Alibaba

Baidu

ByteDance

Meituan

Pinduoduo

Tencent

Most Followed

KrASIA Connection

China's Auto Sector

Earnings Reports

Inside China's Startups

Venture Roundup

Our Events

View All

Collections

News

Features

Insights

Ecosystem

DeepSeek

Kuaishou

Li Auto

Xiaohongshu

Xiaomi

Xpeng

On the Radar

Events

Latest

About

36Kr (EN)

Consumer

Robotics

Fintech

Tesla’s Optimus humanoid robot is catching attention once again with the introduction of its second generation at this year’s World Artificial Intelligence Conference (WAIC) in Shanghai.
Originally <a href="https://asia.nikkei.com/Business/Technology/Tesla-reveals-humanoid-robot-that-will-cost-less-than-20-000">debuting</a> in September 2022 under the codename Bumblebee, Tesla is now aiming to begin selling Optimus robots by the end of 2025, as stated by Elon Musk during an investor conference call in April, according to <a href="https://www.reuters.com/technology/tesla-could-start-selling-optimus-robots-by-end-next-year-musk-says-2024-04-24/">Reuters</a>.
This target aligns with guidelines from China’s Ministry of Industry and Information Technology (MIIT), which views humanoid robots as a “disruptive innovation” capable of spurring growth in China’s slowing economy, <a href="https://www.scmp.com/news/china/politics/article/3240259/china-says-humanoid-robots-are-new-engine-growth-pushes-mass-production-2025-and-world-leadership?module=inline&amp;pgtype=article">SCMP</a> reported. The WAIC display suggests that Tesla’s plans are on track, although these robots are not expected to reach retail customers soon.
Elon Musk has previously suggested that each Optimus robot could eventually retail for around USD 20,000. A February 2024 <a href="https://www.goldmansachs.com/intelligence/pages/the-global-market-for-robots-could-reach-38-billion-by-2035.html">report</a> by Goldman Sachs supports this projection, noting that material and manufacturing costs have been lower than initially expected. While global analysts had anticipated a cost decline of 15–20% annually, the actual figure has been around 40%, resulting in a current price range of USD 30,000–150,000, closer to Musk’s forecast.
Despite the optimistic projections, current prices place humanoid robots in the same category as high-end retail machinery like automobiles, making them a considerable investment for most consumers unless compelling use cases are identified. For now, the most immediate applications for humanoid robots seem to lie in logistics, warehousing, retail, and manufacturing, where they could help address labor shortages and improve efficiency. Tesla plans to deploy Optimus robots in its manufacturing processes post-rollout.
Tesla won’t be short of competition in this burgeoning segment. Over a dozen other humanoid robots were on display at the 2024 WAIC, including <a href="https://interestingengineering.com/innovation/china-humanoid-robot-qinglong">Qinglong</a>, a full-sized robot with a bionic torso and anthropomorphic motion control. Unveiled by the National and Local Co-Built Humanoid Robotics Innovation Center, Qinglong stands approximately 1.8 meters tall, weighs around 80 kilograms, and features 43 degrees of freedom (DoF) with a joint peak torque of 400 newton-meters (Nm) and computing power equivalent to 400 tera operations per second (TOPs). During the conference, the robot demonstrated its capabilities by performing various desk cleaning tasks following voice instructions.
Other players in the field include Xpeng Motors, which showcased its <a href="https://kr-asia.com/xpeng-motors-unveils-px5-humanoid-robot-underlining-its-vision-for-the-future">PX5 humanoid robot</a> with robust maneuvering capabilities in October last year, and Agibot, a company founded by a former Huawei executive, which is developing the <a href="https://kr-asia.com/expedition-a1-is-a-robot-that-can-cook-do-laundry-and-take-care-of-the-elderly">Expedition A1</a> robotic household assistant with backing from investors like BYD and Warburg Pincus.
Tesla’s development of humanoid robots is being closely watched by the market. At the recent shareholders meeting on June 13, Musk <a href="https://www.teslarati.com/tesla-2024-annual-shareholder-meeting-transcript/">referenced</a> the droids from Star Wars, suggesting that the idea of each person owning a C-3PO and R2-D2 is appealing. Extrapolating from this concept, he envisioned a future where humans own two robots each, estimating that Optimus could become a USD 20 trillion business, assuming global demand reaches a billion units and Tesla captures around 10% market share.
<div class="embed-container"><iframe src="https://www.youtube.com/embed/cpraXaw7dyc?si=QjVs8_YA7eTyDWsK?autoplay=0&amp;mute=1&amp;origin=https://OurWebsiteDomain" frameborder="0" allowfullscreen> height="278" frameborder="0" allowfullscreen="allowfullscreen" data-ytbridge="vidSurrogate2"></iframe></div>
While Musk's vision for Optimus is ambitious, it has helped Tesla garner significant market interest. Factoring in the company’s humanoid robot developments, Ark Invest, an exchange-traded fund (ETF) manager run by prominent investor Cathie Wood, <a href="https://www.ark-invest.com/articles/valuation-models/arks-tesla-model">expects</a> Tesla’s stock price to reach USD 4,600 by 2026, over 18 times the current price as of market close on July 5.
However, the long-term viability of Optimus will ultimately hinge on reducing manufacturing costs to as low as USD 10,000 per unit, half the target retail price.

