Daiichi Sankyo

Ortent Life Sciences Radar, research beta. Evidence to 6 October 2026, rule set v0.4.3. Scope: AI and computation in drug discovery (T1), and research-lab digitalisation and scientific data systems (T2), across the whole Daiichi Sankyo group. Each ID in brackets links to its online source, and the source index at the end of this brief lists them. Several sources are in Japanese. They are quoted in Japanese, with an English rendering in square brackets marked as Ortent's translation; these renderings are Ortent's own.

Disclosure: Andrew Wyatt, Ortent's principal, was Chief Growth Officer at Sapio Sciences until April 2026. Software vendors, including Sapio and its competitors, appear in this brief only where public sources name them.

Status: research beta. Ortent has reviewed this brief; it has not been independently validated. It summarises activity identified in public sources. Search coverage varies by company and is set out under Coverage limits, and zero identified items does not mean no activity. It does not predict buying intent or measure operational maturity. For this radar, process and CMC development are grouped with manufacturing and excluded from the research-lab theme (T2); such work appears as context where it helps.

Summary

The reviewed sources identify two T1 initiatives with an AI-discovery commitment change involving Daiichi Sankyo first disclosed in the year to 4 October 2025, and one in the year to 5 October 2026 [>DSK-E01, >E12, >E14]. They are a Smart Research Laboratory in San Diego that the company says will support AI-based approaches in drug discovery, a research collaboration on Nosis Biosciences' AI drug design platform, and an expansion of a Turbine collaboration to Daiichi Sankyo's own ADC discovery programmes [>DSK-E01, >E12, >E14]. In company lab digitalisation (T2), the San Diego laboratory counts in the year to 4 October 2025, and two new Research Division units, Smart Research Laboratories I and II, count in the year to 5 October 2026; a new Shinagawa research building labelled Smart Lab counts only by first disclosure [>DSK-E01, >E10, >E11]. The 6 October recruitment snapshot contains 4 relevant R&D requisition postings across the two themes, none with ambiguous vacancy status [>DSK-R01 to >R04]. These observations identify programmes and functions for further investigation; they do not establish procurement opportunity, deployment completion or net hiring.

Announced commitments

Daiichi Sankyo published three of these five items itself [>DSK-E01, >E10, >E11]. The other two were announced by the partner with no Daiichi Sankyo quote, and we found no Daiichi Sankyo release or other confirmation of either [>DSK-E12, >E14]. Both were found by one reviewer, through the partners' own newsrooms.

The San Diego laboratory counts in both themes: T2 for lab automation and data orchestration, and T1 because the release separately states the AI connection [>DSK-E01]. The Nosis deal counts in T1 because the release states AI design work under the collaboration, so it is not an in-licensed asset under C42-2 [>DSK-E12]. The Turbine row is an expansion, because the release adds Daiichi Sankyo's proprietary ADC programmes to the earlier work (C42-1) [>DSK-E14]. The earlier feasibility programme has no stated start date, so it is recorded as a disclosure and not counted [>DSK-E13].

Creating Smart Research Laboratories I and II is a leadership event under rule D4. They count in T2 only. The same notice says the R&D division will "データ駆動型の創薬探索を推進する" [Ortent's translation: promote data-driven drug discovery], but it states no AI or computation, which T1 needs under AR7a [>DSK-E10]. The Shinagawa building is placed in an FY2025 to FY2030 range on the slide, so its event date is uncertain and it counts only by first disclosure (C42-14). No amount, date or design is stated [>DSK-E11].

By event date, we counted two T1 initiatives with a qualifying commitment change in W0 (5 October 2024 to 4 October 2025), the San Diego laboratory and Nosis, and one in W1 (5 October 2025 to 5 October 2026), the Turbine expansion. By first disclosure, the T1 count is also two in W0 and one in W1. T2 is one in W0 in both views: the San Diego laboratory, which is tagged to both themes and so also appears in T1 [>DSK-E01]. T2 in W1 is one by event date, Smart Research Laboratories I and II, and two by first disclosure, adding the Shinagawa building [>DSK-E10, >E11]. No milestones were disclosed in either window. Existing capabilities first disclosed numbered two in T1 and one in T2 in W0 [>DSK-E02, >E08], and six in T1 and four in T2 in W1 [>DSK-E03, >E04, >E05, >E09, >E13, >E15]. These have no stated start date and are not counted as changes. Under the opposite reading of borderline clarification C42-2, treating the Nosis option as an in-licensed AI-derived asset, T1 W0 would be one in both views [>DSK-E12]; under the opposite reading of C42-4, treating the BioMed X call's encouragement of machine learning tools as AI discovery scope, T1 W0 would be three in both views [>DSK-E20].

