Qianfeng Interview with LianLian DigiTech: From 'Cross-Border Payment Expert' to a 'Trusted Global Intelligent Financial Infrastructure'

【Ebrun Original】LianLian DigiTech, a third-party payment company, has since 2026 clearly defined its 'AI Native Globalization' strategy, evolving from a 'cross-border payment expert' to a 'trusted global intelligent financial infrastructure.' The company's business covers more than 200 countries and regions, supports transaction settlement in over 140 currencies, holds 68 global payment licenses and a Hong Kong VATP license, and connects to more than 180 e-commerce platforms worldwide. In this Qianfeng Award interview, LianLian DigiTech shared its practices and judgments in intelligent payments, enterprise-level Agent platforms, global payment networks, and localized services, centered on the modal evolution brought by AI and going global.

Ebrun: This Qianfeng Award takes 'the modal evolution brought by AI and going global' as its core theme, focusing on the deep restructuring of industrial capabilities. In the context of the third-party payment industry, how do you understand this round of industrial modal evolution jointly driven by AI and globalization? What are LianLian DigiTech's core strategic arrangements in AI capability building and global expansion? Could you specifically introduce the company's positioning and business model?

LianLian DigiTech: AI and globalization are simultaneously driving changes in how enterprises operate. On the one hand, the way enterprises use AI is gradually moving from single model calls, content generation, and assisted decision-making into deeper business links such as enterprise operations, risk management, capital management, and commercial transactions. The question enterprises focus on has also shifted from 'whether we have AI capabilities' to 'how to use AI stably, efficiently, and controllably.' On the other hand, enterprise globalization is also moving from merely exporting products and traffic into a deeper stage of localized operations. Enterprises not only need customers and orders, but also have to face payment system regulatory rules, capital management, compliance requirements, and local service systems in different countries and regions.

Based on these changes, since 2026 the company has proposed and clarified the 'AI Native Globalization' strategy, evolving from a 'cross-border payment expert' to a 'trusted global intelligent financial infrastructure.'

In terms of globalization, LianLian DigiTech continues to improve its global payment network, license compliance system, and localized service capabilities. At present, the company's business has covered more than 200 countries and regions, supports transaction settlement in more than 140 currencies, holds 68 payment licenses and a Hong Kong VATP license, connects to more than 180 e-commerce platforms worldwide, and has served more than 13.3 million customers cumulatively. Around enterprises' global operations, the company provides services such as collection, payment, acquiring, foreign exchange, virtual bank cards, and capital management, and continuously expands scenario-based solutions for cross-border e-commerce, international trade, logistics, and trade in services.

Ebrun: You mentioned the 'AI Native Globalization' strategy. Which main lines is the AI-native strategy currently advancing around? What is the core logic of each main line?

LianLian DigiTech: In terms of AI, LianLian DigiTech's AI-native strategy has moved from vision to productization and implementation, and is currently advancing around three core main lines:

First, the new infrastructure for Agentic Payment. As AI Agents gradually enter enterprise procurement, business decision-making, and transaction execution, the payment system also faces new requirements such as identity authentication, authorization management, quota control, risk control, and transaction traceability. Around this direction, we have carried out agentic payment cooperation with global ecosystem partners such as Visa and UnionPay International, and have promoted relevant capabilities into real B2B transactions, global procurement, and other commercial scenarios.

Second, the AI product matrix and service upgrades. We continue to advance the productization of AI capabilities, further applying AI to scenarios such as payment integration, marketing, customer service, compliance, foreign exchange, and enterprise operations. At present, 'Xingxiaohui' applies AI Agents to foreign exchange services such as market queries, foreign exchange transactions, and account management; intelligent marketing, intelligent customer service, and intelligent compliance further cover scenarios such as marketing customer acquisition, customer service, and compliance management.

Third, the enterprise-level Agent platform. On this main line, we continue to build enterprise-level AI capabilities from underlying model infrastructure to upper-layer Agent platforms. Among them, 'LianLian Zhishu,' as an enterprise-level AI computing power infrastructure platform, uniformly connects and schedules more than 300 mainstream models, providing enterprises with model access, intelligent routing, and operations management capabilities.

At the same time, AI technology has also been further integrated into existing global payment business. Capabilities such as AI KYC, AI KYB, AI KYT, and intelligent operations support have been deeply embedded in product R&D, risk control, daily operations, and customer service processes.

Therefore, our 'AI Native' is not about adding a few AI products on top of payments, but about continuously integrating AI capabilities into the global payment and enterprise operation systems around the core payment chain, enterprise services, and enterprise-level Agent infrastructure.

