Johnson & Johnson

Ortent Life Sciences Radar, research beta. Evidence to 6 October 2026, rule set v0.4.3. Scope: Innovative Medicine R&D, covering AI and computation in drug discovery (T1), and research-lab digitalisation and scientific data systems (T2). MedTech items are excluded, because no reviewed source tied them to drug discovery; the NVIDIA robotic-surgery collaboration is recorded as excluded for this reason [JNJ-E15]. 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 four AI-discovery initiatives with a qualifying commitment change involving Johnson & Johnson in the year to 5 October 2026, and two in the year before [>JNJ-E01, >E02, >E05, >E07, >E09, >E11]. Two of the four in the latest year are J&J Innovation QuickFire challenges, counted under a borderline clarification [>JNJ-E07, >E09]. J&J's chief information officer said in April 2026, as reported by Reuters, that AI had halved the company's lead optimisation time [>JNJ-E12]. We identified no commitment change in company lab digitalisation (T2) in either year in the searched sources. The 6 October recruitment snapshot contains 26 relevant R&D requisition postings across the two themes, 0 with ambiguous vacancy status (IDs under Identified recruitment). Two further lab data and AI postings, in Cell and Gene Therapy Analytical Development and in Therapeutics Development & Supply, are CMC and process development work, so under the v0.4.3 boundary they are outside T2 [>JNJ-R057, >R196]. These observations identify programmes and functions for further investigation; they do not establish procurement opportunity, deployment completion or net hiring.

Announced commitments

J&J published the two QuickFire launches itself through J&J Innovation [>JNJ-E07, >E09]. The AISB Bind release was issued by the AISB Network and quotes J&J's Global Head of In Silico Discovery [>JNJ-E05]. The other three were announced by the partner, with no J&J quote [>JNJ-E01, >E02, >E11]. J&J Innovation reposted the XtalPi release on 23 October 2024 and the Isomorphic Labs release on the day it was issued; these reposts count as company confirmation, not company authorship [>JNJ-E01, >E11]. We found no J&J confirmation of either AISB initiative [>JNJ-E02, >E05].

The two QuickFire rows rest on clarification C42-4, which is marked borderline. Under it, a company-run challenge counts as a T1 commitment where the source states AI or computational discovery scope and company funding or collaboration. The Data & AI challenge states a funding pool [>JNJ-E09]. The CAT-PD challenge states no prize money and counts on the stated possibility of collaboration for finalists, so it is the weakest counted row [>JNJ-E07].

Later steps under these initiatives are recorded as milestones. A J&J Innovation blog of 27 January 2026 says the CAT-PD challenge "was run in partnership with NVIDIA" and that finalists pitched at the JPM conference [>JNJ-E08]. On 20 July 2026 J&J named four awardees of the Data & AI challenge, who will receive grants from the pool already announced [>JNJ-E10]. Apheris announced on 1 October 2025 that Astex, Bristol Myers Squibb and Takeda were joining the OpenFold3 initiative; nothing new was stated for J&J [>JNJ-E03]. A consortium results report followed on 14 September 2026 [>JNJ-E04]. BigHat Biosciences announced on 28 October 2025 the completion of its machine-learning antibody design collaboration with J&J. The source does not say whether the end was scheduled, so it is recorded as a milestone, not a negative change [>JNJ-E06].

By event date, we counted two T1 initiatives with a qualifying commitment change in W0 (5 October 2024 to 4 October 2025), the Ailux licence and the OpenFold3 initiative, and four in W1 (5 October 2025 to 5 October 2026), AISB Bind, the two QuickFire challenges and Isomorphic Labs. By first disclosure, the T1 count is also two in W0 and four in W1. No counted initiative has an uncertain event date. T2 is zero in both views and both years. One T1 milestone was disclosed in W0 [>JNJ-E03] and four in W1 [>JNJ-E04, >E06, >E08, >E10]. Existing capabilities first disclosed numbered none in W0 and three in T1 in W1 [>JNJ-E12, >E13, >E14]. These have no stated start date and are not counted as changes. Under the opposite reading of clarification C42-4, which would treat company challenges as context, the T1 count in W1 would be two on both views instead of four; if only the CAT-PD challenge were dropped, it would be three.