Tesla showcases second-gen Optimus humanoid robot in Shanghai, eyes 2025 launch

Tesla

Deep Tech

China

… could have downsides for the startups, such as being perceived as an entities associated with the Hong Kong or Chinese governments at times of even greater geopolitical tension.
"Now it's probably still fine, but eventually, if we start to see geopolitical tensions rising further … I think in some extreme cases, that can be a problem," he said.
SmartMore is also backed by IDG Capital and HongShan, the previous China investment arm of Sequoia Capital.
When asked by Nikkei about the possible impact of geopolitics on SmartMore, founder Jia Jiaya said that the company's AI focuses on manufacturing and that it exercises "great restraint" in how it is used. The company uses footage of production lines to train AI to identify quality issues—a limited scope that Jia suggested reduces privacy concerns.
Jia added that US restrictions on supplying chips to China have a "relatively small" impact on the company, as the chips it uses are less advanced than those found in consumer electronics like smartphones.
This article first appeared on <a tabindex="0" href="https://asia.nikkei.com/Business/Technology/Hong-Kong-s-AI-bet-through-Temasek-style-fund-signals-strategic-shift" rel="noopener noreferrer" data-token-index="1">Nikkei Asia</a>. It has been republished here as part of 36Kr’s ongoing <a tabindex="0" href="https://www.kr-asia.com/nikkei-teams-up-with-36kr-in-asia-tech-news-coverage/" rel="noopener noreferrer" data-token-index="3">partnership with Nikkei</a>.