Ebrun: In which core business links has your company currently implemented AI/large model technology? What kind of distinctive AI or digital-intelligence competitiveness system has been formed?

LianLian DigiTech: At present, LianLian DigiTech has driven AI into multiple links including payments, compliance, risk management, capital management, operations, customer service, marketing, and commercial transactions, and has formed a capability system corresponding to the three main AI strategic lines.

First, in the core payment chain and the new infrastructure for intelligent payments, capabilities such as AI KYC, AI KYB, AI KYT, and intelligent operations support have been deeply applied in processes such as product R&D, enterprise review, risk control, transaction monitoring, operations, and customer service. On this basis, we are further advancing the building of Agentic Payment capabilities. As AI Agents begin to participate in procurement and transaction execution, the payment system needs to adapt to new transaction entities and business processes. Through cooperation with Visa, UnionPay International, and others, we have applied relevant capabilities to real B2B agentic transactions, global procurement, and other scenarios, promoting AI to move from assisted decision-making further into commercial transaction links.

Second, in the AI product matrix and service upgrades, we continue to apply AI capabilities to specific enterprise operation scenarios. For example, LianLian Open API integrates Skills to encapsulate cross-border payment API capabilities into Skills that can be called by AI programming tools, improving the efficiency of payment technology integration; 'Xingxiaohui' applies AI Agents to foreign exchange services such as market queries, foreign exchange transactions, and account management; intelligent marketing, intelligent customer service, and intelligent compliance further cover scenarios such as marketing customer acquisition, customer service, and compliance management.

Third, in the enterprise-level Agent platform, we are building platform capabilities from underlying model resource management to upper-layer agent operation. Among them, LianLian Zhishu mainly addresses the issues of 'how to connect, how to select, how to tune, and how to manage' in enterprises' use of multiple models. At present, the platform can uniformly connect and schedule more than 300 models including DeepSeek, Qwen, MiniMax, and GLM, and provides scheduling strategies such as cost priority, quality priority, latency priority, and high availability. It also has capabilities such as multi-model and multi-vendor redundancy, failover, call records, visualized billing, and usage management, providing underlying support for enterprises to move from single-point trials to multi-scenario, large-scale AI applications.

The company is forming a multi-layer capability system from AI model resource management and enterprise-level Agent platforms to specific business applications and intelligent payment transactions, enabling AI not only to be called but also to further enter enterprises' actual business processes.

Ebrun: You mentioned that LianLian Zhishu uniformly connects and schedules more than 300 models and provides scheduling strategies such as cost priority, quality priority, latency priority, and high availability. What specific problems does this platform solve in enterprises' use of multiple models? Why do enterprises need such a 'model scheduling layer'?

LianLian DigiTech: LianLian Zhishu mainly addresses the issues of 'how to connect, how to select, how to tune, and how to manage' in enterprises' use of multiple models. At present, the platform can uniformly connect and schedule more than 300 models including DeepSeek, Qwen, MiniMax, and GLM, and provides scheduling strategies such as cost priority, quality priority, latency priority, and high availability. It also has capabilities such as multi-model and multi-vendor redundancy, failover, call records, visualized billing, and usage management. For enterprises, if each business line connects to models and manages vendors on its own, the problems of duplicate construction and fragmented governance will quickly emerge; the value of LianLian Zhishu is to make model access, intelligent routing, and operations management a shared foundation, providing underlying support for enterprises to move from single-point trials to multi-scenario, large-scale AI applications.

Ebrun: Could you share one of the most representative innovative practice cases of the 'AI value chain'? What common industry pain points did it solve, and what perceivable and quantifiable value did it create?

LianLian DigiTech: At present, a relatively representative practice is the first real B2B agentic transaction in Greater China completed by LianLian DigiTech in cooperation with Visa. In this scenario, LoopXPay, an AI agent developed by LianLian, completed procurement demand identification, supplier search and comparison, ordering, and payment within a single workflow. The entire process operated within the spending limits and approval parameters pre-set by the enterprise. The significance of this case is that AI is no longer only undertaking information queries or assisted decision-making functions, but has begun to truly enter the execution process of B2B commercial transactions.

When AI Agents participate in enterprise procurement, the payment system needs to solve a series of new basic issues, such as how to identify and verify the identity of agents, how to limit agents' transaction permissions, how to control payment quotas, how to conduct risk management, and how to ensure that the entire transaction process is transparent and traceable.

Ebrun: You also mentioned advancing agentic payment cooperation with UnionPay International to promote the implementation of scenarios such as the 'human-in-the-loop' model for global procurement and AI Token top-ups. What directions does this cooperation specifically focus on? Compared with the cooperation with Visa, what are the different priorities?

