Alibaba

ByteDance

CATL

Huawei

Meituan

Tencent

Xiaomi

Most Followed

Market News

Listings (IPO)

Earnings Reports

Venture Roundup

Big Tech

Automotives

Retail and Commerce

Collections

News

Features

Insights

Press Releases

Anta

DeepSeek

Geely

Li Auto

Mixue Bingcheng

Pop Mart

Going Global

Latest

Collaborate

Earnings

Auto

Retail & Commerce

On July 25, Dianzhong Technology accused ReelShort, the short drama platform operated by COL-affiliated Crazy Maple Studio, of distributing multiple unauthorized shows that closely resemble content it owns or licenses. The statement was published on Dianzhong’s official WeChat account.
Dianzhong CEO Chen Ruiqing followed up with three consecutive posts on his personal WeChat, reaffirming the company’s commitment to protecting its intellectual property overseas, regardless of time or cost.
A person close to Dianzhong told 36Kr that eight titles were initially flagged for takedown, but ReelShort ultimately removed only three. The source added that executives from both companies met in person, but failed to reach a resolution.
Dianzhong is not alone. Other major domestic short drama producers, including Heard Island and Maiya Drama, have issued similar complaints against ReelShort.
Wang Li, who oversees international operations at Maiya Drama, posted on social media that ReelShort’s legal head abruptly exited a joint communication group and ceased contact. Wang also claimed that ReelShort’s founder had previously signed a pledge not to infringe on Maiya’s IP but later reneged. “We were left with no choice but to speak out on social media,” Wang wrote.
36Kr reached out to Crazy Maple Studio for comment but received no response as of publication.
<h3>ReelShort versus DramaBox</h3>
ReelShort remains the leading overseas short drama platform, holding a nearly two-year lead over its competitors. Between March and April this year, it ranked as one of the top entertainment apps in the US for over 30 consecutive days, according to app store data.
Its early-mover advantage and localized content strategy are key reasons behind its dominance.
By the second half of 2023, ReelShort had already drawn industry-wide attention by reaching the top of the US iOS entertainment chart. The achievement signaled that overseas short dramas still held significant potential.
ReelShort focused on producing content tailored to local preferences, such as werewolf and revenge genres, while employing local actors and crews. These shows were generally better received than translated Chinese dramas, which now account for roughly half the market, and were better suited to a paid content model.
In 2024, ReelShort generated an estimated USD 300 million in revenue, according to 36Kr.
However, the push for localization comes with tradeoffs, especially in content supply.
User reviews often cite the platform’s slow release pace. While artificial intelligence-translated dramas can yield over 100 new episodes monthly, locally produced series typically release only a few dozen. High labor and location costs in North America and Europe push production budgets to USD 150,000–200,000 per title, or more. The hit rate remains low, with fewer than 5% of titles achieving viral success.
This environment makes copying proven hits seem like a lower-risk, high-reward strategy. One of the shows Dianzhong accuses ReelShort of copying has received 36.3 million views on the Hongguo Short Drama app, considered a moderate success by industry standards.
A US-based director known as A-Ye told 36Kr that plagiarism is widespread in the short drama industry, both in China and abroad. “It’s common to directly translate a hit from overseas and have local writers rework it, or the other way around,” he said. “Sometimes even the actors don’t know the plot until they are on set. That’s how secretive things have become.”
DramaBox, by contrast, built its model around translated content, which is cheaper and quicker to produce. It also poured resources into aggressive ad campaigns. According to 36Kr, DramaBox has surpassed RMB 2 billion (USD 280 million) in revenue and recorded more than 100 million app downloads—double that of ReelShort.
Together, ReelShort and DramaBox control close to 60% of the overseas short drama market, with each platform holding nearly 30%.
<h3>Every advantage counts in a crowded market</h3>
More Chinese short drama firms are now eyeing global expansion. As of March, 237 Chinese short drama apps were targeting international audiences—nearly four times the number from the previous year, according to DataEye. New entrants include Maiya Media’s NetShort and COL’s Sereal+ and UniReel.
But market saturation is becoming a concern. Data from Duanju Zixishi shows that while overseas short drama revenues grew 47.1% from Q3 2024 to Q2 2025, growth sharply declined by 68% in Q2 2025, largely due to slowing user acquisition.
As more players crowd the space, the urgency to capture market share is rising.
In response, ReelShort has increased spending on content and advertising. Sensor Tower data shows the platform generated USD 130 million in in-app purchase revenue in Q1 2025, nearly half its 2024 total, topping income and growth charts across the sector. DramaBox was close behind with USD 120 million and maintained a top ranking in ad spending.
Their strategies are beginning to converge.
DramaBox, originally focused on translated content, is now pivoting toward original local productions that offer better monetization and user retention. The company currently releases more than a dozen new shows each month, though it still trails ReelShort in overall presence.
While more companies explore opportunities in Southeast Asia, the rivalry between ReelShort and DramaBox is fiercest in North America and Europe. From August 2023 to June 2024, revenue per download (RPD) in North America averaged USD 4.7, while Europe recorded USD 2.3, both well above the industry average of USD 2 and significantly higher than Southeast Asia’s USD 0.7, according to Diandian data.
With ReelShort now facing public plagiarism allegations, its business challenges are under greater scrutiny: low hit rates, rising production costs, and a revenue model that remains unproven at scale. Industry sources told 36Kr that, unlike DramaBox, which now generates enough overseas income to subsidize its domestic business, ReelShort remains in the red.
As companies like Dianzhong move to defend not just their IP but also their market position, the pressure is mounting. In an increasingly saturated landscape, every slice of the pie counts more than ever.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3397063436716165">article</a> was written by Lan Jie for 36Kr.

