Aiming to Create AGI in the Pharmaceutical Industry, "Deepwise" Secures Another $60 Million in Funding | Exclusive from 36Kr
Text by | Hu Xiangyun
Edited by | Hai Ruojing
36Kr learned that Deepwise, an AI pharmaceutical unicorn enterprise, has recently completed a new round of financing of $60 million. The new investors in this round of financing include Xinchen Capital, Jinyi Capital, and Kaitai Capital. Existing shareholders CDH Baifu and New Summit Capital continued to increase their investment, and Exponent Capital continued to serve as the exclusive financial advisor. The raised funds will be mainly used for the upgrade of the core technology system.
It is worth mentioning that in December last year, Deepwise just officially announced a Series D financing of $50 million. In just two months, the company has cumulatively raised more than $100 million in financing.
Currently, limited by technological capabilities and other factors, most enterprises still mainly apply AI in the pre - clinical stage of new drug development. In contrast, Deepwise can provide global pharmaceutical companies with a full - stack intelligent solution from pre - clinical drug research, registration and declaration, clinical trials to post - marketing research. According to Deepwise, in past cooperation cases, the clinical trial protocols written by the company's AI "brain" have achieved "zero rework" and passed the approval of regulatory authorities at one time.
In the pharmaceutical industry, in the past, the market was often more willing to pay for pipeline assets with certain value and was more cautious about pure technology platforms. The breakthroughs made by Deepwise in the underlying technical architecture may also be the reason for its successive recognition by investment institutions. Specifically:
First, in the new drug development process, Deepwise no longer focuses on single - function replacement. Instead, it proposes the "theory of single - point ineffectiveness of electronic neurons". Based on a profound understanding of brain science, it cultivates AI into an "autonomous digital scientist team with professional 'consciousness' in the field". Through the decentralized collaboration and self - evolution of the "bionic brain", it creates an autonomous AI intelligent agent cluster that can complete the entire clinical trial process more efficiently on the premise of extremely high precision.
The traditional clinical R & D process includes multiple links such as registration and declaration, trial design, clinical operation, and data management. Deepwise chooses to build this complex clinical medical logic into a "bionic brain" composed of high - precision "Atomic Agents" in units of ten thousand, including the Protocol Agent (protocol design intelligent agent) responsible for top - level design, the Statistical Agent (statistical analysis intelligent agent) performing precise calculations, and the Regulatory Agent (regulatory intelligent agent) ensuring compliance.
When performing actual clinical development tasks, the "bionic brain" reaches the level of top - notch scientists through a profound hypothesis - verification - reflection - correction closed - loop of basic specific micro - tasks (data provided by Deepwise shows that the accuracy rate of systematic medical terminology exceeds 99%); and the agents achieve a dynamic collaborative network connection.
For example, the Planning Agent and the Validation Agent can perform a unique "Planning - Execution two - way verification". Through cross - verification between the design end and the execution end, logical loopholes are eliminated from the source to ensure the scientificity and executability of the protocol.
The ability of the intelligent agents to imitate the human thinking process during task execution, conduct self - reflection, learning, and improvement is also a major highlight of Deepwise's technology platform.
Based on a profound understanding of brain science, the system will automatically trigger relevant "reflection" processes to complete work such as attribution analysis (autonomous judgment of knowledge deficiency or logical fallacy), traceability positioning (precise positioning to specific intelligent agents or decision - making nodes), and self - rewriting (the system can autonomously rewrite, differentiate, and reconstruct agents). In other words, as the frequency of use increases, the problems that the system itself once had will be continuously self - corrected.
In addition, to address the hallucination problem when AI models process serious medical scenarios, Deepwise has also built a four - layer protection mechanism including training, reasoning, cross - verification, and post - processing. Data provided by the company claims that it has "increased the industrial - level accuracy of AI to 99.9% and basically overcome the hallucination problem".
It is worth mentioning that based on the current technology platform, Deepwise has now launched a new Protocol Rehearsal (protocol rehearsal intelligent agent). Its function is to conduct a "virtual clinical trial" before human patients are enrolled, that is, to pre - simulate the possible results in actual clinical trials through AI, so as to accurately predict the enrollment rate, dropout risk, and resource bottlenecks. This is expected to accelerate the implementation and application of AI technology in the human clinical trial stage and promote the transformation of new drug development from a labor - intensive industry in the past to an autonomous and intelligent decision - making - intensive industry.
Investors' Views
Zhao Lei, a partner of Xinchen Capital, said that the investment in Deepwise is based on our firm optimism about the long - term structural trend of AI empowering clinical CRO. As a leading AI - native company in the industry, Deepwise's AI capabilities are iterating very rapidly. Its Multi - Agent collaboration capabilities are profoundly changing the industry's cost structure and operation model, which also constitutes the company's most solid moat. Xinchen Capital will serve as a long - term partner, leveraging its rich medical industry resources and post - investment empowerment experience to help Deepwise build a long - term leading advantage in technological iteration and global market expansion.
Li Hui, the founder and managing partner of Jinyi Capital, said that AI is driving a new round of productivity revolution globally. As a globally leading AI - driven drug development platform, Deepwise provides innovative pharmaceutical companies with a full - stack intelligent solution from pre - clinical drug research, registration and declaration, clinical trials to post - marketing research, realizing the automation and intelligence of the entire clinical process, and helping pharmaceutical companies reduce costs, increase efficiency, and accelerate drug development. The company itself has achieved rapid commercial implementation and a closed - loop of profitability, becoming the core force driving the transformation of pharmaceutical R & D from labor - intensive to intelligent - intensive. It is rapidly subverting the old human - power paradigm of innovative drug development in the AI era and achieving a smart leap in the new paradigm. We are optimistic about the huge development potential of Deepwise in the AI + medical field and look forward to joining hands with Deepwise in the future to jointly create a new global leader in the AI pharmaceutical industry and start a new era of intelligent healthcare.
Zhao Tianjun, the vice - president of Kaitai Capital, said that "AI is the medium of the future". The value of Deepwise lies not only in applying AI to all aspects of clinical R & D but also in fundamentally changing the decision - making logic and execution paradigm of clinical trials through its self - evolving intelligent system. The intelligent agent ecosystem built by the company, which has the ability of continuous learning and reflection, is reshaping the traditional R & D process that relies on experience and manual labor into a new model centered on accurate prediction and intelligent collaboration. We especially appreciate its promotion of the industry's leap from "process optimization" to "process reconstruction", which is not only an improvement in efficiency but also an iteration of the R & D paradigm. Kaitai Capital looks forward to and supports Deepwise as a key innovator in continuously leading the evolution of clinical R & D to a new stage driven by intelligence.