Funding increases by 130%: Can AI-driven drug development become a reality?

Deep within a space laboratory, AI with unparalleled computational power beyond human civilization shatters the long cycle of traditional drug development. It instantly locks onto the unique solution among billions of molecular combinations, from gene sequences to protein structures, from disease targets to cures—all through algorithmic simulation. No trial and error, no waiting. Lines of code condense into drug entities within holographic light and shadow, opening an unprecedented fast lane for human disease treatment.

This future scene, once only in sci-fi blockbusters, is gradually becoming reality. AI-driven drug discovery is increasingly clear in its approach, stepping into the real world. Under capital support, it is now at a development inflection point.

However, amidst the hype, the capital market remains cautiously optimistic, favoring selective deployment over broad investment, significantly raising project entry barriers and no longer paying for mere technological concepts and narratives.

▌ Capital Focus Shifts from “Widespread Casting” to “Targeted Selection”

Hot money is accelerating into the AI drug discovery sector.

Earlier this month, Deep Intelligent Yao, along with Sinophi Intelligent (03696.HK), Jingtai Technology (02228.HK), and Jitai Technology, completed a $60 million funding round. Investors include Xinchen Capital, Jinyu Capital, Kaita Capital, with existing shareholders Dinghui Baifu and New Ding Capital continuing to add funds. Less than two months prior, the company had just completed nearly $50 million in Series D funding led by Dinghui Investment, with New Ding Capital and Sequoia China participating.

The continuous capital inflows into Deep Intelligent Yao are not isolated but reflect the overall hot trend in AI drug discovery.

By December 2025, over 350 global AI pharmaceutical companies will exist, with at least 100 in China. Most are still in early startup or embryonic stages; only about ten are in growth or maturity phases. Since 2025, several leading companies have secured large funding rounds, accelerating capital deployment.

In August 2025, Jitai Technology completed a Series D round of 400 million yuan, led by Beijing Medical and Health Industry Investment Fund and Daxing District Industrial Investment Fund. In December, Deep Shihui Technology raised over 800 million yuan in Series C funding, jointly invested by Dachen Venture Capital, Jingguorui Fund, Beijing AI Industry Investment Fund, Beijing Medical and Health Industry Investment Fund, Lenovo Venture Capital, and Yuanhe Puhua.

Listed at the end of last year, Sinophi Intelligent raised HKD 2.277 billion, the highest in Hong Kong biotech IPOs that year. Its cornerstone investors include Tencent, Eli Lilly, Temasek, Huaxia Fund, and Taikang Life. Earlier, Jingtai Technology raised about HKD 2.65 billion through a new share placement.

Data shows that in 2025, there were 32 financing events in China’s AI drug discovery sector, totaling over 6.7 billion yuan, a 130.5% year-over-year increase.

Secondary market investors told the “Science and Technology Innovation Board Daily” that the investment landscape has become multi-layered: core investors are leading VC/PE firms like Sequoia China and Hillhouse Capital backing many AI drug companies. Multinational pharma and tech giants such as Eli Lilly, Sanofi, Tencent are increasing their industry and technological investments. Government-guided funds and local industrial funds actively participate, focusing on supporting domestic innovation and industry chain implementation. Long-term capital like insurance funds and public funds are gradually entering as well.

Miao Tianyi, Managing Partner of Puzhu Capital, told the “Science and Technology Innovation Board Daily” that, based on 2025 domestic AI drug financing data, there is a “structural divergence of slight decline in transaction volume but a sharp rebound in financing amount,” with capital shifting from “broadly casting nets” to “picking the best.”

On one hand, funds are highly concentrated in top companies, with the industry’s CR5 approaching 50%, reinforcing the head effect. On the other hand, over 70% of early-stage projects (Series B and earlier) are favored, with capital preferring high-quality firms that have “closed-loop technology capabilities and proven commercialization.” These companies typically possess autonomous generative AI platforms, have pipelines in small or large molecules, and can secure high milestone payments and sales sharing from multinational pharma, with full-chain capabilities from target discovery to preclinical research.

Liu Lihua, Managing Director of CIC Zhuoshi Consulting, also said: “Currently, AI drug discovery financing shows a clear tiered pattern, with the market valuing companies that have ‘deliverable metrics,’ such as improved druggability, shorter experimental turnaround times, and higher efficiency in producing clinical candidates.”

Based on technological paths and business focus, AI drug companies mainly fall into three categories: first, platform-based firms centered on AI algorithms and drug discovery, focusing on target identification, molecular design, and druggability optimization; second, integrated companies combining AI with automation experiments, closing the loop between computational simulation and wet lab experiments; third, vertical firms specializing in specific drug types like small molecules, biologics, gene therapy, or nanodelivery, commercializing through in-house pipelines or external licensing.

Miao Tianyi pointed out that different types of AI drug companies have different valuation metrics: “First, technological barriers—AI-native biotech firms that deeply integrate AI technology have a median valuation of $78 million in 2024, nearly twice that of traditional biopharma, with companies holding proprietary algorithms or exclusive data enjoying higher premiums. Second, clinical translation progress—companies with pipelines in Phase II/III or AI-led discovery drugs are valued significantly higher than those only in preclinical stages. Third, commercialization ability—companies with milestone revenues or large overseas licensing deals have stronger valuation support.”