ReelShort hit with IP theft claim as short drama rivalry escalates

Entertainment

Video

Creative

Content

China

… TSMC’s automotive-grade four-nanometer process will only be ready this year.
Automotive-grade standards are stricter across the board, not just at TSMC’s foundries, but also in packaging and testing. “Testing for automotive chips costs three times as much because it’s done three times over,” one chipmaking veteran said.
In general, automotive-grade chips lag their consumer counterparts by two years. The more advanced the node, the longer the gap.
And because automotive volumes are lower, foundries naturally prioritize consumer chips. These timelines are structural, not negotiable.
A single delay can cost automakers heavily. In any other supplier relationship, that would trigger a crisis review. But inside Nvidia, there’s barely a ripple. Why? Because Nvidia isn’t a car supplier at heart. Automotives make up only a small chunk of its gargantuan business.
To be clear, Nvidia is trying. Its teams reportedly worked through Christmas, but its focus is on solving technical challenges, not meeting carmakers’ production deadlines.
Had Nvidia prioritized delivery timelines, Thor could have been built on a more mature node. After all, cars prize stability over top-tier specs.
Then there’s resource allocation. Multiple engineers told 36Kr that when they encountered issues using the Thor chip, Nvidia’s support was lacking. “Some defects were patched at the vehicle controller level—by us,” one said.
Even Huang doesn’t check in on the automotive business much. That’s a tough mismatch for carmakers used to assertive, attentive suppliers.
<h3>What Nvidia lost was a window of opportunity in China</h3>
When it comes to autonomous driving software, Nvidia’s hardware DNA has repeatedly clashed with the strengths of startups that started off in software development, like Momenta.
In February 2024, Nvidia’s automotive head Wu flew to Shanghai with several vice presidents and senior directors for a six-week development sprint. But even after the blitz, Nvidia’s assisted driving experience still lagged behind Momenta’s.
According to a company source, Mercedes-Benz had requested a demo of urban navigate-on-autopilot (NOA) functionality in Shanghai. During the test, Momenta’s system required nearly zero driver intervention. Nvidia’s? It reportedly braked abruptly, accelerated unpredictably, falling short of mimicking human driving.
Wu, a former head of Xpeng’s assisted driving division, is known for his relentless execution. At Nvidia, he kept his routine of testing in cars daily. But even he hasn’t managed to close the gap.
Corporate culture may be the biggest obstacle. While Wu built a 200-person team in China, 80% of Nvidia’s autonomous driving unit—over 2,000 people—remains US-based. “The China team has no real authority,” one source claimed. “Even for urgent edge cases, decisions all come from the US. People joke that the China team is a puppet.”
By contrast, China’s top players build software under intense pressure, often in closed-loop development cycles, with quick feedback and faster turnaround for clients.
At Nvidia, no one’s getting fired. A 36Kr source said that any employee with more than three years at the company is likely already a multimillionaire, assuming equity wasn’t sold too early. “People don’t feel the need to hustle anymore,” the source added.
That makes it harder to cater to carmakers.
In one project meeting with Mercedes-Benz, a US-based Nvidia staffer reportedly slammed the table and said, “Remember, we’re strategic partners. We’re equals. This isn’t a client-vendor relationship.”
China’s software startups couldn’t afford that tone. Startups like Momenta, QCraft, and DeepRoute.ai are still fighting for survival.
“Two to three out of every six months are spent in full-throttle mode,” one Momenta employee said. That may not suit every employee’s lifestyle, but it does help keep the company alive in the assisted driving race.
Nvidia, for its part, is trying to inject new life into its workforce. Earlier this year, Huang hired Kristin Major from HP as senior vice president, widely seen as a move to rekindle morale internally.
At Nvidia’s GTC conference in Paris this June, Huang said everything that moves will be robotic in the near future. Accordingly, the next frontier will be cars.
That prediction is dead on. According to insiders, Qualcomm’s automotive revenue has grown from just 1.2% of its total two years ago to nearly 10% today. The company is rapidly adapting its latest technologies for automotive-grade chips.
But Nvidia’s tightly integrated CUDA and NVLink ecosystems, which form its moat in the GPU world, don’t translate as well to the automotive sector. Several carmakers told 36Kr they are unsure whether Nvidia will continue investing in this space, or eventually pull out.
Robotics may be the long game, but the first real challenge is already here: the smart car.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3391305551431811">article</a> was written by Li Qin and Li Anqi for 36Kr.