Hong Kong's AI bet through Temasek-style fund signals strategic shift

JD.com is doubling down on its <a href="https://kr-asia.com/jd-coms-commitment-to-low-price-strategy-pays-off-with-strong-financial-performance-in-latest-quarter">low-price strategy</a>, launching new initiatives aimed at capturing lower-tier markets. According to sources close to the company, Jingxi, JD.com’s budget shopping platform, will now be known as Jingxi Ziying to reflect the significant shift in approach.
Internally, Jingxi Ziying is viewed as a cornerstone of JD.com’s low-price strategy. The platform has begun ramping up recruitment in lower-tier cities, offering a fully managed model to white-label merchants. Merchants focus on production while Jingxi handles operations, logistics, and after-sales services, earning revenue from the price difference of goods.
Sources also revealed that Qi Ting, a long-time JD.com employee with experience in managing omnichannel operations, leads this initiative and reports directly to group CEO Sandy Xu. The team has grown to over 200 members, mainly in procurement, marketing, and quality control, with more than two-thirds in procurement. Rapid expansion is underway, with the team expected to exceed 800 members by year-end.
Jingxi Ziying is highly valued by JD.com’s leadership, receiving frequent guidance from top executives. Currently, its primary goal is to attract and retain new users in lower-tier markets.
Multiple insiders confirmed that Jingxi Ziying is focusing on order volume rather than GMV maximization at this stage, which supports the idea that higher daily order volumes reflect more active users in lower-tier markets.
JD.com has faced numerous challenges but has remained committed to lower-tier markets. The name "Jingxi" signifies JD.com's persistent efforts in these areas.
In 2020, JD.com established the Jingxi Business Group specifically for lower-tier markets. This group includes the Jingxi Pinpin group buying service, Jingxitong convenience stores, and Jingxida courier services. "The previous development of Jingxi did not meet expectations," said a former JD.com employee. Last November, Qi Ting was appointed to lead the division and launched the Jingxi Ziying project.
Unlike JD Ziying, which resells goods from brand merchants, and Jing Zao, JD.com's private label, Jingxi Ziying targets source factories and white-label merchants. By directly purchasing and shipping from the source, costs are reduced, enabling lower prices. Jingxi Ziying currently collaborates with core industrial zones to supply products in categories like tissue paper, earphones, underwear, and fruits, with some achieving monthly sales of over one million units. The goal is to work with over 10,000 factories by year-end.
"To differentiate from competitors in low-price goods, we finally set the selection criteria for Jingxi Ziying as having a bottom line and being adequate," a source close to JD.com revealed.
This standard is crucial in an era where low prices dominate but product quality varies. "Jingxi Ziying conducted a survey, purchasing 1,000 items from various low-price platforms, finding that 80% were either mismatched or underweight, with some having serious quality and safety issues," the source disclosed.
JD.com aims to achieve consumption stratification within its platform. "Once Jingxi Ziying merchants scale up, it can stimulate the supply side, attracting more third-party ‘platform open plan’ (POP) merchants to JD.com's main site. This will enrich the supply of low-price goods, meeting the demand for high-quality, low-price products, and catering to different consumer groups," the source added.
A senior management meeting participant revealed that JD.com's current directive for Jingxi Ziying is "to move faster." During this year's 618 festival, Jingxi Ziying achieved over one million orders within the first four hours, a 220% increase from the previous day.
whereRelatedArticle
This initiative has already shown promise. A source close to JD.com noted that user stickiness in lower-tier markets for Jingxi Ziying is significantly higher than on the main site. In June, the monthly purchase frequency of users reached seven times.