The counts table marks all four theme and year cells as partial coverage. The coverage table shows comparable coverage between the years only for the venture-arm, one-search-per-event-type, consortium, academic and negative-event search families; the press-release, annual-report, results and newswire families are not comparable. No comparison between the years is made.

Identified recruitment

On 6 October 2026 two reviewers screened J&J requisitions on its Workday careers site, one 145 and the other 245, with 90 in common. We classified the union of 300 requisition numbers from their duties; 298 were listed that day and 2 were named only inside other postings. We observed 26 relevant requisition postings in the R&D theme class, 0 with ambiguous vacancy status; each is cited in the two lists below. The T1 split is 10 and the T2 split is 18; the two roles tagged both themes appear in each split [>JNJ-R086, >R118]. A further 7 listed postings are research training posts (6 tagged T1), 51 are internships or graduate placements (3 tagged to the themes), 54 are enterprise roles, 28 are plant, quality or manufacturing roles and 132 are out of scope. These are requisition postings listed on one day, not hires, headcount or a trend.

Workday shows no openings field, so vacancy status is not recorded except where the posting text states a count. Two R&D postings each state two openings [>JNJ-R190, >R193]. One more T2 requisition number is named only inside another posting and is not counted [JNJ-R085].

The T1 postings, and the two tagged both, describe AI and modelling roles in discovery: - in Biologics Discovery, a Principal Agentic Workflow & AI Engineer to "help build the agentic layer of our Biologics Discovery data engine", connecting experimental outputs to models across the design-make-test-learn cycle (tagged both) [>JNJ-R118] - in Biologics Discovery, a Senior Data Scientist for featurisation and model-ready datasets for discovery models, posted under two requisition numbers [>JNJ-R117, >R132], and a Senior ML Engineer, posted under three requisition numbers, to deploy and operate models "developed by partner organizations" for discovery scientists [>JNJ-R186, >R203, >R204] - in the In Silico Discovery department, two Senior Scientist positions in CADD technology development, to build structure-based components for "autonomous design-make-test-analyse (DMTA) cycles" [>JNJ-R190] - in the Data, Data Science and AI (DDSAI) organisation, a Principal Scientist, AI & Autonomous Discovery Systems, to build AI-enabled discovery workflows and "scalable multi-agent architecture for AI-driven laboratory orchestration" (tagged both) [>JNJ-R086], and a Principal Scientist to architect an AI-reasoning platform for translational safety decisions in drug discovery [>JNJ-R200] - a Senior Principal Scientist for translational PK/PD and PBPK modelling on discovery project teams to advise candidate selection [>JNJ-R047]

The PK/PD role is mechanistic modelling, counted as computation applied to discovery. It was screened by one reviewer only and is marked borderline [>JNJ-R047].