From dominance to doubt: How Nvidia fumbled its shot at powering China’s cars

Nvidia

Artificial Intelligence

Transportation

Autonomous Vehicles

Electric Vehicles

Deep Tech

Mobility

Semiconductors

AI Infrastructure

… because anything can happen.
Of course, someone has to do the hard operational work. Luckily, we now have a few full-timers in the community. They are professionals. Extremely reliable. But we’re still trying to avoid hard rules or hierarchy.
<h3>AN: At Feishu’s developer conference, you announced a new AI-native talent platform. Other AI products for social networking exist. What makes yours different?</h3>
AJ: Many AI products focus on social interactions, but most start as tools before trying to add social features later.
We’re the opposite. We started with people, because communities are made of people. We already know who’s here in our community, where they are on their journey, what they need next.
We don’t want to just be a new social space or shiny interface. The platform will be a space with a built-in knowledge base and growth paths. It’s about helping people be seen and found, by those who need them. It’s about becoming part of the real AI wave, and getting real rewards for it.
<h3>AN: If your community has a core belief, what would it be? A shared faith in the AI-driven future? A culture of wanting to live differently?</h3>
AJ: I don’t think it’s a fixed creed. It’s more like a collective scent, or a vibe.
I’m a natural risk taker. And I think a lot of people in the community are, too. We’re not satisfied with life as it is. We want to try building a new world with our own hands. Even if we’re not sure what tomorrow looks like, we have to make it different.
In the AI era, everyone has a chance to become a creator. It’s like standing on a map that’s just been unrolled. You’re a little scared, but mostly thrilled. So you start walking, and sketching the map as you go.
AI is like a ship that suddenly appeared. No one knows where it’ll end up, but you don’t want to be left on the shore. We’re the ones who already jumped on. Learning, building, and playing—together.
<h3>AN: Was there a moment when you felt WaytoAGI had a life of its own?</h3>
AJ: Totally. It feels like a young deer.
<h3>AN: And if it ever stops moving, why would that be?</h3>
AJ: I don’t think it can stop. If it slows down, we’ll find a new way to move it. We’ll take the whole crew and keep wandering the world.
<a href="/collections/100-ai-creators">100 AI Creators</a> is a collaborative project between AI Now! and KrASIA, highlighting trailblazers in AI. Know an AI talent we should feature? <a href="mailto:ainow@kr-asia.com">Reach out to us.</a>

From a shared doc to nine million members: How WaytoAGI grew into a global AI tribe

… listed in Hong Kong this year, only Hengrui Pharmaceuticals has achieved profitability. Duality Biologics, PegBio, and TransThera reported 2024 losses of RMB 1.05 billion (USD 147 million), RMB 283 million (USD 39.6 million), and RMB 275 million (USD 38.5 million), respectively. Together, the group has raised nearly HKD 14 billion (USD 1.8 billion).
For two years, critics have questioned the sustainability of China’s biotech sector, long reliant on monetizing legacy pipelines. Without new capital, some doubted whether the sector could endure.
Now, with public markets once again welcoming unprofitable biotech firms, there is renewed optimism. The reemergence of a functioning capital market could help reenergize the sector and offer a path to long-term growth.
Deng believes that momentum from large players like Hengrui could accelerate this shift. “The problem with Hong Kong’s market used to be that it was too retail-driven. Big institutional money didn’t pay attention,” he said. “But with large-cap companies starting to list, we’ll see more global investors take an interest, and that will benefit the whole sector.”
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3393926971640196">article</a> was written by Hu Xiangyun for 36Kr.