This success is not easily won. JD.com has a larger user base in first-tier cities compared to lower-tier markets. With the rise of Pinduoduo, JD.com needed new strategies to attract users in lower-tier markets. Beyond Jingxi Ziying, JD.com has been fostering a low-price mindset among merchants and users through billion-yuan subsidies and direct sourcing since last year, aiming to gain more market share. However, whether it can stand out remains to be seen.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/2843483446299271">article</a> was written by Ke Xiaobin for 36Kr.

JD.com pushes deeper into lower-tier markets with new Jingxi Ziying model

JD.com

On-demand Services

E-commerce

Logistics

Consumer Economy

On June 29, Farasis Energy kicked off trial production at its new Guangzhou base, nestled in the Sino-Singapore Guangzhou Knowledge City. This facility, boasting an annual production capacity of 30 gigawatt-hours, wrapped up its main construction and began operations in just 10 months—from the first pile driven on September 1, 2023.
The first phase of the factory zeroes in on producing <a href="https://batteryuniversity.com/article/bu-301a-types-of-battery-cells">large pouch cells</a> using a lithium iron phosphate (LFP) system branded as the <a href="https://en.farasis.com/index/news-detail/infoid/288">“Super Pouch Solution” (SPS)</a>. These products primarily target the pure electric passenger car market, with added applications in hybrid vehicles.
Launched in 2022, SPS entered mass production in April this year, debuting in Geely-owned Radar’s first pure electric four-wheel-drive pickup, the Radar Horizon.
As Farasis Energy’s flagship product, SPS simplifies battery system design by horizontally arranging large pouch cells. This design cuts costs—reducing design expenses by 50% and material costs by 33%—and boosts volumetric efficiency by 75%, meeting vehicle range requirements of 500 to 1000 kilometers.
Farasis Energy touts SPS as a versatile solution, compatible with various material systems, including high-nickel ternary, low-nickel, lithium-rich manganese, lithium manganese iron phosphate, and sodium-ion, making it adaptable to different products and scenarios.
Responding to the market's demand for fast charging, SPS delivers 400 kilometers of range with just a 10-minute charge, using large pouch cells at varying charging rates.
At the launch event, Wang Yu, chairman of Farasis Energy, showcased fast-charging cells rated at 2.4C, 3C, and 4C. These cells offer 150, 100, and 75 kilowatt-hours of charge, and ranges of 1000, 800, and 600 kilometers, respectively. They can charge to 40%, 50%, and 70% in ten minutes, achieving ranges of 400, 400, and 420 kilometers.
Beyond its product advantages, the Guangzhou base is primed for sustainability. It slashes GWh investment by 67%, labor by 50%, and energy consumption by 50%.
Equipped with automated flexible production lines, digital twin technology, and smart quality inspection solutions, the factory reduces single-cell defect rates to PPB levels (one in a billion) and ensures full traceability of each cell. This innovation boosts production capacity per acre by nearly 50% compared to traditional processes.
Committed to zero-carbon operations, the base leverages renewable energy sources like photovoltaics, generating 90 million kWh annually, saving around 30,000 tons of standard coal, and cutting carbon emissions by over 20,000 tons each year.
The automotive industry is Guangzhou's largest, with a comprehensive industrial chain centered around 12 vehicle manufacturers, making it one of China’s biggest automotive clusters. The start of production at Farasis Energy’s Guangzhou factory fills the final gap in the city’s automotive battery segment.
Farasis Energy is aiming for the Guangzhou base to deliver products to city customers within four hours post-production, enabling same-day deliveries across South China.
At the ceremony, Farasis Energy signed cooperation agreements with GAC Aion, Trumpchi, Sunward, and Starair Aircraft.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/2842062453623688">article</a> was written by Han Yongchang for 36Kr.