Coverage is partial for both themes in both years. Seven of the nine source families are complete. The results-material family is partial because Daiichi Sankyo publishes audio and video webcasts, not text transcripts, and the webcasts were not reviewed. The partner newswire family is partial because Business Wire and GlobeNewswire were blocked or not fetched and the newswire search was not run for lack of search budget. The table marks the annual report family as not comparable between the years, because the Value Report 2026 was not published by the retrieval date, so no comparison between the years is made.

Eighteen items are recorded as context and one as excluded (see Context).

Identified recruitment

On 6 October 2026 the careers.daiichisankyo.com sitemap listed 258 job URLs; one reviewer counted 205 distinct requisition numbers among them. The standard English title filter matched no listing, because most Daiichi Sankyo R&D postings carry Japanese titles. One reviewer applied Japanese equivalents of the filter terms, which matched 6 requisitions, and the other read every title. The two reviewers classified 20 requisitions from their duties. We observed 4 relevant requisition postings in the R&D theme class (1 T1, 3 T2), none with ambiguous vacancy status [>DSK-R01 to >R04]. No posting is tagged to both themes. A further 5 are enterprise roles, 1 is a plant quality control role and 10 are out of scope. These are requisition postings listed on one day, not hires, headcount or a trend.

All four R&D theme postings show a "募集人数" [Ortent's translation: number of openings] field reading "若干名" [Ortent's translation: a few], so each is recorded as listed with openings [>DSK-R01 to >R04].

The T2 postings describe lab automation and lab software roles in the new units, all in Shinagawa, Tokyo: - an automation engineer in Smart Research Laboratory I, whose duties begin with "創薬研究業務の自動化システムの構築や最適化の検討" [Ortent's translation: building and optimising automation systems for drug discovery research] and include tuning and maintaining automated research instruments [>DSK-R01] - a software engineer in Smart Research Laboratory I, for "ラボのスマート化を支える社内利用アプリケーションの設計・開発・運用" [Ortent's translation: the design, development and operation of in-house applications that support lab smartification] in drug discovery research [>DSK-R02] - a high-throughput synthesis role in Smart Research Laboratory II, to build and run "自動合成システム（合成ロボット、自動分注等）" [Ortent's translation: automated synthesis systems (synthesis robots, automated dispensing and similar)], move manual synthesis to automation and conduct "国内外ベンダーとの技術折衝" [Ortent's translation: technical negotiations with domestic and overseas vendors] [>DSK-R03]

The automation engineer posting says the company is pursuing "創薬研究領域のデジタル変革として実験の自動化とデータ収集のリアルタイム化" [Ortent's translation: lab automation and real-time data collection as the digital transformation of drug discovery research] [>DSK-R01].

The T1 posting is a senior scientist in Modality Research Laboratory II, for drug discovery through "計算科学・機械学習などを用いた各種バイオモダリティデザイン" [Ortent's translation: biomodality design using computational science, machine learning and similar methods] and for "社内モダリティデザイン基盤整備" [Ortent's translation: building an internal modality design platform] [>DSK-R04].

Enterprise postings include a Cloud Center of Excellence engineer in the Digital & Technology department. The posting says a "セルフサービス型クラウド基盤" [Ortent's translation: self-service cloud platform] supports the transformation of drug discovery research and is being extended by "全社展開" [Ortent's translation: company-wide rollout] [>DSK-R05]. Its duties name no discovery activity, so it is an enterprise role under C42-21. The other enterprise postings are an IT and DX audit role, two GxP system roles and a Munich digital transformation lead for corporate platforms [>DSK-R06 to >R09]. The out-of-scope postings are three pharmacovigilance automation roles, two real-world evidence programming roles under C42-22, four biologics research roles with no AI or lab-systems duties under C42-20, and an R&D budget system role in finance [>DSK-R16 to >R18, >R19, >R20, >R11 to >R14, >R15].

Japanese postings show a "掲載開始日" [Ortent's translation: posting start date] field, which may record a refresh, so posting dates are not reliable.

No LinkedIn company insights were collected for Daiichi Sankyo.