LianLian DigiTech: Around the direction of agentic payments, we are also advancing cooperation with UnionPay International to promote the implementation of scenarios such as the 'human-in-the-loop' model for global procurement and AI Token top-ups, and are carrying out cooperation in directions such as overseas agentic payment scenarios, merchant expansion, and collaborative R&D of AI technology. The first real B2B agentic transaction in Greater China completed with Visa focuses on verifying the feasibility of AI agents in the real B2B transaction execution process; the cooperation with UnionPay International focuses more on the implementation of specific scenarios such as the 'human-in-the-loop' model in global procurement scenarios and AI Token top-ups, as well as overseas merchant expansion and technical collaboration. The two are complementary and jointly advance agentic payments from verification to scenario-based implementation.

Ebrun: In terms of overseas layout, which core overseas markets has LianLian DigiTech currently entered? Based on the local industrial environment, regulatory rules, and market characteristics, what key localization adjustments have been made?

LianLian DigiTech: Globalization has always been one of the core directions of LianLian DigiTech's development. With the Chinese market as our foundation and Hong Kong, China as our international hub, we continue to improve our payment network, compliance system, and localized service capabilities covering major global markets, serving enterprises as they move from cross-border transactions further toward global operations.

As of June 30, 2026, the company's service scope has covered more than 200 countries and regions, supports transaction settlement in more than 140 currencies, holds 68 global payment licenses and a Hong Kong VATP license, and connects to more than 180 e-commerce platforms worldwide. In the first half of 2026, the company further improved its layout in the North American and Middle East markets. In April, it obtained Canada MSB-related qualifications through the acquisition of an entity, and in May it obtained the Category 3D payment license from the Dubai International Financial Centre, continuously enhancing compliance access and local service capabilities in key markets.

The global layout has also been reflected in business growth. In the first half of 2026, the company's global payment total payment volume (TPV) reached RMB 249.9 billion, up 25.9% year on year; global payment revenue was RMB 601 million, up 27.1% year on year; global payment gross profit was RMB 426 million, up 23.8% year on year. On the basis of B2C cross-border e-commerce, we have further deepened B2B trade, as well as service trade scenarios such as aviation and travel, logistics, advertising, and overseas study payment, and service demand has continuously extended from single cross-border collection and payment to multi-currency settlement, capital management, and overseas local operations.

LianLian DigiTech's 'localization' is not merely about establishing local teams or completing language adaptation, but about forming comprehensive service capabilities in different markets that match local regulatory requirements, payment systems, and enterprise operating habits. On the one hand, we continue to improve the global license and compliance network, providing a foundation for entering different markets; on the other hand, with the overseas headquarters established in Hong Kong, China, the company continues to integrate domestic and overseas resources, strengthen local operations, compliance, risk control, and product capability building in key markets, and enhance service depth through the global payment network, local financial institutions, and industry ecosystem partners. At the same time, we also continue to promote synergy between domestic payment capabilities and the global payment network, providing more complete payment support for enterprises' cross-market operations.

At the specific business level, we continue to provide services such as collection, payment, acquiring, foreign exchange, virtual bank cards, and capital management around enterprises' global operation needs, and adapt products and services according to the account systems, payment habits, regulatory rules, and industry characteristics of different countries and regions.

For LianLian, globalization is not simply expanding geographical coverage, but moving from 'covering more markets' further toward 'deeply serving local markets.' As Chinese enterprises move from 'cross-border trade' to deeper 'global operations,' we are also continuously strengthening the synergy among the global payment network, compliance system, localized operations, and technical capabilities, providing infrastructure support for enterprises' long-term and stable operations in different markets.

Ebrun: In the process of overseas expansion, what core support do AI and digital capabilities provide for cross-border business?

LianLian DigiTech: AI and digital capabilities have now entered multiple core links of our global payments and enterprise services, and have formed synergy with our global layout.

First, in the global payment, compliance, and capital management systems, we continue to promote the embedding of AI capabilities into product R&D, risk control, daily operations, and customer service processes. Capabilities such as AI KYC, AI KYB, AI KYT, and intelligent operations support have been applied in links such as identity verification, enterprise review, transaction monitoring, and risk management; at the same time, we are also promoting AI to further enter enterprise capital management scenarios. For example, the foreign exchange AI assistant 'Xingxiaohui' has covered services such as market queries, account management, and foreign exchange transactions. By combining AI with payment, compliance, and capital management processes, the level of digitalization and intelligence in cross-market services can be further improved.

