Jingtai Technology (02228) 2026 Mid-term Performance Decoding: AI4S Commercialization Breakthroughs at High Speed, Multimodal Pipeline Accelerates to Realize Value

On August 19, JingTai Technology (02228) released its mid-year performance announcement for 2026. This is the first interim report since JingTai upgraded from an "AI Pharmaceutical Service Provider" to an "AI4S platform that integrates digital research and physical experimentation." According to the financial report, the company achieved revenue of 394 million yuan (RMB) in the first half of the year, down from 517 million yuan in the same period last year, primarily due to the high base effect from a significant upfront payment of 51 million USD for a pipeline licensing project that was recognized last year; after excluding this impact, operating revenue increased by 73.8% year-on-year. This adjustment corresponds to the shift in revenue structure from "one-time pipeline licensing" to "platform-based, recurring revenue" the rapid growth of 136.4% in AI4S business, the intelligent Siasun Robot & Automation laboratory, and intelligent services is a direct reflection of this transition. JingTai is advancing AI from a predictive tool in the digital world to an industrial-grade infrastructure capable of autonomously completing scientific tasks in real experiments. In the first half of the year, the company reported a net loss of 225 million yuan, with an adjusted net loss of 106 million yuan, mainly due to a 66% year-on-year increase in R&D expenses, which were directed toward independent laboratories, intelligent system developments, ongoing pipeline projects, and multi-modal technology platform construction. As of June 30, 2026, the company held a total cash balance of 8.671 billion yuan, providing robust funding reserves to support continued R&D investments and consolidate the companys leading position in the AI4S field. The cash reserve of 8.671 billion yuan, combined with the 66% year-on-year increase in R&D expenses, indicates that the company has entered a synchronized investment phase in platform-pipeline-infrastructure a typical capital rhythm for transitioning through concept validation to the commercialization of large-scale industries. It is understood that in drug development, JingTai Technology has the most comprehensive multi-modal AI drug discovery platform in the industry, covering four major categories: small molecules/molecular glue, antibodies, peptides, and small nucleic acids. The platform has accumulated a large amount of proprietary standardized data, training industry-leading generation and prediction models, significantly outpacing peers in pipeline iteration speed. Pipeline iteration efficiency is a major competitive advantage for JingTai Technology, compressing "trial and error-driven" processes into "design-driven" engineering processes. The company disclosed for the first time that several self-developed projects have achieved key experimental progress: the molecular glue pipeline achieved picomolar-level target protein degradation activity within a single quarter, whereas similar industry projects typically take 12-18 months from target validation to obtaining high-activity degraders; oral cyclopeptide and IgA nephropathy small nucleic acid projects have quickly obtained high-quality efficacy data, with the IgA nephropathy siRNA project requiring only about seven months from sequence design to in vivo efficacy validation in non-human primates the industry usually requires 12-24 months. Currently, three self-developed and enabled pipelines have entered the clinical stage, with over ten in the IND approval and IND-enabling stages; it is expected that by 2027, more than five additional pipelines will enter the clinical stage, and about twenty pipelines will reach the PCC stage. On the commercialization front, the company is simultaneously advancing two pathways: asset BD licensing and traditional R&D service collaborations. Domestic and international BD efforts have yielded fruitful results, securing strategic partnerships with international pharmaceutical companies for AI drug discovery with a potential total value exceeding 400 million USD, significantly raising expectations for asset value release; traditional R&D service collaborations are progressing in an orderly manner, forming a dual-driven model of "asset licensing + platform R&D services" that provides a clear competitive advantage over most peers. The Siasun Robot & Automation laboratory and synthesis business of JingTai Technology have achieved large-scale implementation and are among the industrys top tier. They have already secured contracts and repeat orders from leading pharmaceutical companies such as Eli Lilly, extending from compound management to the HTE platform. The overseas project with JW in Korea has completed all delivery and acceptance, while multiple benchmark projects worth tens of millions have landed domestically, with business boundaries extending into new material tracks such as perovskites and lithium batteries. The company has built an industry-leading full-process AI autonomous decision-making and synthesis capability, with related businesses experiencing rapid growth. On the technical side, the SureRoute self-evolving retrosynthetic model has lowered the chemical illusion rate to 4.6%, while the first synthetic route recommendation accuracy reached 74.3%; simultaneously, the two self-developed solutions, AgenticSynthesis (intelligent autonomous molecular synthesis) and AgenticHTE (intelligent autonomous high-throughput experiment), have achieved commercialization, realizing fully autonomous operations from AI solution design to Siasun Robot & Automation experimental execution. At the platform level, over 300 automated workstations have been deployed globally. Leveraging a four-layer industrial-grade technology system composed of intelligent agents, scientific AI, physical intelligence, and a high-quality data foundation, the company has established a closed-loop integration of digital simulation with real Siasun Robot & Automation experiments; more than 300 automated workstations now cover over 20 types of R&D scenarios; more than 500,000 real experimental records have been accumulated, with approximately 80% being rare negative samples from public literature; over 50,000 new reaction yield data points and 300,000 process data points are added monthly; and over 100 drug and material projects have been serviced cumulatively. In July 2026, the company publicly launched the XtalPiScience scientific intelligence platform and the GeniusAgent scientific intelligent matrix, opening its internally validated complete set of research infrastructure to global research institutions, outputting underlying AI4S technical capabilities. JingTai's barrier lies not in a single model, but in the system of data generationexperimental executionmodel iteration. Even if competitors obtain similar models, they find it challenging to replicate the real experiments and data flywheel. From an investment perspective, there is a significant expectation gap concerning JingTai Technology in the secondary market. Industry talents such as Jeff Dean have successively laid out the field of AI autonomous scientific research, validating the long-term development potential of this sector, but many overseas peers are still in the concept validation phase. Among overseas benchmark companies, according to market news, AI drug development leader Isomorphic Labs has a valuation of about 20 billion USD, yet its commercialization path has not been fully disclosed; LilaScience, which focuses on AI autonomous laboratories, has a valuation of approximately 8 billion USD but lacks large-scale commercial orders. In contrast, JingTai Technology boasts actual business scenarios, confirmed revenues, and rapidly iterating pipeline reserves. In reference to first-tier market benchmarks, market analysis suggests that JingTai Technology has already achieved large-scale commercialization rather than just telling scientific stories, and its comprehensive progress affords a 3-5 year lead over similar enterprises. The current valuation of over 30 billion HKD may be significantly underestimated. Looking ahead, the AI for Science industry is evolving from point tools to autonomous scientific discovery. With a complete business model that has already been validated, JingTai Technology stands at a critical inflection point for the release of technological dividends and at the forefront of the industry; the commercialization of XtalPiScience, the sequential entry of pipelines into clinical stages, and the realization of BD collaboration milestones will constitute the core value-driving factors for the company, solidifying its long-term growth logic.