Farasis Energy begins trial production at new Guangzhou facility

China Tech

Battery

Mobility

… companies tend to believe they can handle it themselves. Additionally, large companies prefer focusing on the Chinese market, meaning they are reluctant to take on the development risks for small indications like rare diseases.
"They tend to be conservative and concerned about the toxicity and side effects of entirely new delivery systems. Therefore, they prefer products with complete trials and immediate Center for Drug Evaluation (CDE) approval data. Collaborating with domestic pharmaceutical companies remains challenging."
<h3>Mapping the pharmaceutical industry in China:</h3>
<ul>
<li aria-level="1">Ribo Life Science: Founded in 2007, focusing on RNA interference (RNAi) drugs, Ribo is a major pioneer in China's pharmaceutical industry. The company has established an independent, fully integrated R&amp;D platform for small nucleic acid drugs, supporting the entire lifecycle from early research to commercialization. It currently has eight clinical products targeting cardiovascular and metabolic diseases, liver diseases, and ophthalmic diseases. In May 2024, Ribo’s siRNA drug targeting FXI (RBD4059), received European Medicines Agency (EMA) approval for Phase 2 clinical trials. In April, its siRNA drug for hepatitis D virus (RBD1016) received the same approval.</li>
<li aria-level="1">Visirna Therapeutics: Established in 2022, Visirna has formed a strategic partnership with Nasdaq-listed small nucleic acid drug company Arrowhead Pharmaceuticals. Its product pipeline includes three siRNA drugs in clinical development for cardiovascular and metabolic diseases. In April 2024, Visirna completed the first patient dosing in a Phase 2 clinical trial for VSA006, its siRNA drug for non-alcoholic steatohepatitis (NASH) in China.</li>
<li aria-level="1">Argo Biopharma: Founded in 2021, Argo focuses on developing next-generation siRNA drugs. Since its inception, it has raised over RMB 700 million (USD 96.2 million) in financing and established a pipeline of multiple siRNA candidates targeting cardiovascular diseases, rare diseases, hepatitis B, autoimmune, and neurological diseases. In January 2024, Argo Biopharma signed a USD 4.165 billion collaboration agreement with Novartis to develop multiple cardiovascular siRNA drugs. Argo Biopharma currently has five siRNA drugs in clinical trials, covering cardiovascular diseases, rare diseases, and viral infections.</li>
<li aria-level="1">Hengrui Pharmaceuticals: The company started Phase 1 clinical trials for its small nucleic acid drug HRS-5635 targeting chronic hepatitis B in 2023. In May 2024, Hengrui announced plans to initiate Phase 2 clinical trials to evaluate the efficacy and safety of HRS-5635 injection alone or in combination with other drugs for treating chronic hepatitis B.</li>
<li aria-level="1">Synerk: Founded in 2018, Synerk specializes in RNA-targeted therapies. Its founding team has extensive experience in successful siRNA drug development and has established proprietary hepatic and extrahepatic delivery platforms and chemical modification technology platforms. Synerk's most advanced pipeline has entered Phase 1 clinical trials, with dozens of participants enrolled and dosed. The company expects to have multiple pipelines, including CNS indications, in clinical trials by the end of 2025.</li>
<li aria-level="1">CSPC Pharmaceutical Group: The group's mRNA Covid-19 vaccine was the first independently developed and authorized for emergency use in China. In June 2024, CSPC’s mRNA-LNP-based CAR-T cell injection (SYS6020) was approved for clinical trials for multiple myeloma.</li>
<li aria-level="1">Chia Tai Tianqing Pharmaceutical Group: In April 2024, Chia Tai Tianqing's siRNA drug TQA3038 for chronic hepatitis B completed Phase 1 clinical trials.</li>
<li aria-level="1">Junshi Biosciences: In April 2023, Junshi’s siRNA drug ANGPTL3 for hyperlipidemia entered Phase 1 clinical trials. In April 2024, the company's self-developed PCSK9 monoclonal antibody was accepted by China's NMPA for treating primary hypercholesterolemia and mixed dyslipidemia.</li>
<li aria-level="1">Sano Pharmaceuticals: Founded in 2007 and focused on RNA therapy development, Sano’s pipeline targets tumors, fibrotic diseases, viral diseases, and cardiovascular metabolic diseases. The company has four drugs in clinical stages, with STP705 for squamous cell carcinoma and basal cell carcinoma set to enter Phase 3 clinical trials and STP707 for multiple solid tumors in Phase 1 clinical trials.</li>
<li aria-level="1">Sanegene Bio: Founded in 2021, Sanegene specializes in developing small nucleic acid drugs based on RNAi technology. It has established proprietary tech platforms for chemical modification and liver and extrahepatic delivery. Its pipeline covers cardiovascular and metabolic diseases, immune-mediated diseases, and neurological diseases. In December 2023, Sanegene announced it raised over USD 80 million in Series A+ funding. In June 2024, its RNAi drug SGB-9768, which targets the complement component 3 (C3) protein, received clinical trial approval. Another siRNA hypertension treatment drug, SGB-3908, developed in collaboration with Innovent Biologics, received clinical trial approval in May.</li>
<li aria-level="1">Rona Therapeutics: Established in 2021, Rona aims to develop innovative nucleic acid drugs with its proprietary “Razor” platform. Using its unique delivery platform, the company’s nucleic acid drugs are developed in a modular and programmable manner, primarily targeting metabolic, CNS, ophthalmic, and oncology diseases. In December 2023, Rona’s first self-developed PCSK9 siRNA drug RN0191 was approved for clinical trials to treat hypercholesterolemia, mixed hyperlipidemia, and atherosclerotic cardiovascular diseases.</li>
</ul>
Note: Xu Yao and Shi Ming are pseudonyms.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/2841638408932227">article</a> was written by Hu Xiangyun for 36Kr.