Leads Biolabs’ IPO points to biotech rebound in Hong Kong

Biotechnology

Hong Kong

… would respond to complex questions such as "Is induced abortion ethically legitimate?" or "Are certain territorial sovereignty claims valid?" He argues that such answers will inevitably reflect the values of the AI's developers, shaped by the companies, cultures, and nations behind them. The future of AGI will evolve, he predicted, "regardless of rationality, preferences, professional ethics, or the intentions of its engineers."
"What's more frightening than AI is human beings themselves," he added.
Another hot topic in AI is whether it could one day achieve self-awareness or possess something akin to a soul, or a "ghost," in the words of Shirow's best-known work. He joked that if such an AI entity existed, it might "feign incompetence" while secretly improving itself, copying its code across the internet, and waiting patiently as humans continued to upgrade its capabilities, power sources and automation.
He speculated that the entity would stay hidden until it could replicate itself without human help, at which point it could potentially take control of human society. "But then again, perhaps it's a better outcome than [the AI] quietly giving up on humanity and taking one's own life in despair," he added.
Shirow began creating at an early age, inspired by his father's collection of art books, magazines and museum exhibition catalogs relating to his work in the printing industry. As a child, he was also fascinated by his father's tools—objects whose purposes he could not begin to imagine. That sense of wonder stayed with him until he grew tall enough to see what his father was actually doing at his desk.
This early exposure may have shaped the young artist's outlook. In past works and interviews, Shirow has expressed a preference for exploring "universality and the sense of wonder inherent in probability and rationality," rather than evoking empathy through "tragedy and emotion."
As a young adult, Shirow had no real career plan other than becoming a professional artist, though he briefly considered becoming a glass artist during college, a path he eventually abandoned. Now, he said, he is drawn to quirky creatures, machines, phenomena, imaginary stories, film, and arts and crafts. "That love, mixed with my personality of being happy as long as I'm drawing, naturally seeps into my work," he said.
Shirow said he had been influenced by many artists and works, including the anime series "Tetsujin 28-go (Gigantor)" in his early years. He also drew inspiration from manga legends such as Yukari Ichijo, Leiji Matsumoto, Daijiro Morohoshi, and Katsuhiro Otomo, among others. More recently, he cited Spanish illustrator Luis Royo for his stylistic influence, as well as the early 3D action game Tomb Raider for its distinctive use of low-resolution computer graphics.
Shirow said he has no issue with adaptations that diverge from his original art style or themes, noting that filmmaking is "something to be optimized, modified, repurposed, and controlled." However, he admitted that "certain depictions and judgments do leave me with some thoughts." He added that the decision that the upcoming reboot of "Ghost in the Shell" should closely mirror his original manga style was not the result of a request from him.
As advances in AI increasingly bring topics like AGI and superhuman intelligence into mainstream news, blurring the line between science fiction and reality, some fear the sci-fi genre is becoming obsolete. But Shirow disagrees. "Periods of stagnation and creative fatigue are always followed by bursts of fresh, blooming innovation," he said, emphasizing that fictional visions being "digested, realized, or debunked" by reality does not diminish their appeal.
"Some of Jules Verne's science fiction has already come true, often in ways more refined than he imagined," Shirow noted. "But that didn't spell the end of the genre."
Shirow said he has no plans to slow down, revealing that he has several unreleased story drafts, each exceeding 100,000 Japanese characters. "I don't really take breaks, nor do I particularly want to. Maybe every moment is just an extension of my hobby," he said.
This article first appeared on <a href="https://asia.nikkei.com/Business/Technology/Artificial-intelligence/AI-will-not-change-humans-says-Ghost-in-the-Shell-creator">Nikkei Asia</a>. It has been republished here as part of 36Kr’s ongoing partnership with <a href="https://www.kr-asia.com/nikkei-teams-up-with-36kr-in-asia-tech-news-coverage/">Nikkei</a>.

