Produced by | Bullet Finance

Art Editor | Qianqian

Reviewed by | Songwen

After 27 years of deep involvement in financial IT, Yusys Technologies has significantly accelerated its AI layout (strategic AI deployment).

Recently, Bullet Finance observed that Yusys Technologies launched the SSP platform, aiming to connect originally fragmented data, knowledge, tools, and business scenarios within banks, propelling AI applications from isolated use cases toward full-process intelligence. Concurrently, the company has successively secured multiple bank AI projects. Its self-developed "Xingrui Intelligent Credit Investigation Agent" was also included in the "2026 China AI Agent Pioneers" list.

Notably, Yusys Technologies previously made a strategic investment in Yuezhi Anmian through a dedicated fund, extending its business reach into areas such as embodied intelligence, model inference, and computing power services. From AI agent products to foundational AI capabilities, and from banking scenarios to intelligent applications, Yusys is gradually completing its capability map across the financial AI industry chain.

Public data shows that Yusys Technologies (300674.SZ) serves clients including the People's Bank of China, China Development Bank, other policy banks, the six major state-owned banks, twelve joint-stock banks, as well as hundreds of small and medium-sized banks and rural credit cooperatives. According to IDC's "China Banking IT Solution Market Share, 2024," the company ranks second in this market. A 2025 report by the Zhongguancun Internet Finance Institute indicates that it holds the largest overall market share among listed companies and leads in segments such as regulatory reporting, retail credit, online lending, and corporate online banking.

Why is a financial IT company that has long built banking systems now intensively focusing on AI?

A closer look at current AI+finance trends reveals that Yusys Technologies aims to transform its 27 years of accumulated financial business understanding and engineering capabilities into platform assets that can be leveraged by more banks, thereby reopening its growth trajectory.

As banking digitalization transitions from the "system construction phase" to the "intelligent operations phase," the competitive logic for fintech companies is evolving. The financial business understanding, industry experience, and engineering delivery capabilities accumulated over the past two decades are becoming critical assets in the AI era.

Over the past twenty-plus years, banks' digital investment has followed a clear trajectory: migrating offline operations into systems and completing process digitization and standardization through platforms for core banking, credit, risk control, and regulatory reporting. During this infrastructure development wave, financial IT companies primarily generated revenue through project contracts.

However, as major systems are now largely in place, this growth model is approaching its limits. While banks still have upgrade, migration, and maintenance needs, simply adding more systems no longer delivers significant efficiency gains. Under pressure from narrowing net interest margins, stricter risk management, and rising demands for refined operations, banks are increasingly prioritizing AI integration into workflows as an urgent imperative.

(Image / Shutterstock, under VRF protocol)

Credit investigation exemplifies this transformation. To assess a company's operational capacity and credit risk, account managers must analyze multi-source information including business registration, tax, financial reports, judicial records, public sentiment, and supply chains—much of which exists in unstructured formats like contracts, reports, invoices, and web text. For complex projects, due diligence on a single enterprise often exceeds one week, with most time consumed by repetitive tasks like data collection, information entry, and report compilation, leaving limited time for actual risk assessment.

Large models and intelligent agents now offer a new solution. AI can first perform data parsing, cross-verification, and risk clue extraction, then generate corporate profiles and due diligence reports according to existing bank rules, freeing account managers to focus on judgment and decision-making. Similar shifts are occurring in knowledge management, compliance review, and marketing operations: AI's value is expanding from answering questions to orchestrating multiple systems, invoking different tools, and executing complete tasks.

Consequently, banks' evaluation criteria for AI projects are changing. Previously, system procurement focused on functional realization and successful deployment. Today, banks prioritize metrics like reduced due diligence cycles, lower labor costs, more accurate risk identification, and sustained business value creation through technology. Technology suppliers must now deliver capabilities that continuously improve business outcomes.

This is also disrupting the financial IT industry's longstanding "headcount × project" pricing model. Traditionally, project acceptance marked the end of phased revenue; sustained growth required winning new projects and increasing delivery staff, with revenue expansion often accompanied by rising costs. However, as AI reshapes business workflows, new value indicators emerge: platform utilization sustainability, capability reusability across clients, and long-term service revenue potential.

This does not imply the immediate disappearance of traditional project-based models, but the market must reevaluate fintech growth quality. Companies that can institutionalize banks' systems, knowledge, and processes into replicable AI capabilities are more likely to break through human-driven growth limits.

Yusys Technologies' intensive deployment of financial intelligent agents and the SSP platform aligns precisely with this trend.

In the AI era, banks are unlikely to lack models.

From open-source models to cloud vendor services, financial institutions now have an expanding array of technical foundations to choose from, with model capabilities rapidly iterating. Simply integrating a large model offers little differentiation and may not solve business problems. The real challenge lies in making AI understand financial operations, integrate into complex workflows, and comply with banks' security, compliance, and business rules—this is the crux of AI implementation in finance.

(Image / Shutterstock, under VRF protocol)

Yusys Technologies' strength lies in knowing which business processes AI should enter, what rules to follow, and how to securely connect with banking systems.