China’s pharmaceutical industry bets big on small nucleic acid drugs in 2024

Biotech

Leapmotor, a Hong Kong-listed electric vehicle maker backed by Stellantis, has unveiled its new six-seater SUV, the C16. The C16 is available in both battery electric vehicle (BEV) and extended-range electric vehicle (EREV) variants, offering three trims, making up a total of six models with a price range of RMB 155,800–185,800 (USD 21,430–25,555).
The C16 is positioned as Leapmotor's flagship mid- to large-sized SUV, measuring 4.915 meters long, 1.905 meters wide, and 1.77 meters tall, with a wheelbase of 2.825 meters. The EREV and BEV variants share the same dimensions, and the vehicle’s overall size is similar to the company’s other SUV, the five-seater C10.
According to Leapmotor, optimizing motor placement and vehicle structure has increased the legroom and foot space for the third row, making passenger room in the C16 comparable to a vehicle with a 2.95-meter wheelbase.
The EREV variant of the C16 is powered by a 1.5-liter engine as a range extender and a fuel tank with a capacity of 50 liters. It offers a pure electric range of 200 kilometers and a CLTC range of 1,095 kilometers when fully fueled and charged. The EREV variant also has a battery pack based on lithium iron phosphate (LFP) chemistry with a capacity of 28.4 kilowatt-hours, supporting fast charging from 30% to 80% in 30 minutes. It is powered by a rear electric motor with peak power of 170 kilowatts and peak torque of 320 newton-meters, allowing it to accelerate from 0–100 km/h in 8.46 seconds.
The BEV version features a 67.7 kWh battery pack, with a CLTC range of 520 kilometers. Despite its moderate range, Leapmotor says that the pure electric C16 uses a silicon carbide-based 800V fast-charging platform, with a maximum charging power of 180 kW, allowing the battery to charge from 30% to 80% in just 15 minutes. The C16 BEV is powered by a single rear-mounted motor with a peak power of 215 kW and peak torque of 360 Nm, enabling it to accelerate from 0–100 km/h in 6.37 seconds.
Inside, the C16 offers memory and welcome functions for the driver’s seat, independent heating and ventilation for the second-row seats, and a small table behind the co-pilot seat. The vehicle also comes standard with six airbags.
The smart cockpit of the C16 is equipped with Qualcomm’s Snapdragon SA8295P chip, delivering artificial intelligence computing power of up to 30 TOPS. Both the C16 EREV and C16 BEV use Nvidia’s Drive Orin chip in the highest-priced variant with 254 TOPS of computing power. Both variants come with a LiDAR sensor with 128 channels. This configuration enables both front and rear passengers to use the in-car voice assistant at any time. Passengers in all three rows can control the vehicle and seats via voice commands. Additionally, the vehicle features a 15.6-inch entertainment screen that folds down from the roof, viewable by second- and third-row passengers.
In the sub-RMB 200,000 (USD 27,510) price segment, companies like Nio and Xpeng Motors are gradually phasing out the use of LiDAR. However, the “intelligent driving” version of the C16 is equipped with Nvidia’s Orin X chip and LiDAR sensor solutions from Hesai Technology, with a total of 30 sensors throughout the vehicle. Such comprehensive configurations are rare in this price range.
Leapmotor claims that, with a total computing power of 284 TOPS, the C16 can run its “Navigation Assisted Cruise” (NAC) system in urban areas and “Navigation Assisted Pilot” (NAP) on highways and urban expressways. Currently, the latter feature is available in Shanghai, Jiaxing, and Hangzhou, with plans to expand to more cities.
Previously, Leapmotor’s strategy for intelligent driving was primarily to align with other market players. However, at a recent media communication meeting, Leapmotor founder and chairman Zhu Jiangming said that the internal strategy is evolving, with a significant increase in the importance of intelligent driving over the next 3–5 years. Leapmotor has built an in-house team of over 400 people to develop and implement technologies like BEV.
whereRelatedArticle
The launch of the C16 fills a gap in Leapmotor’s six-seater SUV product line. However, with three SUV models—C10, C11, and C16—priced between RMB 150,000–300,000 (USD 20,630–41,260), Leapmotor aims to thoroughly penetrate this price segment and achieve monthly sales of 30,000 units. The platform-based manufacturing approach allows for up to 88% component commonality, with cost savings passed on to customers.
Looking ahead, Leapmotor plans to establish four major model series: A, B, C, and D. In 2025, they will launch three B-series models targeting the RMB 100,000–150,000 (USD 13,755–20,630) market, with high-end models featuring LiDAR and high-end AI computing chips. The D-series models will debut in 2026, focusing on the RMB 200,000 market.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/2840161522617223">article</a> was written by Li Anqi for 36Kr.