Technology mentions in recruitment

Mentions are classified from the exact posting wording. They do not establish installation, an incumbent contract or company-wide use. The table records 15 Daiichi Sankyo mentions: 2 explicit use, 4 required experience, 2 preferred experience and 7 alternatives. - Named as existing systems: AWS, which the Cloud CoE posting names as the cloud platform the team is to design, build and run: "クラウド基盤 （AWS）の設計・構築・運用" [Ortent's translation: design, build and operation of the cloud platform (AWS)]. This is an enterprise role and role-specific wording [>DSK-R05]. Outside R&D, a plant quality control lab posting in Pfaffenhofen names the analytical data software the role uses; the product is recorded in the technology mentions table and is outside the themes [>DSK-R10]. - Required experience: understanding of research instruments, robots and their control systems [>DSK-R01]; public cloud and UNIX/Linux [>DSK-R02]; and building a data management platform [>DSK-R04]. - Preferred experience: SiLA 2, a lab automation standard [>DSK-R01], and cloud-native architecture on AWS [>DSK-R02]. The software engineer posting does not state that the employer uses AWS. A plan slide separately lists "Building AWS infrastructure and data utilization systems" under research DX [>DSK-E09]. - Lists of examples: Chemspeed and Unchained Labs automated synthesis equipment, and Python and Pipeline Pilot software [>DSK-R03]; computational science and machine learning for biomodality design [>DSK-R04]; Terraform and Docker [>DSK-R05].

No integration responsibility, planned implementation or unclear mention is recorded. No category (ELN, LIMS, cloud) is recorded as explicit use in the R&D postings, and no ELN or LIMS product is named.

Deployment evidence

• San Diego laboratory. A Lab Design News interview of 27 February 2025 quotes the site head: "We leased a 10,000 sf (1000 sm) facility", with robots and automated research equipment then being installed [>DSK-E02]. This is a media source quoting company staff. A company story of 9 December 2025 says the team is "laying the foundation for integrating generative AI into our workflows" and that its first milestone is "getting the lab fully operational" [>DSK-E04]. Science and Technology Day slides of 16 December 2025 state "Smart Lab is the cornerstone of AI-driven drug design success." [>DSK-E05]. An undated R&D page says the facility "is set to begin full operations in 2026" [>DSK-E06]. We found no source stating that the laboratory is fully operational.
• Lab smartification. A Japanese company story of 30 October 2025 names "ラボスマーティフィケーション" [Ortent's translation: lab smartification] as one of the important R&D measures in the company's DX vision, and names a head of the initiative with no appointment date [>DSK-E03]. An undated Japanese DX page says that in drug discovery research the company is working to realise a smart research lab using "ロボット、データ基盤、AIなどを活用するスマートリサーチラボ" [Ortent's translation: a smart research lab that uses robots, a data platform, AI and similar], with robot-run experiments, metadata stored with results as standardised data products, a mechanism for compiling results into an electronic notebook, and image AI built into lab apparatus [>DSK-E17]. The page names no product.
• Discovery AI tools. The same undated page says Daiichi Sankyo has introduced "OpenFold-3をはじめとする最先端のAIモデルを社内環境に導入し" [Ortent's translation: cutting-edge AI models, starting with OpenFold-3, into its internal environment], provided to researchers as cloud applications, and has introduced "AIエージェントをバイオインフォマティクス業務に導入し" [Ortent's translation: AI agents into bioinformatics work] [>DSK-E16]. Because the page is undated, these two disclosures fall in neither window (C42-13) [>DSK-E16, >E17].
• Programme statements. In the Value Report 2025, an outside director says introduction of "DX and generative AI is progressing, particularly in drug discovery" and clinical development [>DSK-E08]. The May 2026 business plan slide places an "AI-driven molecular design research platform" and "Building AWS infrastructure and data utilization systems" on an FY2020 to FY2030 research DX timeline, with "Expanding and strategizing AI-driven drug discovery into multi-modality research" as a forward item with no stated resource [>DSK-E09]. The same plan states an aim to accelerate targeted protein degradation molecule creation by integrating medicinal chemistry with AI, with no unit, resource or date [>DSK-E15].
• Partner work. Turbine's release says an initial feasibility programme evaluating its predictive modelling for ADC development was completed before the September 2026 expansion; its dates are not stated [>DSK-E13]. ExaWizards and Daiichi Sankyo jointly announced on 18 March 2024, before both windows, that their joint AI drug discovery project "discovered a hit compound on a challenging disease-causing target protein"; the collaboration dates from 2019, and we found no W0 or W1 action [>DSK-E18].
• Annual-report dates rest on the investor news feed posting dates, and the cited Value Report 2025 file is a later revision [>DSK-E08].
• We found no adoption or benefit metric for any in-window item in the reviewed sources.