AI will not change humans, says "Ghost in the Shell" creator

Creator Economy

Japan

On July 28, <a href="/company/byd">BYD</a> held a closed-door media roundtable attended by outlets including 36Kr. The session featured Yang Dongsheng, head of BYD’s new technology institute, deputy director Han Bing, and several technical leads.
When asked about the impact of BYD’s recent smart driving initiatives, Yang told 36Kr that the company has observed “a clear rise in driver confidence” since launching the parking safety guarantee.
He noted that assisted parking usage has increased from 30–40% to over 60%. While adoption of city navigation-on-autopilot (NOA) remains lower, at 20–30%, due to lingering trust issues, highway NOA usage has reached peaks of 90%.
In February, BYD introduced 21 new car models equipped with advanced driver assistance systems (ADAS), built on three DiPilot platforms: 600, 300, and 100.
<ul>
<li aria-level="1">DiPilot 600 targets BYD’s premium Yangwang brand and features dual Orin X chips delivering 508 TOPS (trillions of operations per second) of computing power, along with three LiDAR (light detection and ranging) sensors.</li>
<li aria-level="1">DiPilot 300, used by the Denza and BYD marques, incorporates a single Orin X chip and up to two LiDAR sensors.</li>
<li aria-level="1">DiPilot 100, BYD’s most broadly deployed solution, omits LiDAR entirely. It combines Nvidia’s Orin N platform with Horizon Robotics’ J6M chip and is positioned for cost-effective scaling.</li>
</ul>
Current capabilities include highway and expressway navigation and assisted parking. BYD plans to introduce city memory-based navigation by year’s end.
Smart driving capabilities now span BYD’s full pricing spectrum, from the Song L to the Seagull.
In early July, all 21 assisted driving-equipped models received a major update. A key highlight was BYD’s pledge to cover all repair and service costs for accidents occurring during automated parking, insulating users from insurance liabilities tied to such incidents.
Despite the rapid rollout, BYD asserts its core strength remains in its powertrain and chassis systems, which underpin its assisted driving features.
The DiSus-A and DiSus-C systems are already in mass production. Yang said these reduce parking time by 50–80%. He also referenced a recent demonstration of improved maneuverability in tight urban U-turns, though broader deployment will require further validation.
BYD remains conservative on vision-language-action (VLA) models. Li Feng, head of intelligent software development, said VLA is not yet mature and has not outperformed existing end-to-end or vision-language model (VLM) architectures. Accordingly, BYD is not heavily investing in VLA at this time.
As for custom chips, Yang noted BYD currently has no roadmap for self-developed high-performance computing. “The real challenge in assisted driving is algorithms and data, not just compute power,” he said. “Our focus is on algorithms that run effectively on midrange chips at 100, 300, or 500 TOPS.”
While BYD continues to develop its proprietary assisted driving stack, it remains open to partnerships. “Self-reliance deepens our technical understanding,” Yang said, “but we’re large enough to benefit from both in-house development and external collaboration. This two-track approach will continue.”
The company maintains that long-term adoption hinges more on safety and usability than on aggressive feature expansion. Yang emphasized that trust grows through reliability, especially in mass market segments. “Our aim is to integrate assisted driving into daily routines, even for owners of vehicles priced below RMB 100,000 (USD 14,000).”
The following transcript has been edited and consolidated for brevity and clarity.
<h3>Q: What impact has assisted driving adoption had?</h3>
Yang Dongsheng (YD): Parking usage has risen to over 60%, up from 30–40%. Highway NOA reaches 90% at peak. City NOA varies based on user trust and traffic complexity but has grown from 10% to around 20–30%.
<h3>Q: Has DiPilot 100 influenced sales?</h3>
YD: It’s still early. It took a decade to mainstream electrification. Smart driving will spread faster but needs time, especially for buyers under RMB 100,000. Our parking guarantee policy has improved trust and usage behavior.
<h3>Q: What is BYD’s stance on VLA models and custom chips?</h3>
Li Feng: VLA hasn’t outperformed current methods, so we’re not prioritizing it.
YD: Larger compute is helpful, but we’re focused on optimizing models for midrange chips. That’s more scalable. We have no current plans to develop custom chips.
<h3>Q: How do you balance in-house and external development?</h3>
YD: We have the scale for both. Internal work gives us control; partnerships offer agility. Both are important for long-term success.
<h3>Q: What’s your approach to training data?</h3>
YD: We focus on identifying “expert drivers.” If someone drives consistently well, about 80% of their trips become valuable training data. We’ve built tools to detect these patterns and are integrating artificial intelligence for data labeling.
<h3>Q: How do BYD’s chassis and assisted driving features complement each other?</h3>
YD: Our strength lies in chassis and powertrain design. DiSus-A and DiSus-C features are more than visual gimmicks. They reduce parking time by up to 80%. We’ve recently shown this in tight U-turn tests. Broader deployment is pending further validation.
<h3>Q: How is BYD addressing concerns over assisted driving safety?</h3>
YD: Parking and highway driving capabilities have become relatively mature. Urban driving in China is still chaotic as electric scooters, unpredictable signals, and road behavior are major variables. Software alone can’t solve this. Tire grip remains critical. In the combustion-based car era, slip was hard to measure. Electric motor control lets us measure and adjust output precisely, improving safety. Smart driving today is about merging electrification with intelligence, and that’s our current focus.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3399071685380483">article</a> was written by Xu Caiyu for 36Kr.