The T2 postings describe lab systems, lab automation and scientific data roles: - a Senior Director, Head of Technology for the Therapeutic Areas and External Scientific Innovation, in the Research Systems organisation, posted under three requisition numbers. The role is to act as "the primary Technology partner to DDSAI" for that portfolio and to run in-scope platforms, "partnering with suppliers and Technology Services" [>JNJ-R211, >R233, >R235] - a Technical Product Manager, Chemistry IT Systems, in a Business Technology team, to deliver "cheminformatics and broader discovery informatics platforms" for the molecule discovery organisation [>JNJ-R077] - in Biologics Discovery, a Principal Data Engineer for discovery data products and scientific data models [>JNJ-R184], and a Senior Ontologist, posted under three requisition numbers [>JNJ-R185, >R205, >R206] - discovery lab automation roles: a Principal Scientist, Biologics Optimization, leading automation integration [>JNJ-R056]; an Associate Director, Expression Sciences, for automated expression screening [>JNJ-R084]; a Director, Omics Enabling Platforms, to build "closed-loop, autonomous experimentation workflows" [>JNJ-R160]; and a Principal Engineer in the Pooled Screening Core, Cell & Genetic Medicines, to connect lab automation, instruments and scientific data for cell therapy discovery [>JNJ-R232] - two R&D software engineers in Hyderabad for lab automations, lab systems and digital platforms supporting DMTA cycles [>JNJ-R095, >R096] - two Principal Developer openings for Oncology Data & AI Systems, covering pre-clinical, biomarker lab, omics and clinical data [>JNJ-R193] - a Senior Manager for programme management of the Preclinical Sciences & Translational Safety digital, data, AI and technology portfolio. This row was screened by one reviewer only and is marked borderline [>JNJ-R252]

Two postings that were T2 under v0.4.2 are CMC and process development work, so under the v0.4.3 boundary they are out of scope: a DDSAI Director for AI solutions in Therapeutics Development & Supply, which covers CMC, device development and clinical supply, including model-informed process design and lab-in-the-loop workflows [>JNJ-R196], and a Scientific Manager Lab (Data) Systems in Leiden, for validated data systems and automated instrument data capture in Cell and Gene Therapy Analytical Development [>JNJ-R057]. The Leiden role is recorded as borderline: the boundary names CMC development but not analytical development as such, and this unit supports products moving to the clinic under GMP audit.

The six T1 research training posts are postdoctoral roles: agentic AI for neurodegeneration in the Machine Intelligence team [>JNJ-R018], multimodal representation learning, posted under three requisition numbers [>JNJ-R094, >R112, >R113], machine learning on multi-omic data for target evaluation in Immunology [>JNJ-R198] and co-folding models in In Silico Discovery [>JNJ-R291]. Three internships are tagged to the themes: a computational immunology intern (T1), and interns for toxicology data pipelines and proteomics automation (T2) [>JNJ-R297, >R276, >R289].

Four R&D-wide AI roles without stated discovery duties are classed as enterprise roles: AI literacy, Immunology data science strategy, GenAI evaluation and a Director, World Model & Agentic Learning for the R&D agentic AI platform [>JNJ-R071, >R107, >R228, >R273].

Posting dates come from a Workday field that may record a refresh, so they are not reliable.

No LinkedIn company insights were collected for Johnson & Johnson.

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 126 J&J mentions: 1 explicit use, 17 integration or environment, 3 required experience, 6 preferred experience and 99 alternatives. - Named as existing systems: LabVantage LIMS, in a quality role whose duties are for "the LabVantage LIMS Product within Innovative Medicine (IM) globally". This is role-specific wording in a plant, quality or manufacturing role, outside both themes [>JNJ-R281]. - Integration responsibilities: the Head of Technology postings are to design integrations "across laboratory instruments, ELN, LIMS, research data platforms, knowledge graphs" [>JNJ-R211, >R233, >R235]. The omics director is to "Integrate LIMS/ELN, laboratory execution systems, and QC frameworks" [>JNJ-R160]. The Pooled Screening Core engineer is to connect laboratory automation, analytical instruments and enterprise systems [>JNJ-R232]. The Hyderabad engineer is to work with enterprise capabilities "(AWS, data platforms, workflow engines, lab systems)" [>JNJ-R095]. The ML engineer works between "model-ready data in our data warehouse" and models for discovery scientists [>JNJ-R186]. No ELN or LIMS product is named in these duties. - Required experience: AWS [>JNJ-R232], and HPC and GPU computing in a postdoctoral role [>JNJ-R291]. - Preferred experience: hands-on ELN experience, "especially Benchling" [>JNJ-R232]; LabVantage, in a skills list in the Analytical Development posting, which is outside the themes from v0.4.3 [>JNJ-R057]; LIMS implementation [>JNJ-R084]; lab automation and robotic systems [>JNJ-R118]. - Lists of examples: Schrödinger, BIOVIA, Revvity, OpenEye, ChemAxon, RDKit, Oligokit, KNIME, Pipeline Pilot and Spotfire [>JNJ-R077]; Tecan and Hamilton as liquid-handling platforms [>JNJ-R084]; Neo4j, GraphDB, Amazon Neptune, Stardog and Protégé [>JNJ-R185, >R205, >R206]; Snowflake, AWS, Azure and BigQuery [>JNJ-R184]; MLflow and Weights & Biases [>JNJ-R186, >R203, >R204]; LangGraph, DSPy and mem0 [>JNJ-R200]; PyTorch, TensorFlow and JAX [>JNJ-R086]; AWS, GCP or Azure, and Nextflow or Snakemake [>JNJ-R160].

