Sanofi

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). Each ID in brackets links to its online source, and the source index at the end of this brief lists them.

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 three AI-discovery partnership commitments involving Sanofi first disclosed in the year to 5 October 2026, with Earendil Labs, Owkin and Aqemia, and three Sanofi Ventures investments in companies with a stated computational or AI discovery platform across the two years [>SAN-E02, >E03, >E08, >E25, >E26, >E27]. Sanofi company sources describe a new R&D function for computational and AI strategy, a biologics AI programme and a digital R&D accelerator launched in 2024; none has an exact start date [>SAN-E11, >E12, >E14]. A partner release of October 2024 reports that one AI-discovery programme was taken back by the partner "following a reprioritization by Sanofi", with no date for the change [>SAN-E10]. We identified no commitment change in company lab digitalisation (T2) in either year in the searched sources, including the T2 event-type and negative searches run in a top-up on 6 October 2026. The robotics and AI tools in Sanofi's Synthetics CMC Development lab are process development, so under the v0.4.3 boundary they are context, not T2 evidence [>SAN-E28, >E29]. The 6 October recruitment snapshot contains 4 relevant R&D requisition postings across the two themes; Sanofi job pages show no openings field, so vacancy status is not recorded [>SAN-R10, >R11, R29, >R30]. These observations identify programmes and functions for further investigation; they do not establish procurement opportunity, deployment completion or net hiring.

Announced commitments

Sanofi published three of these ten items itself [>SAN-E11, >E12, >E14]. The other seven were announced by the partner or investee, with a Sanofi executive quoted in three [>SAN-E03, >E25, >E27]. Sanofi confirmed the Earendil collaboration in its Q4 2025 results presentation and the QuantX investment in its Q1 2026 results presentation [>SAN-E02, >E26]. We found no Sanofi confirmation of the Insilico programme return, the Aqemia expansion, the Owkin K Pro collaboration or the Character Biosciences and Diagonal investments [>SAN-E10, >E08, >E03, >E25, >E27].

The Insilico release gives no date for the reprioritisation and no further detail of it. The event is dated only as falling between the November 2022 agreement and the 30 October 2024 release, so it counts only by first disclosure [>SAN-E10]. It concerns one programme; the release says the collaboration continues on other targets.

BioAIM and CompAI have no stated creation date, so they count only by first disclosure [>SAN-E11, >E12]. The Digital R&D Accelerator was launched in 2024, a year that spans the start of W0, so its event date is uncertain [>SAN-E14]. Its launch date comes from the 2024 Form 20-F and its discovery scope from an undated company page [>SAN-E14, >E15]. The 20-F date shown is the filing date given in a Sanofi press release, not the PDF creation date [>SAN-E14].

The Character Biosciences and Diagonal investments are counted under clarification C42-3 because each release states a computational or AI discovery platform. Both are recorded as borderline: Character describes itself as a precision medicine company, and the Diagonal round funds clinical development [>SAN-E25, >E27].

Sanofi and Earendil Labs announced on 17 April 2025 a licence giving Sanofi exclusive worldwide rights to two bispecific antibodies, HXN-1002 and HXN-1003, which the release says both use Earendil's "proprietary AI and high-throughput discovery & research platform" ($125 million upfront and up to $1.72 billion in milestones) [>SAN-E01]. The release states no AI discovery work, platform access or model access for Sanofi, so under clarification C42-2 the licence is context, not a T1 commitment. Sanofi's 2025 half-year report, of 31 July 2025, confirms the agreement [>SAN-E01]. The January 2026 collaboration does not say it extends the licence, so it is counted under its own parent (C42-1) [>SAN-E02].