Having participated in extensive credit, risk control, marketing, and regulatory system construction, Yusys has developed deep insights into banks' operational rules, workflows, and compliance boundaries. However, this expertise previously resided primarily with consultants, product managers, and project teams: each new bank engagement required reassembling personnel, clarifying requirements, and completing custom development.

Large models now provide a new transformation path. Yusys attempts to convert processes and experience into foundational data such as knowledge bases, generating various agents and skills to transform expertise previously reliant on senior staff into software-callable capabilities. Financial experience thus evolves from serving individual projects to continuous accumulation through product iteration.

Yet new challenges arise: while bank AI applications grow rapidly, many agents remain siloed. Customer service agents cannot access credit system data, marketing agents cannot invoke compliance rules, and knowledge bases built by different departments remain isolated. Increasing agent numbers does not necessarily improve overall efficiency and may create new information islands.

Recently, Yusys Technologies introduced SSP (Smart Scene Partner). Simply put, this is an "all-scenario intelligent operations platform" for financial institutions' intelligent transformation. It comprises a knowledge middleware layer and an intelligent delivery foundation: the former processes regulatory documents, institutional rules, and expert experience into AI-understandable knowledge cards and skills; the latter orchestrates data, knowledge, tools, and personnel into workflows for credit, compliance, and report generation.

Thus, SSP establishes a loop where "data becomes knowledge, knowledge evolves into capabilities, capabilities enter scenarios, and scenario outcomes enrich the platform." Banks gain not isolated AI tools but an intelligent operations platform that continuously accumulates knowledge, reuses experience, and enables process tracking and auditing.

Critically, this product offers replication potential. While banks' data and internal rules remain isolated, platform architecture, common workflows, and financial business understanding can be reused. Each bank deployment further enriches the platform with general business knowledge and intelligent capabilities, presenting subsequent institutions with a more industry-savvy expert agent.

This embodies Yusys' envisioned knowledge flywheel: more projects yield richer general capabilities; more mature products accelerate deployment efficiency for new banks. Experiences previously dispersed among individuals and projects thus become platform assets that continuously accumulate and are repeatedly utilized.

AI's deeper impact on Yusys Technologies lies in transforming its underlying business model. Understanding this layer reveals the true direction of its potential valuation logic shift.

Traditional financial IT services face inherent human efficiency ceilings: each new order typically requires proportional delivery staff increases, with scale expansion accompanied by rising management costs and delivery risks. When valued via PS or PE multiples, such companies receive relatively low valuations, reflecting the assumption that growth depends on continuous capital and labor input.

Conversely, AI and platformization are fostering a distinct value chain. As previously described, once the "platform deployment—business utilization—knowledge accumulation—capability optimization—sustained operations" knowledge flywheel begins spinning, Yusys' partnership with banks will evolve from "turnkey projects" to "companion-style intelligent operations," potentially exploring new business models including platform licensing fees, annual service subscriptions, API call metering, and model iteration maintenance—sustained, high-margin revenue streams that would improve market valuation expectations.

More importantly, AI is compelling Yusys to revolutionize its internal delivery methods. Bullet Finance learned that the company has developed AI-assisted tools for its delivery teams, already piloted in multiple projects with notable efficiency gains in code generation and testing.

From an industry competition perspective, Yusys maintains relatively clear positioning advantages in the AI era. Compared to AI-native startups seeking scenarios from scratch, Yusys possesses over 20 years of financial core system construction experience, deep understanding of complex banking processes like risk control, credit, and anti-money laundering (the "tacit knowledge"), and established production-grade trust relationships with hundreds of financial institutions including the six major state-owned banks and twelve joint-stock banks. This provides a scalable service foundation where AI implementation requires "embedding" rather than "ice-breaking."

Evidently, Yusys Technologies is exploring its role as a key participant in financial AI infrastructure. This explains its investments in KIMI and embodied intelligence deployments—seemingly technical positioning moves, but substance (in essence) building value connectors spanning "upstream models and computing power—middleware intelligent engines—downstream diversified financial scenarios."

Of course, business model transformation will not happen overnight. Yusys Technologies still faces a transitional period between old and new growth drivers. Bullet Finance learned that the company began strategic adjustments a year ago, focusing on large and medium-sized clients while increasing AI R&D investment—initially manifesting as revenue pressure and profit contraction. Simultaneously, AI product scalability validation, cultivating banks' willingness to pay for operations, and ramping up new revenue structures all require time and patience.

In the future, market valuation benchmarks for Yusys Technologies may gradually shift from "financial IT service provider" to "financial AI infrastructure service provider with platform replication capabilities and long-term operational value." This transition represents both the ultimate return on technological investment and the best test of strategic resolve.

(Image / Shutterstock, under VRF protocol)

The fintech competition in the AI era has only just begun. The true differentiator in this industrial restructuring will be the ability to institutionalize industry experience into reusable intelligent capabilities and transform project accumulations into long-term value assets.