No category (ELN, LIMS, cloud) is recorded as explicit use for J&J, and no ELN or LIMS product is named as in use in the R&D postings.

Deployment evidence

• Reuters reported on 27 April 2026 that J&J "is using artificial intelligence to slash by half the time it takes to generate new leads for developing drugs", citing the chief information officer, Jim Swanson [>JNJ-E12]. This is a media report of executive remarks. It gives no start date, method or basis for the figure.
• Apheris published consortium results on 14 September 2026 reporting that five members, including J&J, "each fine-tuned OpenFold3 Preview2 locally on their own structures", with more than 20,000 structures in total. The report states that "Johnson & Johnson provided funding for the consortium" [>JNJ-E04]. This is a consortium source, and a J&J employee is a named author.
• The AISB Bind release says each partner "receives the resulting model to run and further tailor within its own environment". The model is still to be trained, so this describes intent, not use [>JNJ-E05].
• BigHat's completion notice says the next stage includes "advancing multiple novel sequences" in neuroscience. It does not say who advances them [>JNJ-E06].
• A posting of 7 July 2026 says J&J's Machine Intelligence team works "in partnership with the c-brAIn academic network" on agentic AI for neurodegeneration drug discovery. No start date is stated, and the posting date may be a refresh date [>JNJ-E13].
• The In Silico Discovery posting states an objective "to advance the core automated functionality from its current prototype stage to full portfolio deployment over the coming years" [>JNJ-R190]. The agentic workflow posting ties its work to "lab-automation readiness as new automation infrastructure comes online" [>JNJ-R118].
• We found no adoption metric for any of these items in the reviewed sources. The CIO's lead-optimisation figure is the only stated benefit found [>JNJ-E12].

Context

Justin Scheer is named as Vice President, Global Head of In Silico Discovery, in the AISB Bind release of 22 September 2026; we found no appointment date [>JNJ-E14]. The Director, World Model & Agentic Learning posting describes "a newly created leadership role within the Generative AI organization", reporting to the Head of Generative AI [>JNJ-R273].

Under rule AR7a, data access is context, not T2. A J&J Innovation blog of 14 November 2025 discusses the UK Biobank whole-genome sequencing dataset, to whose study its author contributed, and says J&J is a founding member of Our Future Health [>JNJ-E16]. A blog of 27 January 2026 describes HERACLES, a Dutch oncology patient-data project launched in October 2022 with twelve other stakeholders; it does not say the project has ended [>JNJ-E17]. Neither source states AI or computational discovery work for the access, so neither is counted in T1.

Other context rows: two TU Delft PhD projects on computational modelling for continuous manufacturing [>JNJ-E18]; a Radboud University and Radboudumc collaboration covering AI, multiple myeloma research and early lung cancer detection, which is clinical in focus [>JNJ-E19]; and an enterprise-wide statement in the 2024 Form 10-K that J&J uses data science, machine learning and other forms of AI across varying parts of its business [>JNJ-E20]. BetaKit reported on 20 August 2025 that J&J is withdrawing its support for the JLabs Toronto incubator by the end of 2025, reportedly part of a wider JLabs scaleback; the report links it to neither theme, so it is context, not a negative change [>JNJ-E21].