In August 2025, Jingtai Technology partnered with DoveTree for BD cooperation, using an “AI + robotics” end-to-end platform to develop small molecules and antibodies targeting tumors and autoimmune diseases. DoveTree gained exclusive global development and commercialization rights, with a total deal value of $5.99 billion, making it one of the largest BD transactions in China’s AI drug sector in 2025.

Within six days of listing, Sinophi Intelligent signed an $888 million oncology drug R&D collaboration with Sviya, with its four oncology projects fully or partially licensed out.

“Institutions favor ‘platform + pipeline’ dual-driven companies,” Miao Tianyi said. “These firms can generate scalable revenue through technology licensing and validate platform value via self-developed pipelines, forming a sustainable growth cycle.”

▌ Challenges in Late-Stage Clinical Trials and Commercialization

New drug development has long been seen as a “blind man feeling the elephant” gamble—each step critical, one mistake can derail the entire process. Data shows that over the past decade, global pharma invested hundreds of billions of dollars, yet only a few hundred new drugs were approved, with an overall clinical success rate below 10%. AI has broken the “double ten law” (10 years and $1 billion) that long constrained drug R&D, reshaping the development paradigm.

According to Guojin Securities, “AI can reduce new drug R&D costs by four times and increase R&D return fivefold. The commercial value of AI drugs will be 20 times higher than standard drugs and 2.4 times that of the best precision drugs.”

Li Yang, Chairman of Xili Technology, said: “AI’s most significant contribution currently is in preclinical screening, effectively solving the traditional bottlenecks of low-throughput high-efficiency screening and lengthy molecular optimization cycles.”

He added: “In the past, from hit compounds to lead compounds and preclinical candidates, the process took years—costly in traditional small molecule R&D. With AI, exemplified by two rare disease molecules RTX-117 and RTX-317, discovery cycles have been compressed to months. Relying on self-developed AI models and robotic collaboration, high-throughput screening and efficient optimization have shortened drug entry into clinical trials. This not only accelerates clinical development but also significantly improves clinical translation rates through precise modeling, directly reducing sunk R&D costs.”

In January, Xili’s small molecule pipeline RTX-117 received approval for clinical trials in spinal muscular atrophy and leukoencephalopathy, becoming China’s first Class 1 innovative drug pipeline approved for these indications. This was developed in collaboration with Jingtai Technology, leveraging AI + robotics.

Li Yang also said: “AI is gradually adding value in other drug development stages, such as integrating RNA and genomics data to more precisely select trial participants, enabling earlier and more efficient proof-of-concept. We are exploring AI’s role in post-trial data analysis and long-term patient management, which can improve trial success rates and provide real-world evidence for subsequent commercialization.”

Despite high expectations, AI drug discovery still faces long roadblocks. Currently, no AI-designed drug has been approved for market, with only a few projects in Phase III, including Generate Biomedicines’ entirely AI-designed antibody GB0895 and Jitai’s self-developed AI drug MTS004.

This indicates that deeper technological implementation and commercialization face multiple challenges. Although efficiency gains in preclinical stages are validated, the transition from preclinical to late-stage trials remains difficult.

Moreover, issues such as scarce high-quality annotated data, limited model interpretability, and incomplete wet-lab and dry-lab feedback loops continue to restrict final drugability. For many early-stage AI companies, ongoing R&D investment and BD negotiations test their survival and self-sustainability.

Regulatory adaptation is also crucial. Currently, global standards for AI-driven drug R&D, data traceability, and algorithm transparency are still evolving. How AI drugs can achieve faster approval and clearer pathways while ensuring safety and efficacy remains unresolved.

Li Yang admitted that “the AI drug industry is still in an early stage, requiring more validation. The key to industry reshuffling lies in the core steps that remain unverified—those that can early demonstrate concept validation in patients, and achieve breakthroughs in clinical trial design and post-trial data management will gain a competitive edge and seize opportunities.”

Miao Tianyi pointed out that future industry competition will favor companies with clinical validation capabilities. “Pure tech firms lacking clinical proof will be eliminated. Leading companies will expand through M&A, increasing industry concentration. Capital will form a ‘dumbbell’ layout—early-stage capital focusing on disruptive tech platforms, later-stage capital preferring mature clinical data. Overseas licensing and global clinical networks will become valuation core indicators, pushing the industry from ‘domestic competition’ to ‘global competition’.”

Regarding future opportunities, Liu Lihua believes that “the focus will shift to the deep application of multimodal technologies and biological foundational models. Breakthroughs will no longer rely solely on algorithm optimization but on integrating multi-dimensional data such as genes and proteins through multimodal tech, combined with iterative upgrades of biological models. This will enable more precise target prediction, molecular design, and druggability assessment, further shortening drug development cycles, reducing trial-and-error costs, and accelerating BD collaborations.”