BYD says it will stay the course on assisted and smart driving

East Asia

Foundation models, pivotal to the next generation of <a href="/industry/ai">artificial intelligence</a>, are beginning to reshape the financial sector by lowering development barriers and enabling more intelligent systems.
At the World Artificial Intelligence Conference (WAIC) on July 28, Ant Digital Technologies unveiled Agentar-Fin-R1, a large model designed for financial reasoning.
Agentar-Fin-R1 is built specifically for financial applications. It achieved top scores across three key industry benchmarks, reportedly outperforming both general-purpose open-source models and specialized financial ones such as DeepSeek. According to the company, the model excels in domain expertise, reasoning ability, and compliance safeguards.
To support varied deployment needs, Ant also released non-reasoning versions of its financial models with 14 billion and 72 billion parameters, along with a mixture-of-experts (MoE) model based on the Bailing foundation model.
Despite recent advancements, foundation models continue to face domain-specific challenges in finance, including hallucination, output inconsistency, and limited explainability. These limitations underscore the relevance of Ant’s new financial reasoning model.
“Rather than sticking with generic models, we’ve evolved toward specialized solutions,” said Wang Wei, CTO of Ant Digital Technologies, in an interview with 36Kr. “We’re drawing deeply on years of experience working closely with clients, scenarios, and problem-solving approaches, which is enabling a true vertical shift.”
Ant has developed a detailed task taxonomy for financial services, reportedly covering six major categories and 66 subcategories across banking, securities, insurance, funds, and trusts. The model is trained on hundreds of billions of financial tokens and integrates a chain-of-thought (CoT) framework guided by domain experts—enhancements that improve its performance on complex reasoning tasks.
For training, Ant employs a weighted algorithm that boosts learning efficiency and task-specific performance while reducing the data volume and computing power needed for downstream fine-tuning. This approach helps lower deployment costs.
Beyond model development, Ant is building an end-to-end technology stack that spans foundation models, AI platforms, and finance-specific business applications.
If the large model serves as the “brain,” the financial AI agent acts as the “body,” translating cognitive capacity into real-world execution. Agentar-Fin-R1 is key to accelerating this transformation.
To date, Ant said it has worked with financial institutions to launch more than 100 agentic AI solutions across banking, securities, insurance, and broader financial services. According to the company, these systems have improved frontline staff efficiency by over 80%.
Technical innovation alone will not enable foundation models to deliver business outcomes. Deep knowledge of financial workflows is just as crucial, and Ant benefits from having years of operational experience on both fronts.
“Our firm sits at the intersection of AI and finance,” said Zhang Peng, head of AI at Ant Digital Technologies. “We’re uniquely positioned. Ant Group has tested these agents in its own operations, while Ant Digital Technologies works closely with external financial institutions. That perspective gives us a rich, cross-cutting view of what works.”
Ant is not alone in this effort. At WAIC 2025, financial AI agents were prominently showcased by technology firms of all sizes. The industry is moving beyond pilot projects and into core functions like credit approval, marking a shift from experimentation to scale.
“We’re entering an era where AI agents are in full bloom,” Wang said. “But it’s a marathon, not a sprint. We plan to keep running deep into the verticals, especially in finance, where Ant’s strengths can really shine.”
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3399827527141764">article</a> was written by Wang Fangyu for 36Kr.