By event date, we counted one T1 initiative with a qualifying commitment change in W0 (5 October 2024 to 4 October 2025), Character Biosciences, and five in W1 (5 October 2025 to 5 October 2026): Earendil, Owkin, Aqemia, QuantX and Diagonal. By first disclosure, the T1 count is four in W0, adding Insilico, BioAIM and the Digital R&D Accelerator, and six in W1, adding CompAI. Insilico counts only because of the negative change, and the negative count by event date is zero in both years [>SAN-E10]. T2 is zero in both views and both years. Under the opposite reading of clarification C42-2, which would count in-licensed AI-derived assets in T1, the Earendil licence would count and T1 W0 would be two by event date and five by first disclosure. Clarification C42-4 changes no count as written; if it were read to cover the internal Digital R&D Accelerator, T1 W0 by first disclosure would be three. Three T1 milestones were disclosed in W0 and one in W1 [>SAN-E04, >E05, >E09, >E06]. Existing capabilities first disclosed numbered three in T1 in W0 [>SAN-E16, >E18, >E19], six in T1 in W1 [>SAN-E07, >E13, >E17, >E20, >E22, >E23] and none in T2. Under rule set v0.4.2 the T2 W1 figure was two; the CMC development lab and its named head moved to context in v0.4.3 because they are process development [>SAN-E28, >E29]. The only T2 capability rests on an undated page, so it has no first disclosure window [>SAN-E43]. These have no stated start date and are not counted as changes.

Every theme and window cell is partial. The shared web-search budget ran out during this batch, and a top-up of ten searches on 6 October 2026 completed only some of the missing searches (see Coverage limits). The results-materials and partner-newswire families are also marked not comparable between the years. Coverage is therefore not comparable, and no comparison between years is made.

Identified recruitment

On 6 October 2026 one reviewer counted 851 job URLs in the Sanofi careers sitemap and fetched 364 shortlisted by title keyword. The two reviewers recorded 43 and 39 postings, 23 of them in common, giving 59 requisitions in the table. We observed 4 relevant requisition postings in the R&D theme class (3 T1, 1 T2) [>SAN-R10, >R11, >R30, R29]. No requisition is tagged to both themes. Sanofi job pages show no openings field, so vacancy status is "not recorded" for all 59 and none is classed as ambiguous. A further 18 are internships or graduate placements, 3 are postdoctoral or research training posts, 13 are enterprise roles, 8 are plant, quality or manufacturing roles and 13 are out of scope. These are requisition postings listed on one day, not hires, headcount or a trend.

The T1 postings describe AI and modelling roles in discovery: - a Head of Agentic Biology in the Computational Biology group of CompAI, Cambridge, Massachusetts, to lead "first-generation agentic bioinformatics workflows", advance novel target discovery from multi-modal disease data, lead genetic, transcriptomic and proteomic data lakes "in close collaboration with Digital/AI teams" and "Evaluate and onboard AI/ML vendors" [>SAN-R30] - two Associate Director QSP postings in Morristown, New Jersey, with the same title under separate requisition numbers, in the Disease Modeling group of the Translational Medicine Unit, to develop mechanistic models "that inform drug discovery and development decisions" [>SAN-R10, >R11]

The two QSP postings are recorded as borderline because the group sits in Translational Medicine. Under a stricter reading of clarification C42-21 they would be out of scope, and the R&D theme count would be 2 (1 T1, 1 T2) [>SAN-R10, >R11].

The T2 posting describes a lab automation role: a Principal Automation Software Engineer in the Bioanalytics Automation Lab, Framingham and Waltham, Massachusetts, to build software that automates "laboratory instruments, robotics, sample handling systems, and associated data workflows" and to integrate automation platforms with LIMS, ELN, databases and enterprise systems [SAN-R29].

A Senior Associate Scientist in CMC Bioprocess Analytics, Framingham, is to use automated lab systems and high-throughput assays to inform process development decisions. This is process development, so under the v0.4.3 boundary it is out of scope; it was a T2 posting under v0.4.2 [>SAN-R32]. A second posting with the same Bioprocess Analytics title has analytical testing duties and is also out of scope [>SAN-R33].