Context

Leadership and enterprise programmes: a company story of 29 October 2024 describes James Felix as "Recently appointed Head of Global Digital Transformation", leading the first global IT strategy and internal generative AI platforms; no discovery remit is stated, so this is governance context under C42-7 [>DSK-E23]. From 1 April 2026 Global DX split its Data Intelligence Department to create a "データファウンデーション& AI-ML 部" [Ortent's translation: Data Foundation & AI-ML Department]; no discovery remit is stated [>DSK-E22]. From 1 July 2026 biostatistics and clinical data management moved from Global DX to a new R&D department, the "生物統計＆データマネジメント部" [Ortent's translation: Biostatistics & Data Management Department] [>DSK-E28]. The May 2026 plan release says a Business Transformation function was established to drive company-wide operational excellence by enhancing "productivity through AI and digital technologies" [>DSK-E26]. The FY2024 securities report, filed 20 June 2025, describes AI governance work including "グローバルAIガバナンスポリシーの策定" [Ortent's translation: formulating a global AI governance policy] [>DSK-E24]. The company was selected as a Japanese government DX Brand for the third consecutive year in April 2025 [>DSK-E25].

Daiichi Sankyo established a Research Institute in San Diego on 2 October 2025, its third research institute outside Japan; the release states no AI or lab-systems scope and does not link it to the Smart Research Laboratory [>DSK-E19].

Under C42-8, six AI items for biomarkers, patient selection or clinical development are context, not T1: Tempus models for patient selection in an ADC programme (25 March 2026), BostonGene translational analytics in an ADC development programme (17 February 2026), Lunit digital pathology for biomarker discovery in two oncology programmes (16 December 2025, media report), computational pathology for the TROP2 biomarker, digital pathology with Merck & Co, and the plan's use of AI tools in clinical development [>DSK-E32, >E31, >E33, >E29, >E30, >E27].

Other context: BioMed X announced a funded research call on multispecific biologics with Daiichi Sankyo in January 2025 and the start of the research team in October 2025; the call encourages machine learning tools but does not state AI discovery scope, so it is context under C42-4 [>DSK-E20, >E21]. Daiichi Sankyo Group appears on the member list of the Life Intelligence Consortium (LINC), with no join date and no stated AI scope for the membership [>DSK-E34]. The open innovation page seeks partners for the "de novo design of antibody-like binders" using computational approaches; seeking partners is not a commitment [>DSK-E35]. A General Proximity collaboration with the Boston research institute is excluded, because no AI or computational scope is stated [>DSK-E37].

We identified no qualifying pause or cancellation in the two themes in the searched sources. The February 2026 notice abolishes the Hit Discovery Platform Lab and Modality Lab III, but moves their functions into the new units, so it is not a negative change under N1 [>DSK-E10]. Both reviewers ran a negative-event web search for each theme. That is a search result, not evidence that none occurred.

Coverage limits

• Seven source families are complete in every cell: the company press release archive, annual reports and filings, consortium, academic collaboration, corporate venture (the company states it has no corporate venture capital function), one web search per event type, and one negative-event search per theme. Two are partial in every cell: results presentations and transcripts, and partner newswires.
• Results presentations were read for every quarter in both windows, but Daiichi Sankyo publishes audio and video webcasts, not text transcripts, and neither reviewer reviewed the webcasts. This family is partial in every cell.
• Business Wire blocked access and GlobeNewswire failed to connect; neither was used. The Nosis and Turbine releases were read on the partners' newsrooms, and the Tempus and BostonGene releases through 1stOncology reproductions [>DSK-E12, >E14, >E32, >E31]. tempus.com, investors.tempus.com, nasdaq.com, pharmaceutical-technology.com, genomeweb.com and mt.com also blocked access, HPCwire disallows all agents in robots.txt, and the pharmafocusasia.com and indianpharmapost.com result pages sit under paths their robots.txt disallows. The daiichisankyo.com R&D pages served a disclaimer page that the reviewers accepted.
• No Japanese-language web search was run. Japanese company newsroom feeds, organisation notices, securities reports, stories and DX pages were read directly, but Japanese-only announcements outside the company newsroom, such as partner releases and trade press, may be under-covered.
• Searches not run for lack of search budget: the newswire search "Daiichi Sankyo" businesswire AI OR "machine learning" OR "lab automation" 2025 2026; the Japanese-language partnership searches 第一三共 AI創薬 提携 2025 2026 and 第一三共 AI創薬 共同研究 2025; a search for an earlier announcement of the Turbine feasibility programme and a Business Wire check of the Turbine expansion; and a search for a MeDDeNovo Drug Design collaboration. Searches to date the Shinagawa building decision and to settle OpenFold consortium membership were also not run.
• Whether the in-house OpenFold-3 use comes through consortium membership is not established. The OpenFold new-members release on Business Wire was blocked and HPCwire was not fetched [>DSK-E16].
• The Value Report 2026 was not published by the retrieval date. W1 annual-report coverage rests on the FY2025 securities report and the Sustainability Report 2026, so that family is not comparable between the years.
• Five rows rest on undated pages, so their disclosure window is uncertain and they are not counted in either window (C42-13) [>DSK-E06, >E16, >E17, >E34, >E35]. The logo-grid row below is also undated.
• On the open innovation page, partner names appear only in logo images, so that row is not verified and is not used as evidence [>DSK-E36].
• Patents are not among the searched source families. Publication counts are raw Europe PMC query counts and are not events; the two reviewers used different term sets.
• Recruitment: both reviewers enumerated the 258 job URLs in the careers sitemap. Only the 20 requisitions classified are in the table. Four requisitions on English or German pages show no openings field, so their vacancy status is not recorded [>DSK-R09, >R10, >R19, >R20].
• Two independent reviewers searched Daiichi Sankyo, and the final rows are their source-verified union. That union contains rows that each reviewer missed, but it does not show that both did not miss the same events. The Nosis and Turbine rows were found by one reviewer only [>DSK-E12, >E14].