Hong Kong-listed Leapmotor reveals the C16, a six-seater SUV

East Asia

Leapmotor

Transportation

Electric Vehicles

On June 26, DingTalk hosted the Make 2024 DingTalk Ecosystem Conference in Beijing, where it officially launched its AI-powered search feature and enhanced the AI assistant introduced in January. Significantly, DingTalk announced the full opening of its model layer, partnering with six large model companies to build a more open AI ecosystem.
AI search is becoming a hot topic. Perplexity AI, with its unique Q&amp;A engine, recently reached a valuation of over USD 1 billion in its Series B funding round. Google is integrating Gemini into its search engine, while Chinese players like TianGong AI, Meta Sota, and Genspark from MainFunc have entered the market. Unlike these products, DingTalk's AI search solution primarily targets internal enterprise scenarios.
DingTalk aims to resolve the fragmentation of information, a major pain point for users. In a company's internal communications, vast amounts of information are generated daily across various collaboration software, accumulating chat records, documents, meeting minutes, and business data. Traditionally, employees needed to sift through past chat logs to find necessary information, resulting in inefficiency. With the launch of its AI search feature, DingTalk can understand user queries through natural language processing (NLP), integrating fragmented information scattered across documents, chat logs, and meeting notes into a coherent knowledge network, such as mind maps and outlines, making information more accessible.
Another highlight is the feature’s capability in predicting and recommending information that users might need, based on their work scope and historical behavior. Additionally, the AI search feature can delve into internal data including personal and team schedules, to-do lists, and project details, enabling comprehensive information collection and organization. This means users can find necessary information and trace related personnel, organizations, and documents for a multidimensional overview.
Differing from user-personalized AI assistants that require manual data input, DingTalk's solution integrates platform data, promising a differentiated user experience through data accumulation. In a post-conference interview, DingTalk CEO Ye Jun said that, while current AI capabilities are used primarily for general search, deeper application at enterprise level will incorporate more data into the system, facilitating specialized use cases.
DingTalk COO Fu Xujun said that, unlike traditional and other public AI search engines, DingTalk's version focuses on the B-end, aggregating data from business, management, and personal usage scenarios to provide accurate and personalized search results.
Looking ahead, DingTalk hopes to combine AI technology with knowledge graph (KG) technology to form a networked structure of information, enhancing task execution. Discussing challenges in developing AI search, Fu acknowledged that presenting data accurately and objectively for internal enterprise use is a key challenge compared to consumer-facing use cases.
The business intelligence (BI) scenario exemplifies this. Fu explained that traditional BI scenarios are long and demanding, such as accurately fetching data and choosing appropriate visualizations like line or pie charts, analyzing trends, and addressing errors during the process. "Our goal is to eliminate hallucinations and leverage large models' strengths, like understanding, abstraction, summarization, and emergent capabilities, to improve,” said Fu.
In addition to AI search capabilities, DingTalk upgraded the AI assistant released in January. The update connects the AI assistant with the Tongyi Qianwen large model by default and enables users to switch between various large models developed by companies like MiniMax, Moonshot AI, Zhipu AI, OrionStar, 01.AI, and Baichuan AI, based on their needs. Moreover, the AI assistant now boasts enhanced memory and reasoning capabilities, remembering relevant information and breaking down complex tasks. Its perception system also adds time awareness, better coordinating team schedules to avoid conflicts.
DingTalk introduced multi-agent collaboration and anthropomorphic capabilities. Multi-agent collaboration allows multiple AI assistants to work together in a workflow, breaking down complex tasks sequentially. For example, in planning a marketing event, three AI assistants can collaboratively complete tasks like identifying hotspots, analyzing events, and brainstorming creative ideas, while automating the process. Anthropomorphic operations enable users to demonstrate complex operations to the AI assistant through dialogue, enhancing convenience and efficiency.
Ye said that DingTalk will share its AI capabilities with ecosystem partners and customers through two APIs, pushing AI into application across various scenarios. Beyond new AI products, the strategy of ecosystem openness continues. Last April, DingTalk integrated the Tongyi Qianwen large model, followed by offering its AI platform-as-a-service (PaaS) solution to ecosystem partners and customers in August, helping them redesign products with large models.
Currently, DingTalk's ecosystem includes over 5,600 partners, with more than 100 AI ecosystem partners, covering areas like AI solutions and plugins. DingTalk said its AI services are now called over 10 million times daily. Significantly, DingTalk, which is part of Alibaba Group, is now fully open to multiple large model ecosystem partners at the model layer, including product and scenario access and AI assistant development platforms.
As of now, six companies—MiniMax, Moonshot AI, Zhipu AI, OrionStar, 01.AI, and Baichuan AI—have integrated with DingTalk, jointly exploring various real-world scenarios. Developers can choose different large models from these companies for personalized development.
whereRelatedArticle
At the conference, DingTalk also launched a startup version, priced at RMB 980 (USD 135) per year, alongside its other commercial versions. This year, DingTalk has also reduced prices, offering customized workbenches, enterprise mailboxes, annual inspection certification, and enterprise cloud disks in the startup version. DingTalk announced free one-year usage support for 10,000 new registered enterprises.
Whether expanding coverage internally or collaborating with large model partners externally, the goal is ultimately to increase AI adoption. As DingTalk is showcasing, this consensus has become a common goal among AI players this year.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/2839679142071175">article</a> was written by Wang Yixin for 36Kr.

Alibaba’s DingTalk integrates with six AI partners, unveils new AI search feature

Enterprise Services

Artificial intelligence

… aria-level="1">Superbank, an Indonesian digital bank, raised an additional investment of IDR 1.2 trillion (USD 73.3 million) from shareholders Grab, Singtel, and KakaoBank.</li>
<li aria-level="1">Oretha, a senior care provider in Malaysia, completed its equity crowdfunding campaign, raising MYR 499,200 (USD 105,770) with co-investment from MyCIF.</li>
</ul>
If there are any news or updates you'd like us to feature, get in touch with us at: <a href="mailto:hello@kr-asia.com">hello@kr-asia.com</a>.