Four internship or postdoctoral postings are tagged to the themes but are outside the R&D count. In T1, a machine learning co-op in Quantitative Pharmacology, Cambridge, Massachusetts, is to build AI/ML models "to enhance decision making across drug discovery and development" [>SAN-R18], and a postdoctoral fellow in Vitry-sur-Seine is to combine structural biology with computational modelling to support molecule optimisation [>SAN-R39]. In T2, two working student posts in Frankfurt include digitisation and archiving of raw data [>SAN-R27, >R28]. Internships in the process chemistry department in Montpellier and the process engineering lab in Sisteron cover high-throughput experimentation on automation platforms and the automation of a continuous chemistry installation. These are process development, so under the v0.4.3 boundary they carry no theme [>SAN-R56, >R59]. Enterprise postings include a product owner for AI-ready unstructured data governance [>SAN-R43] and an Enterprise Data Strategy Lead [>SAN-R20]. A causal inference product owner within Digital Clinical Development is an enterprise role under clarification C42-22 [>SAN-R42].

Posting dates come from a datePosted field in the page data that may record a refresh, so they are not reliable.

No LinkedIn company insights were collected for Sanofi.

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 30 Sanofi mentions: 0 explicit use, 3 integration or environment, 1 planned implementation, 1 required experience and 25 alternatives. - Named as existing systems: none. No posting names a product as one Sanofi uses now. - Integration responsibilities: the automation software engineer is to "Integrate laboratory automation platforms with LIMS, ELN, databases, and enterprise systems". These are categories and no product is named [SAN-R29]. The Head of Agentic Biology is to "Design, build, scale and transform existing bioinformatics infrastructure" [>SAN-R30]. - Planned implementation: genetic, transcriptomic and proteomic data lakes, which the Head of Agentic Biology is to lead the development of [>SAN-R30]. - Required experience: LabVIEW, for the Sisteron internship, a process development posting [>SAN-R59]. - Lists of examples: Hamilton, Dynamic Devices, HighRes Biosolutions and Retisoft as lab automation platforms, and LIMS, ELN or SDMS as lab informatics systems [SAN-R29]; Tecan, Hamilton and SoftMax in the CMC Bioprocess Analytics posting [>SAN-R32]; Chemspeed and LabMan as automation platforms in the process chemistry internship [>SAN-R56] (both process development postings, outside the themes from v0.4.3); MATLAB in a list with R and Julia [>SAN-R10, >R11]; and, for the Head of Agentic Biology, AI for Science tools including Owkin K-Pro, Google Co-Scientist, Phylo (Biomni), Mithrl, EMET, PandaOmics (Insilico Medicine), NVIDIA BioNeMo Agent Toolkit and Claude Science (Anthropic), and models such as ESM, AlphaFold 3, Nucleotide Transformer, HyenaDNA and Evo [>SAN-R30].

The Head of Agentic Biology posting lists Owkin K-Pro and PandaOmics only as examples of tools a candidate may have used. It does not say Sanofi uses them, although Owkin and Insilico have each announced agreements with Sanofi [>SAN-R30, >E03, >E09]. No category (ELN, LIMS, cloud) is recorded as explicit use for Sanofi, and no ELN or LIMS product is named in the classified postings.