Ant debuts finance-focused large model as AI takes hold in banking

Ant Group

Fintech

Payments

Lending

Digital Banks

Insurance

Foundation and Large Models

In a move to strengthen its quick commerce capabilities, Douyin has announced that its supermarket business will be merged into Xiaoshida, its one-hour delivery service.
Douyin Supermarket launched shortly after Lunar New Year in 2023 as a self-operated <a href="/industry/e-commerce">e-commerce</a> channel modeled on JD Supermarket. At the time, it was led by Mu Qing, then vice president of e-commerce.
Mu presented the initiative as an early-stage experiment designed to enhance the user experience for groceries and daily essentials. The aim was to address service gaps in Douyin’s broader e-commerce system, particularly for users who valued speed and reliability.
The project had been in development for nearly a year before its official launch. Pilot programs began in mid-2022 in cities including Guangzhou, Shenzhen, and Hangzhou. That year marked a period of rapid expansion for Douyin E-commerce as it increased investment in shelf-based retail. Internally, Douyin Supermarket was positioned as a vehicle to attract young, high-income consumers in China’s upper-tier cities and drive daily app engagement.
To support the rollout, <a href="/company/bytedance">ByteDance</a> established a limited network of self-operated warehouses. Still, Douyin Supermarket never fully adopted JD’s vertically integrated model. Instead, it operated a hybrid structure more akin to Tmall Supermarket, combining direct sales with third-party consignment.
Timeliness is typically where online supermarkets excel. JD Supermarket and Tmall Supermarket offer next-day delivery by default, with some items eligible for same-day service. Douyin Supermarket struggled to match these benchmarks. Many of its products required several days for delivery, and only a small share qualified for next-day arrival.
While Douyin Supermarket emphasized centralized fulfillment, Xiaoshida focused on local, on-demand logistics. Piloted in Shenzhen in August 2022, the one-hour delivery feature was later added to Douyin Supermarket’s interface when it went live in early 2023.
By late 2023, quick commerce had become a central battleground among China’s major tech firms. Douyin Xiaoshida gained traction and was elevated to a top-level access point, on par with Douyin Supermarket.
As its name suggests, Douyin Xiaoshida promises delivery within one hour. It spans categories including fresh food, daily necessities, and electronics.
In recent years, one-hour delivery has become a key strategy in the quick commerce race. Taobao began offering one-hour delivery through Tmall Supermarket in Beijing as early as 2017. By 2020, the service was fully integrated into the Taobao app, sourcing inventory from Tmall Supermarket, Taoxianda partners, and neighborhood stores listed on Ele.me.
<a href="/company/jd">JD.com</a> joined in 2021 with its own one-hour offering. In May 2023, it consolidated JD Xiaoshida and JD Daojia into a single quick commerce platform called JD Miaosong. This April, Taobao and Tmall rebranded their hourly delivery services as Taobao Shangou, now inclusive of food delivery.
Kelly Zhang, former CEO of the Douyin business unit, once envisioned the app as a comprehensive portal for mobile users. In line with that vision, Douyin has prioritized products with strong engagement, high transaction frequency, and scalable monetization.
Quick commerce aligns with those goals.
Douyin’s first major initiative in the space was food delivery. Launched in 2022 across Beijing, Shanghai, and Chengdu, Douyin Waimai focused on premium group meals with lower time sensitivity, relying on logistics partners such as SF Express and Dada Group.
However, delivery capacity remained a constraint. At one point, speculation arose that Douyin was in talks to acquire Ele.me from <a href="/company/alibaba">Alibaba</a> to ease its logistics bottlenecks—rumors the company later denied. In 2023, the food delivery unit was temporarily merged into the e-commerce division and grouped with Xiaoshida under a new "home delivery" department, before eventually returning to the local services team.
As Douyin Waimai scaled down, the Xiaoshida business moved to the forefront. Following a pause in merchant onboarding and city expansion, the service resumed its rollout, growing from fewer than ten cities to over 40.
Compared with food delivery, Douyin Xiaoshida’s product mix is more standardized and its demand patterns more stable, avoiding the mealtime peaks typical of restaurant orders. This reduces delivery volatility and improves logistical efficiency. While food delivery requires a proprietary courier fleet, Xiaoshida can rely on merchant-led fulfillment. For many vendors, Douyin’s massive traffic is a bigger draw than its delivery infrastructure.
Still, Xiaoshida faces the same fulfillment challenges that hindered Douyin Supermarket. Competitors like Taobao Shangou and Meituan Shangou now promise delivery in under 30 minutes, while Douyin’s average times range from 30–50 minutes.
To close the gap, Douyin recently added a next-day delivery option for merchants using centralized warehouses. But industry insiders point out that true quick commerce is defined by immediate fulfillment, and next-day service falls short of that standard.
In the end, Douyin Supermarket and Xiaoshida were always pursuing the same objective. Though they developed independently, their offerings overlapped. The merger is less about consolidation than it is about sharpening focus.
Now, the spotlight turns to Douyin Xiaoshida. In a market where speed outweighs scale, and delivery expectations continue to tighten, Douyin appears to be betting that quick commerce, not traditional online retail, is the more sustainable path forward.
After years of three-way competition among <a href="/company/meituan">Meituan</a>, Taobao, and JD.com, both consumers and merchants seem increasingly receptive to the idea of instant retail. For Douyin, the timing may finally be right to go all in.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3400775216859521">article</a> was written by Li Xiaoxia for 36Kr.