Two questions to investigate (Ortent's interpretation)

1. Lab automation and lab software roles sit in Smart Research Laboratories I and II in Shinagawa, created on 1 April 2026 [>DSK-E10, >R01 to >R03]. Company sources say the San Diego Smart Research Laboratory is set to begin full operations in 2026, and one of its leaders says most of its staff are contractors in several countries [>DSK-E04, >E06]. The automated synthesis role is to conduct technical negotiations with equipment and software vendors [>DSK-R03]. A Cloud CoE in the Digital & Technology department runs a self-service cloud platform that supports discovery research [>DSK-R05], and Global DX now has a Data Foundation & AI-ML Department [>DSK-E22]. What is the remit of each unit for lab automation, orchestration software and research data systems, and who decides on automation equipment and software vendors? How are the interfaces between the research units, the San Diego laboratory, the Cloud CoE and Global DX governed, including for the planned Shinagawa building [>DSK-E11]?
2. The external AI-discovery work found in this review was announced by partners: Nosis and the Turbine expansion, neither with a Daiichi Sankyo quote or confirmation, and the joint ExaWizards project before the windows [>DSK-E12, >E14, >E18]. Internally, an undated company page says Daiichi Sankyo provides OpenFold-3 and other AI models to researchers on the cloud [>DSK-E16], a biomodality design role is to build an internal modality design platform [>DSK-R04], and the open innovation page seeks partners for computational antibody-like binder design [>DSK-E35]. Who holds decision rights over adopting an external AI discovery platform rather than building in-house, and do such collaborations pass through open innovation, the research units or DX procurement? How is funding for them set within the 2026 to 2030 business plan, which lists an AI-driven molecular design research platform [>DSK-E09]?

Sources for the main claims

• DSK-E01 Daiichi Sankyo, Smart Research Laboratory in San Diego: https://www.daiichisankyo.com/files/news/pressrelease/pdf/202501/20250121_E.pdf
• DSK-E12 Nosis Biosciences, research collaboration and option agreement: https://www.nosisbio.com/news/nosis-biosciences-announces-research-collaboration-and-option-agreement-with-daiichi-sankyo-to-develop-cell-targeted-rna-medicines
• DSK-E10 Daiichi Sankyo organisation notice, 24 February 2026 (Japanese): https://www.daiichisankyo.co.jp/files/news/pressrelease/pdf/202602/20260224_J2.pdf
• DSK-E11 FY2025 Financial Results and 5-Year Business Plan presentation: https://www.daiichisankyo.com/files/investors/library/quarterly_result/2025/FY2025%20Financial%20Results%20and%205Year%20Business%20Plan%20Presentation_E_v2.pdf
• DSK-E13, E14 Turbine, expanded ADC discovery collaboration: https://turbine.ai/news/turbine-and-daiichi-sankyo-expand-adc-discovery-collaboration/
• DSK-E04 Daiichi Sankyo story on the Smart Research Laboratory: https://www.daiichisankyo.com/our_stories/detail/index_4658.html
• DSK-E16, E17 Daiichi Sankyo DX page (Japanese, undated): https://www.daiichisankyo.co.jp/about_us/dx/
• DSK-E18 ExaWizards and Daiichi Sankyo joint release: https://exawizards.com/en/archives/26987
• Other event rows and all requisitions: URLs and archived copies in the Daiichi Sankyo tables. No SRC source files are cited for Daiichi Sankyo.

Source index

Each ID in brackets links to the online source for that row. Job posting links come from a snapshot of the company job site on 6 October 2026; postings that have closed since are marked and no longer linked. Where a company site blocked access, the link goes to the copy that was read.