Deployment evidence

• The 2024 Form 20-F, filed on 13 February 2025, states that "our Target Discovery engines have delivered seven novel drug targets in just one year, while our mRNA modelling solution has cut mRNA design time in half" [>SAN-E16].
• The 2025 Form 20-F, filed on 17 February 2026, lists, under Digital and then R&D, "100% small and large molecule drug design efforts now supported by AI/ML methods, doubling molecular throughput of research teams" and "90% targets in the pipeline now being credentialed using our data and AI systems" [>SAN-E17].
• A Sanofi story of 9 July 2025 says the number of biologics and vaccines using AI in their development at Sanofi "has nearly doubled since 2019" [>SAN-E11]. The same story describes in-house AI models for mRNA design: CodonBERT, mRNA-LM and RiboNN [>SAN-E18].
• An undated Sanofi page describes the Search & Design platform, "a generative interface powered by a proprietary LLM trained on molecular data", which gives chemists access to generative chemistry design [>SAN-E24]. Its first public date is not established.
• A Sanofi story of 28 July 2026 says computational models from Sanofi's Target, Diseases and System Biology team help narrow target lists for its bispecific screening library [>SAN-E22]. A story of 30 June 2026 says Sanofi scientists are using AI to design AAV capsids for delivery to the brain [>SAN-E23].
• Recursion reported that two Sanofi programmes passed initial milestones in 2024, generating $15 million [>SAN-E04], that a fourth milestone generated $7 million for an orally active small-molecule lead in autoimmune diseases [>SAN-E05], and that a fifth in February 2026 generated $4 million [>SAN-E06]. A Sanofi story of 29 June 2026 says the teams "have advanced multiple discovery programs and achieved several development milestones" [>SAN-E07].
• In T2, an undated Genedata webinar page says "Sanofi has integrated Genedata Biologics within its Large Molecules Research Organization to enhance data management and automation", and that the team is building a data governance strategy on FAIR data principles. This is vendor text with no Sanofi quote, and its first public date is not established [>SAN-E43].
• A Q1 2026 results slide lists BioMap's "protein-level artificial intelligence platform" among partners. The slide names no date or terms, and no announcement is in the bundle [>SAN-E20].
• A Sanofi page carrying only the note "Page updated September 2023" says that through Open Targets Sanofi works with academic and industry partners on drug-target identification and prioritisation [>SAN-E21].
• The 20-F figures and the story figures are the company's own statements. We found no independent measure of them, and no other adoption or benefit metric, in the reviewed sources.

Context

Paulo Fontoura was appointed Executive Vice President, Global Head of Research & Development Pharma, effective 1 September 2026, succeeding Houman Ashrafian; the 22 June 2026 release says he was most recently Chief Medical Officer of Xaira Therapeutics, "an AI-native biotechnology company" [>SAN-E36]. From 1 September 2026 Emmanuel Frenehard, Head of Digital, "will continue to report directly to" the CEO, outside the listed Executive Committee, and Finance's expanded remit includes Global Partnering and Business Development [>SAN-E37]. Company sources name Matt Truppo as Global Head of Computational & AI Strategy, R&D [>SAN-E13], Maria Wendt as Global Head of Preclinical Computational Innovation Strategy [>SAN-E19] and Christian Airiau as Global Head of CMC Data Sciences and Global CMC Development [>SAN-E29]. We found no appointment dates for these three. The CMC role is process development, so under the v0.4.3 boundary it is context, not T2 [>SAN-E29].

Sanofi added a $625 million multi-year commitment to Sanofi Ventures on 24 September 2025, taking its assets under management to over $1.4 billion; the release describes the fund as investing in "biotech and artificial intelligence/digital health companies" [>SAN-E31]. Sanofi Ventures investments in QuantHealth (AI clinical trial simulation, October 2025), Trial Library (December 2025) and BranchLab (May 2026) are context under clarification C42-3, because no AI or computational discovery platform is stated [>SAN-E32, >E33, >E34].

Enterprise and clinical AI: the 2024 Form 20-F says Sanofi "deployed the internal application plai" with Aily Labs [>SAN-E38]. A June 2026 story describes AI Foundry as "our internal platform powering AI development" [>SAN-E40]. The 2025 Form 20-F describes products from the Formation Bio and OpenAI collaboration of May 2024 for patient profiles, outreach and IRB-ready content [>SAN-E39]. Biogen joined accelRare, a rare-disease pre-diagnosis tool for physicians, in June 2026 [>SAN-E41]. On 23 October 2025 Medidata and Sanofi announced an expanded enterprise agreement under which Sanofi will use Medidata's Patient, Data and Study Experiences and its decentralised trial capabilities [>SAN-E44]. These are clinical or enterprise uses, so they are context.