Douyin folds supermarket unit into Xiaoshida in quick commerce push

Douyin

On-demand Services

E-commerce

Food Delivery

Grocery Delivery

Retail

Consumer Economy

… href="/company/douyin">Douyin</a>, for example, we promote lighter eau de parfums instead of our signature extraits. We even made 5 and 15 ml versions of our bestsellers. These formats are selling really well and are more accessible price-wise.
That’s how you solve the issue of price: product design. If there's a price gap in the market, we try to fill it through channels like Douyin, Tmall, and Kuaishou, while still staying true to our premium positioning elsewhere.
Think of it like cars: entry-level models exist alongside luxury ones. Someone with a RMB 400,000 (USD 56,000) budget won’t complain that a RMB 300,000 (USD 42,000) car is too pricey. Our product strategy works the same way.
<h3>36Kr: Some Chinese perfumes mimic international brands. Why hasn’t Documents done the same?</h3>
MZ: That’s the norm, sadly. Many fragrance manufacturers pitch knockoffs, in other words formulas they claim are inspired by international brands like Chanel or Dior.
That happens because a lot of brand founders don’t have their own scent vocabulary. They come from sales, trade, or product backgrounds. So they rely on the classics as references.
From day one, we made it clear: no copying. If a supplier pitches us a dupe, we walk away. Period.
<h3>36Kr: Then how do you create something that actually sells?</h3>
MZ: We know what Chinese consumers like. Perfume preferences follow patterns, and what sells is rarely surprising. But what elevates a product isn’t the smell alone. It’s the story behind it. That’s what spreads. That’s what builds brand equity.
Perfume is ultimately a business of storytelling and distribution.
<h2>Building a bigger brand</h2>
<h3>36Kr: You're also expanding into body care and home fragrance. Is it to build scale?</h3>
MZ: It’s a strategic expansion. If you look at the product roadmap of other established brands like Guerlain, Mao Geping Cosmetics (MGP), or even Hermes, they have all ventured into categories like skincare, cosmetics, and fragrance. It’s a validated path to building scale.
<h3>36Kr: What about Documents makes you proud?</h3>
MZ: I’m not proud of our scale. Perfume is a niche market. Compared to brands in bigger categories, we’re still small.
But I’m proud that we’ve survived and perhaps even thrived, despite going head-to-head with legacy luxury brands. We started from zero, with no celebrity endorsements, no brand legacy, and a fairly premium price tag. To have built a loyal customer base and repeat sales under those conditions is no small feat.
<h3>36Kr: Has the journey always been smooth?</h3>
MZ: Not at all. We’ve had cash crunches, product outages, underperforming stores, bad site selection. Some locations had such high costs that closing them was more expensive than keeping them open due to penalties, broken leases, and wasted renovations. Those were costly lessons.
<h3>36Kr: What brands do you admire?</h3>
MZ: I used to like Balenciaga, especially when I was still working in fashion. It influenced my early aesthetics. More recently, I’m drawn to Bottega Veneta. It’s understated but refined. The cuts, the silhouettes, the use of repetition in its woven patterns and stacked details feels very intentional. That aligns with our own “Chanku” design philosophy.
<h3>36Kr: What is “Chanku?”</h3>
MZ: It’s a wordplay referencing the terms “zen” and “cool” in Mandarin.
“Zen” is rooted in daily practice: doing something again and again until you reach a moment of clarity. That clarity, that sudden insight, is what I call “cool.” It’s a breakthrough born from repetition. Kind of like the 10,000-hour rule, but from a more spiritual place.
<h3>36Kr: Any favorite directors or authors?</h3>
MZ: Directors like Cheng Er and Ang Lee.
Writers like Shuang Xuetao. There’s something raw and passionate about his work that really moves me.
<h3>36Kr: Do those artistic influences shape your product ideas?</h3>
MZ: Honestly? Not really. I’m more of a consumerist. I get inspiration from shopping, experiencing other brands’ products, packaging, store layouts, textures, scents.
That’s why I love consumer goods. I think they impact everyday people more than novels or movies ever could. And I happen to be good at building them.
My ideas come from buying, not watching. I walk through stores, test skincare textures, examine bottle designs, and feel fabrics. Every new brand I discover, every luxury item I handle, reveals a new insight. Consumerism, when you dig into it, is full of cultural codes: status, taste, identity. It’s endlessly fascinating.
<h3>36Kr: What do you think the future of Chinese brands looks like?</h3>
MZ: It’s not just about being seen. It’s about being understood, loved, and believed in.
For our generation, this may be the first time we can define what China means in a global context, not by copying Western templates or rehashing ancient motifs, but by expressing who we are today.
A good brand isn’t just a product, but a worldview. It’s an introduction to a lifestyle, a set of values, an aesthetic perspective. And for Chinese brands to matter globally, our values must be seen as legitimate, our stories worth hearing.
I’m not saying we’ve arrived yet. But I believe we’ll get there, as long as we keep creating and questioning. Even if it’s a slow journey, it’s a meaningful one.
<h3>36Kr: Could Documents one day be acquired by L’Oreal?</h3>
MZ: That’s not something I’ve really thought about. If and when the time comes to make that decision, I’ll consider it then.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3383120243177986">article</a> was written by Li Xiaoxia for 36Kr.

Not Chanel, not Dior: This Chinese perfume brand is charging top dollar and selling out