The competition in enterprise AI is ultimately not just a contest of model capabilities, but rather a race in context quality, collaboration infrastructure, and organizational execution.
The automotive industry chain is validating a proposition applicable to many other industries: the competition in enterprise AI ultimately hinges not only on model capability, but more critically on context quality, collaboration foundation, and organizational action capacity.
From New-Energy Automakers to the Entire Industry Chain
Feishu Builds Collaborative Advantages in the Automotive Industry
As the automotive industry shifts from single-vehicle competition to organizational competition and expands from the Chinese market to global markets, what determines an enterprise's pace of development is no longer just a product or technology, but also whether the entire organization and supply chain can respond quickly. An engineering change, material anomaly, or quality issue often spans multiple departments, systems, and companies—wherever the slowest link exists along the chain, it could delay overall vehicle development and delivery.
Zhang Di, Vice President of Commercialization at Feishu, said the company’s automotive service scope has expanded since 2021 from new-energy automakers to larger vehicle groups, component manufacturers and multinational automakers. Feishu names SAIC Motor, Seres, Sunny Optical, Horizon Robotics, Desay SV and Baolong Technologies among its customers. These names, the depth of use and the extent of cross-company collaboration are vendor-reported claims that buyers should verify through reference calls and a scoped pilot.
Feishu says organizations can retain parts of their existing systems while using Feishu Projects, multi-dimensional tables, automation, its open platform and AI features to connect R&D, production, quality and supply-chain workflows. Whether fragmented data can become actionable tasks and usable AI context depends on the actual integrations, permissions and data quality in each deployment.
The automotive industry has become a typical case for enterprise AI transformation: large-scale organizations, complex processes, long industry chains, and extremely high demands for safety, quality, and delivery. If AI can succeed within the automotive industry chain, its experience can potentially be replicated across consumer electronics, smart manufacturing, and other multi-party collaboration scenarios.
From R&D to Delivery
Enterprises Advance OEM-Supplier Collaboration into Real Processes
OEM-supplier collaboration refers to coordination between vehicle manufacturers and component partners across R&D, manufacturing, supply-chain and quality workflows. Cross-company work can otherwise fall back on meetings, email attachments and manual follow-ups. At the conference, NIO, Dongfeng Yipai, Shengyi Electronics, Baolong Technologies and SAIC Group presented their own accounts of using AI and collaboration tools in these processes.
According to NIO’s conference presentation, it built a Partner Industrial Development Management (IDM) platform with Feishu Projects to bring partner tasks, workflows, deliverables and risk information into one operating view. The company described tasks and status moving between upstream and downstream participants; the actual partner coverage, traceability and AI use should be verified with the case owner.
Dongfeng Yipai shared its practice in collaborative material supply planning. Reporting a supplier base of over 400, they described using Feishu multi-dimensional tables, group bots, and automation to convert material shortage alerts into targeted tasks. Suppliers provide online feedback on inventory readiness, vehicle dispatch, and estimated arrival time, with all responses consolidated into a single record. AI then performs root cause analysis based on response data, transportation logs, and historical anomalies, driving review processes. Dongfeng Yipai stated: "Structured business data is where AI enters operations."
According to Shengyi Electronics’ conference presentation, its short-delivery workflow connected order data, processes and responsible staff while retaining existing systems. The team described using RPA, multi-dimensional tables, automation and Doujia Work for data extraction, exception prompts, overdue escalation and report drafts. Buyers should ask how delivery-confirmation time and complaint volume were measured against a documented baseline.
Baolong Technologies’ Vice President Huang Junlin, under the theme “People stay still, data moves,” presented a systematic upgrade of internal R&D collaboration. Leveraging Feishu Projects, Baolong established a project management middleware platform integrating ALM, PLM, FMEA, and design tools, bringing APQP, engineering changes, review management, and risk control into one unified system. Schedule changes trigger automatic updates, engineering changes flow online with full traceability, and risks from past projects become checklist items for new ones. Rather than relying on reported percentages, a CIO evaluating a similar upgrade should define the team's own baseline for process adherence and traceability, then measure against it in a pilot.
Wang Conghe, Dean of SAIC Group’s Innovation Research and Development Institute, pointed out that the automotive industry is shifting from single-technology competition to ecosystem-level competition. Vehicles are rapidly evolving into "super mobile intelligent spaces" and progressing toward embodied intelligence, necessitating a complete restructure of R&D organizations, collaboration models, and industrial relationships. While AI enhancing individual efficiency is just the beginning, the future competitive edge lies in transforming R&D from being "driven by human experience" to being "process-driven," thereby achieving systemic efficiency gains.
According to SAIC’s conference presentation, it is changing R&D workflows for vehicle product changes, model projects and OEM-supplier collaboration. SAIC also described a joint project with United Electronic Technologies involving multiple engineers and process nodes; buyers should verify the project scale and any schedule-sync savings with the case owner. Wang Conghe said, "AI does not lack computing power—it lacks context," and presented continued process capture and an open, collaborative AI-native R&D system as future directions.
The practices of these companies collectively reveal a trend: the traditional way of collaboration—relying on manual information transfer and repeated status chasing—is giving way to a new model driven by automated workflows, self-flowing data, and AI-powered early risk warnings. The value of Feishu is likewise expanding—from internal enterprise collaboration to enabling industry-wide collaboration between OEMs and component suppliers to jointly execute projects.
Feishu Hosts Context
Doujia Work Transforms Context into Productivity
Xie Xin, CEO of Feishu, noted in his keynote speech that most tasks within enterprises do not occur in isolation, but span documents, messages, meetings, spreadsheets, workflows, and interpersonal collaboration. As Agents enter enterprises, work styles will shift from "humans using software" to "humans and Agents completing tasks together within the same environment." Feishu provides the setting—knowledge, processes, and permissions—where work happens, becoming critical infrastructure that enables Agents to understand context, invoke tools, and collaborate with people.
According to the conference announcement, ByteDance launched the Agent product and brand "Doujia Work" on August 25. The vendor says it can draft documents, spreadsheets and presentations, operate browsers and local computers, and work within Feishu; buyers should confirm current regional availability, permission controls and feature scope. The same presentation described Feishu context and an authorized Agent acting on that context as complementary layers rather than substitutes.
The newly announced 2026 "AI Smart Manufacturing, Together We Fly" manufacturing efficiency acceleration program, as described by the vendor, is intended to support Shanghai-based manufacturers through enterprise-grade AI experience, model invocation support, OEM-supplier collaboration incentives, transformation assessments, scenario templates, benchmark visits, and policy alignment. Eligibility criteria and program terms have not been independently confirmed and should be requested directly from the vendor.
The organizer’s account portrays Feishu’s automotive work as expanding from collaboration inside individual organizations to projects involving vehicle groups, component suppliers and some multinational companies. Whether a deployment connects upstream and downstream participants or creates usable AI context depends on its integration scope, permissions and data-governance results.
The automotive industry chain is once again validating a proposition relevant to many industries: the competition in enterprise AI is ultimately not just about model capabilities, but more fundamentally about context quality, collaboration foundation, and organizational action capacity.