Sites and manufacturing: on 15 October 2025 Sanofi launched a Translational Centre of Excellence in Montpellier with IHU Immun4Cure under a five-year partnership; no AI discovery work is stated [>SAN-E35]. An undated Sanofi page says the M&S Accelerator in Lyon "officially opened in May 2025" [>SAN-E42]. A Sanofi story of 13 January 2026 describes a CMC development lab in Cambridge, Massachusetts, where Solutron, a robot "designed by Sanofi scientists", runs experiments chosen with the Solvify solubility prediction tool, and a Kinetic AI platform analyses reaction results; the story uses the present tense [>SAN-E28]. This lab is process development, not a research lab, so under the v0.4.3 boundary it is context; it was T2 evidence under v0.4.2. The same story describes process digital twins that simulate production across manufacturing steps; this is manufacturing, so it is context [>SAN-E30].

Apart from the Insilico programme [>SAN-E10], we identified no qualifying pause or cancellation in the two themes in the searched sources. The dedicated T2 negative search, and a further T1 negative search, were run in the 6 October top-up and found none. That is a search result, not evidence that none occurred.

Coverage limits

• The shared web-search budget ran out during this batch. A top-up of ten searches on 6 October 2026 ran the T2 leadership, partnership, acquisition, consortium, regulatory-programme and open-release event-type searches, the T2 negative search, a T2 academic search and a T1 negative search. The T2 event-type and negative cells are now complete, except open release, which stays partial because the GitHub organisation could not be checked. In both themes and both years, the partner-newswire, consortium and academic-collaboration families are still partial. In T1 the capacity and regulatory-programme event-type searches did not run. Ortent keeps a record of the exact queries.
• The reviewers did not find results transcripts on the results pages for Q3 2024 to Q1 2025 or for Q3 and Q4 2025. The results-materials and partner-newswire families are marked not comparable between the years.
• Business Wire and GlobeNewswire blocked access. The Owkin, Aqemia and Recursion releases were read on owkin.com and through BioSpace and Stock Titan copies. Fierce Biotech and American Pharmaceutical Review also blocked access, and the Earendil releases were read through PR Newswire Asia copies.
• ir.recursion.com blocked access, and the Recursion Q1 and Q2 2026 results releases were not obtained, so any Recursion milestone after February 2026 is unchecked [>SAN-E06].
• SEC EDGAR blocked access and was not used; the Forms 20-F were read from sanofi.com. The proxy refused GitHub, again in the top-up, so the Sanofi-Public organisation was not checked for open releases.
• The Sanofi Digital & AI page is undated, so the first public dates of the rows that rest on it are not established [>SAN-E15, >E24, >E42]. The Open Targets page shows only an update note [>SAN-E21].
• The first public dates for BioAIM, CompAI and the Digital R&D Accelerator are the earliest found. Earlier disclosures may exist in sources the search did not reach [>SAN-E11, >E12, >E14].
• Patents are not among the searched signal types. Publication counts are raw Europe PMC query counts and are not events.
• Recruitment: one reviewer fetched 364 job pages shortlisted by title from 851 sitemap URLs. Only the 59 postings classified are in the table.
• Two independent reviewers searched Sanofi, and the final rows are their source-verified union. The Aqemia expansion and the Insilico programme return were found by one reviewer only, and Character Biosciences, Diagonal and CompAI by the other [>SAN-E08, >E10, >E25, >E27, >E12]. The union does not show that both reviewers did not miss the same events.

