, create websites, and test whether their features work. It can also navigate across web pages, fill out forms, and move through spreadsheets at high speed.

“Computer use is a particularly important part of what’s new; the model can zip through spreadsheets, fill out forms, and navigate across web pages often at superhuman speed,” Brockman said.

In latency simulations on the offline subset of OSWorld 2.0, Astra achieved higher computer-use performance in about 47 per cent less time per task than GPT-5.6 Sol, OpenAI’s current model. OSWorld is a benchmark designed to test how well AI agents operate computers across realistic tasks, rather than simply generate text.

For Southeast Asian companies, this is where the launch may become commercially relevant. Many businesses in the region still run on fragmented workflows: spreadsheets, web dashboards, PDF invoices, WhatsApp conversations, accounting software, customer relationship management systems, and government portals that do not always talk to each other. A model that can reliably operate across these interfaces could reduce manual work in finance, compliance, procurement, and customer service.

That said, reliability will matter more than raw speed. A model that fills forms quickly but makes quiet mistakes could create new operational risks, especially in regulated sectors such as banking, insurance, healthcare, and cross-border trade. For founders, the near-term question is not whether Astra looks intelligent in a benchmark, but whether it can be trusted with repetitive, high-volume workflows where errors are costly.

A stronger model for developers and researchers

OpenAI is also positioning Astra as its best model for software engineering. The company said it performs better on complex tasks in real codebases and, on DeepSWE v1.1, outperforms GPT-5.6 Sol at approximately 57 per cent lower estimated API cost per task when comparing each model’s highest-scoring setting.

That combination, stronger capability and lower task cost, will be watched closely by startups. Engineering talent remains expensive across Southeast Asia, especially for AI, cybersecurity, fintech infrastructure, and enterprise software companies. Tools that help smaller teams understand large codebases, write tests, fix bugs, or ship features faster could shift how early-stage startups allocate resources.

Also Read: The real difference between OpenAI and Anthropic is what happens when AI gets cheaper

OpenAI cited Canva as one early customer example. According to the company, Astra navigated Canva’s codebase of more than 80 million lines, wrote and analysed over 1,000 data queries, and drew on more than 21 internal knowledge sources to recommend improvements. While Canva is far larger and better resourced than a typical regional startup, the example hints at where AI coding agents are heading: not just autocomplete, but systems that can reason across engineering, analytics, and company documentation.

Astra is also being presented as a scientific research tool. OpenAI said an internal version of the model contributed to ten advances in mathematics and theoretical computer science, with proofs formalised in Lean, a programming language and proof assistant used to verify mathematical reasoning. Astra also scores 98 per cent on FrontierMath Tier 4, a benchmark focused on difficult mathematical problems.

If such capabilities hold up outside OpenAI’s own testing, the implications could extend beyond software companies. Universities, research institutes, biotech startups, climate modelling teams, and semiconductor firms in the region may eventually gain access to tools that can support formal reasoning, literature review, experiment planning, and technical validation. But those gains will depend on pricing, local access, data governance, and the ability to adapt models to domain-specific knowledge.

Cybersecurity becomes both use case and risk

One of the more sensitive parts of the launch is cybersecurity. OpenAI said Astra’s stronger cyber capabilities can help defenders find and patch weaknesses, but also create a need for stronger safeguards. The model meets the Critical threshold in cybersecurity under OpenAI’s Preparedness Framework.

The company said it is strengthening protections against misuse. Through OpenAI Daybreak, it plans to expand access and roll out less restrictive safeguards in the coming weeks for work such as vulnerability validation, malware analysis, and detection engineering.

This will be especially relevant in Southeast Asia, where digital adoption has often outpaced security readiness. Banks, e-commerce platforms, healthtech providers, government systems, and small businesses face rising cyber threats, while cybersecurity talent remains in short supply. AI tools that help defenders test systems and detect suspicious behaviour could be useful. But the same capabilities, if poorly controlled, could assist attackers.

For regulators and enterprise buyers, Astra’s launch will likely reinforce a growing tension: the most capable AI systems are also the ones that require the strongest governance. Companies using Astra for cyber work will need clear audit trails, permission controls, and policies on what the model can and cannot do inside production environments.

Rivals are moving quickly

OpenAI is not alone in trying to turn large language models into capable workplace agents. Google has been pushing Gemini deeper into Workspace and developer tools, while Anthropic’s Claude models have gained traction among companies that prioritise coding, reasoning, and safety. Meta continues to compete through its open-weight Llama models, which appeal to developers and companies seeking more control over deployment. Microsoft, OpenAI’s key partner and investor, is embedding AI agents across its enterprise stack, while AWS is advancing Bedrock and its own agent infrastructure.

In Asia, competition is also intensifying. China’s DeepSeek, Alibaba’s Qwen, and Baidu’s Ernie models have pushed the market on price and performance, while open-source communities are giving startups alternatives to closed US models. For Southeast Asian companies, the choice will rarely be ideological. It will come down to cost, latency, language support, data residency, integration, and whether a model performs reliably on local business workflows.

Astra’s launch suggests the next phase of AI competition will be less about chat and more about execution. The winners will not simply be models that write fluent answers, but systems that can safely complete work across messy digital environments.

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

For founders and operators in Southeast Asia, that could be an opportunity — and a warning. The opportunity is to build new products on top of more capable AI agents, automate back-office bottlenecks, and give small teams leverage once reserved for large companies. The warning is that every competitor will get access to similar tools soon enough.

The question, then, is not only what Astra can do. It is how quickly companies can redesign their workflows, safeguards, and teams around a world where software increasingly uses software on their behalf.

The post OpenAI’s Astra aims to turn AI from chatbot into digital worker appeared first on e27.