Second, in enterprise-level AI infrastructure building, we continue to improve enterprise multi-model access, scheduling, and management capabilities through LianLian Zhishu. LianLian Zhishu currently uniformly connects and schedules more than 300 model resources. Through intelligent routing, operational assurance, and operations management, it helps enterprises reduce the complexity of duplicate multi-model access and fragmented management, providing infrastructure support for AI to move from single-point trials to multi-scenario, large-scale applications.

Third, in intelligent commerce and payment links, we have completed the first real B2B agentic transaction in Greater China with Visa, and have carried out cooperation with UnionPay International around directions such as agentic payments for global procurement, promoting AI to further enter the cross-border commercial transaction process.

Therefore, LianLian DigiTech's digital-intelligence building is not separate from payments and globalization business, but continuously embeds AI into specific processes such as payments, compliance, capital management, enterprise operations, and commercial transactions, jointly supporting enterprises' global operations together with the global payment network and localized service capabilities.

Ebrun: Could you introduce the iteration of LianLian DigiTech's profit model? How do you view the evolution trend of profit models for industrial internet companies? How can Token-based payment become a standard industry revenue model?

LianLian DigiTech: At present, LianLian DigiTech's main businesses include global payments, domestic payments, and value-added services. The profit model is gradually evolving from being centered on payment transaction services to a multi-layer revenue structure of 'global payment foundation + value-added service growth + AI technology and enterprise service extension.'

First, global payments remain the company's core source of revenue and profit. In the first half of 2026, the company achieved revenue of RMB 880 million, up 11.9% year on year; among this, global payment business revenue was RMB 601 million, up 27.1% year on year, global payment TPV reached RMB 249.9 billion, up 25.9% year on year, and global payment business gross profit was RMB 426 million, with a gross margin of 70.8%. As the global license network, localized service capabilities, and business scenarios continue to expand, the company has further deepened B2B trade on the basis of B2C cross-border e-commerce and extended to service trade fields such as aviation and travel, logistics, advertising, and overseas study payment. The continued growth of the global payment business provides a stable foundation for the company's overall profit model.

Second, value-added services are becoming an important source of revenue growth. In the first half of 2026, the company's value-added service revenue reached RMB 164 million, up 83.2% year on year, mainly driven by growth in virtual bank card-related business. At the same time, the company is also expanding more enterprise services around scenarios such as intelligent marketing Agents. As customer demand further extends from single fund collection and payment to capital management, operational efficiency improvement, and support for global operations, the company's revenue sources are also further extending to more payment-related value-added services.

Third, AI productization is expanding new service boundaries. In the first half of 2026, the company advanced the implementation of AI capabilities around three main lines: the new infrastructure for Agentic Payment, the AI product matrix and service upgrades, and the enterprise-level Agent platform. It has launched related products and capabilities such as LianLian Zhishu, 'Xingxiaohui,' and intelligent marketing. AI is gradually moving from internal R&D efficiency improvement and business empowerment toward external productization and serviceization, further expanding the space for enterprise technology services and intelligent services.

Ebrun: You mentioned that the company has 'three growth curves.' What do they specifically refer to? What is the relationship among these three curves?

LianLian DigiTech: From the 'three growth curves' proposed by the company, the evolution path of the profit model is relatively clear: the first curve is to continue strengthening the global payment foundation and focus on high-value customers; the second curve is to promote the implementation of AI productization, further extending services from payments to enterprise growth and intelligent operations; the third curve is to continuously improve the Fintech and AI industry ecosystem through industrial investment and ecosystem cooperation. The three curves are not parallel but progressive layer by layer: global payments provide the foundation and customer base, AI productization expands service boundaries on top of the foundation, and industrial investment and ecosystem cooperation supplement capabilities and scenarios for the first two curves.

As an annual bellwether in the TO B field, the '2026 8th Ebrun Industrial Internet Annual Conference and First Industrial Agent Conference' will be held in Beijing on November 20, 2026. The theme is set as 'the modal evolution brought by AI and going global,' focusing on in-depth deliberation of frontier topics such as the implementation of industrial agents, value reconstruction of AI supply chains, data elements and the Token economy, and the new ecosystem of industrial globalization. The conference will bring together academic authorities, benchmark central and state-owned enterprises, industrial leaders, investment institutions, and service ecosystem partners, gather core industrial forces, break down the paths of industrial digital-intelligence transformation, and share frontier practical experience. At the same time, the annual industrial internet report and benchmark cases will be released, and a decision-makers' closed-door meeting and the 'Qianfeng Award' ceremony will be held.

This article was first published on the official website of Ebrun.