Two questions to investigate (Ortent's interpretation)

1. Sanofi describes CompAI as a new R&D function, and its Head of Agentic Biology is to "Evaluate and onboard AI/ML vendors" and build data lakes with Digital/AI teams [>SAN-E12, >R30]. The Owkin K Pro collaboration was announced with a quote from Sanofi's Chief Digital Officer, and its agents are to be deployed through K Pro [>SAN-E03]. From 1 September 2026 the Head of Digital reports to the CEO, and Finance's remit includes Global Partnering and Business Development [>SAN-E37]. Who holds decision rights over licensing an external AI discovery tool or agent rather than building one in house: CompAI, the Digital function or the research units? Does such a licence pass through Global Partnering and Business Development, a Digital procurement route or another route?
2. Lab automation and lab data roles sit in several units. In research, a Bioanalytics Automation Lab engineer is to integrate automation platforms with LIMS and ELN [SAN-R29], and a vendor page says the Large Molecules Research Organization has integrated a biologics data management and automation platform [>SAN-E43]. On the process development side, which is outside T2 under the v0.4.3 boundary, a CMC Bioprocess Analytics scientist is to help implement new automation and analytics capabilities [>SAN-R32], a CMC development lab runs an in-house robot with prediction tools [>SAN-E28], and Sanofi names a Global Head of CMC Data Sciences [>SAN-E29]. The Head of Agentic Biology is to close "the loop between automated wet-lab data capture and cutting-edge predictive modeling" [>SAN-R30]. What is the remit of each unit for lab automation and lab data systems, and who decides on new automation and informatics systems? How are the interfaces between these lab units, CompAI, the R&D Digital Accelerator and the Digital function governed [>SAN-E14, >E15]?

Sources for the main claims

• SAN-E10 Insilico Medicine milestone and returned programme (EurekAlert): https://www.eurekalert.org/news-releases/1063134
• SAN-E14 Sanofi 2024 Form 20-F: https://www.sanofi.com/assets/dotcom/content-app/publications/annual-report-on-form-20-f/2024-01-01-form-20-f-2024-en.pdf
• SAN-E15 Sanofi Digital & AI page (undated): https://www.sanofi.com/en/our-science/digital-artificial-intelligence
• SAN-E11 Sanofi story, AI Across the R&D Value Chain: Drug Discovery: https://www.sanofi.com/en/magazine/ai-in-healthcare/ai-across-the-randd-value-chain-drug-discovery
• SAN-E25, E26, E27 Sanofi Ventures news page (Character Biosciences, QuantX, Diagonal): https://www.sanofiventures.com/news/index.html
• SAN-E02 Earendil Labs collaboration (PR Newswire Asia copy): https://en.prnasia.com/releases/global/earendil-labs-announces-strategic-collaboration-with-sanofi-to-discover-bispecific-antibodies-for-autoimmune-diseases-518612.shtml
• SAN-E03 Owkin K Pro collaboration: https://www.owkin.com/newsfeed/owkin-to-build-ai-agents-as-part-of-a-multi-year-k-pro-collaboration-with-sanofi
• SAN-E12 Head of Agentic Biology posting (CompAI): https://jobs.sanofi.com/en/job/cambridge/head-of-agentic-biology/2649/44654451136
• SAN-E13 Sanofi story, Partner Spotlight with Recursion: https://www.sanofi.com/en/magazine/ai-in-healthcare/partner-spotlight-how-sanofi-and-recursion-are-reimagining-drug-discovery-with-ai
• SAN-E08 Aqemia expansion (BioSpace copy): https://www.biospace.com/press-releases/aqemia-and-sanofi-expand-their-research-collaboration
• SAN-E43 Genedata webinar page (undated): https://www.genedata.com/resources/learn/details/webinar/cpepp-in-action-high-throughput-protein-expression-and-purification-at-sanofi
• SAN-E01 Earendil Labs licence (PR Newswire Asia copy): https://en.prnasia.com/releases/global/earendil-labs-announces-worldwide-exclusive-license-agreement-with-sanofi-for-next-generation-bispecific-antibodies-for-autoimmune-and-inflammatory-bowel-diseases-485805.shtml
• SAN-E17 Sanofi 2025 Form 20-F: https://www.sanofi.com/assets/dotcom/content-app/publications/annual-report-on-form-20-f/2025-01-01-form-20-f-2025-en.pdf
• Other event rows and all requisitions: URLs and archived copies in the Sanofi tables. No SRC source files are cited for Sanofi.

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.