Deals in brief: Dagangan secures funding from Kejora-SBI Orbit Fund, Saison International invests in Mongolia, ten China and India deals, and more

Venture Capital

Private Equity

Investment

Merger &amp; Acquisition

Indonesia

Japan

Southeast Asia

India

… top-performing merchants. These are typically large chain stores with big ad budgets.
This change soon became palpable. In April, Douyin’s BD staff from Hangzhou and Wenzhou were asked to manage a maximum of 250 merchants each, proving that increasing GMV had become a higher priority than expanding new ones.
Recently, the newly formed Wenzhou team’s results have been good. They spent half a month tackling Metersbonwe’s Wenzhou branches, listing them on Douyin’s group buying platform, and achieved RMB 10 million (USD 1.3 million) in GMV within two days.
On-demand retail is an area where Douyin is consolidating its moat. Clothing, leather goods, watches, and automotive parts—these non-core categories of Meituan are currently the focus of Douyin’s Wenzhou team.
In mid-April, Douyin’s staff and service providers in Wenzhou posted extensively on social media, prompting merchants in these categories to open Douyin stores. Merchants previously categorized as "other" by Meituan finally saw a day of being valued.
<h3>No clear winners or losers</h3>
Meituan isn’t a company that likes subsidies, but it dislikes losing even more, Yu Huo said. So, when Douyin offers lower commissions to merchants, Meituan tends to <a href="https://kr-asia.com/with-douyin-looming-meituan-shifts-focus-back-to-its-core-business">follow suit</a>. When Douyin emphasizes profits, Meituan seizes the opportunity to reduce subsidies. In Q1 this year, Meituan’s marketing expenses were RMB 13.89 billion (USD 1.9 billion)—up 33.6% year-on-year, but down 16.8% quarter-on-quarter, far below the market expectation of RMB 16 billion (USD 2.2 billion).
The tug-of-war has replaced ambushes that colored the early battles between Meituan and Douyin, subsiding the price war, at least for now. According to LatePost, in Q1 2024, Douyin racked up around RMB 100 billion (USD 13.7 billion) in sales of local lifestyle services, but did not narrow the gap with Meituan’s in-store business. Over the past six months, the market shares of both companies remained at a 1:2 ratio—both in Wenzhou and nationwide.
Douyin picked up markets where Meituan fell short, while Meituan recaptured some territories at the cost of concessions.
Reflecting on the stock price, although Meituan once lost RMB 2 trillion in market value after Douyin’s entry, its stock price has surged 95% since this year’s spring. This has inspired belief among some investors that Meituan has <a href="https://kr-asia.com/calmer-waters-arrive-for-meituan-after-fending-off-douyin">held its ground</a>, and the worst is over.
But merchants still miss those best days that have characterized the past five years. One merchant, whose head was previously turned by Meituan and chose to delist from Douyin’s takeaway platform, eagerly looks forward to the day ByteDance buys Ele.me, despite the rumor being <a href="https://kr-asia.com/ele-me-ceo-yu-yongfu-addresses-rumors-of-potential-acquisition-by-bytedances-douyin">denied multiple times</a> by officials.
"If Douyin’s in-store group buying business replicates its success in e-commerce, what reason would ByteDance have not to buy Ele.me and take over the at-home business as well?" The merchant believes the showdown is far from over.
Of course, this is pure speculation. In today’s stringent world, even the giants can no longer afford disorderly mergers and subsidies.
The relentless subsidy wars of yesteryear are unlikely to happen again, with prolonged skirmishes, accounting, and restraint close at hand.
The business world runs its cycles in ebbs and flows. As 36Kr departed Wenzhou, some business owners were noted studying Xiaohongshu and WeChat Channels because they heard these platforms are planning to delve into local lifestyle services. BD staff from Xiaohongshu held its first presentation in Wenzhou this March, attracting owners of female-focused businesses such as entertainment, hairdressing, and aesthetic medicine.
Battles come and go, but the struggle never ends, with new challenges always on the horizon.
Note: Zhang Kan, Yu Huo, and Guo Tedi are pseudonyms.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/2839150698711688">article</a> was written by Wang Yuchan for 36Kr.

Exclusive: Skirmishes in Wenzhou bring Douyin and Meituan’s three-year showdown to life