We identified no qualifying pause or cancellation in the two themes in the searched sources. The BigHat completion is not one, because no source says the end was early or not renewed [>JNJ-E06]. The dedicated T2 negative-event searches were run in a later top-up and found no theme-linked pause or cancellation; the JLabs Toronto withdrawal they found is not linked to either theme [>JNJ-E21]. That is a search result, not evidence that none occurred.

Coverage limits

• The shared web-search budget ran out during this batch. The searches neither reviewer ran were completed in a top-up on 6 October 2026: the consortium and PPP family (an IHI project search, a general consortium-joining search, a Structural Genomics Consortium, Target 2035 and AIRCHECK search, and an Illumina Alliance for Genomic Discovery check), two academic-collaboration searches, the dedicated T2 negative-event searches and one reviewer's planned newswire search. They added one context row and no counted change [>JNJ-E21]. We identified no consortium launch in either year listing J&J as a partner in the searched sources. The searches also found items older than the two years (an Odyssey Therapeutics AI collaboration of April 2024, and the closure of the San Francisco Bay Campus R&D site in February 2024), which are not recorded as rows. University newsrooms were not searched one by one. One reviewer had scanned all 245 IHI project factsheets directly and found nothing in scope.
• Neither reviewer enumerated the jnj.com newsroom. www.jnj.com and innovativemedicine.jnj.com served a block page, and investor.jnj.com served a Cloudflare challenge. Company-authored releases may exist for the Isomorphic Labs and AISB rows, and for items neither reviewer found [>JNJ-E02, >E05, >E11]. One reviewer read J&J Innovation items through its news API; the other could not reach it.
• The FY2025 Form 10-K was not retrieved: SEC EDGAR refused automated access and annualreports.com disallows its PDFs by robots.txt. The W1 annual-report cell is therefore partial. The 2024 report's first-public date rests on its PDF creation date and is uncertain [>JNJ-E20].
• Results slide decks were not retrieved, and one reviewer could not read the quarterly releases, presentations or transcripts (Nasdaq blocked access).
• Business Wire blocked access and GlobeNewswire failed to respond; PR Newswire search pages render client-side, so one reviewer read J&J's organisation feed there instead.
• Other blocked or refused sites: www.careers.jnj.com (Cloudflare; the Workday careers site was used instead), web.archive.org and archive.org, archive.ph, Fierce Biotech, Endpoints News (robots.txt disallows Claude agents), Pharmaceutical Technology, MarketScreener (the Reuters report was read through Yahoo Finance), PharmaLive, technode.global, ailux.xtalpi.com and app.edgar.tools. last10k.com and boardroomalpha.com disallow access by robots.txt.
• Two independent reviewers searched J&J, and the final rows are their source-verified union. Agreement between them shows consistency, not completeness. Three of the six counted rows were found by one reviewer only, all through J&J Innovation news API records [>JNJ-E01, >E07, >E09].
• Patents are not among the searched signal types. Publication counts are raw Europe PMC query counts and are not events.
• Recruitment: single-reviewer requisitions keep that reviewer's class. One reviewer used the enterprise role class more widely, so the enterprise count of 54 leans on that convention. Two R&D rows are borderline [>JNJ-R047, >R252].

Two questions to investigate (Ortent's interpretation)

1. Lab and scientific data systems for discovery appear in several units. A Head of Technology in the Research Systems organisation is to act as "the primary Technology partner to DDSAI" and to run in-scope platforms with suppliers and Technology Services [>JNJ-R211, >R233, >R235]. A Business Technology team delivers discovery informatics platforms for the molecule discovery organisation [>JNJ-R077]. Biologics Discovery is recruiting its own data engineering, ontology and agentic workflow roles for its "data engine", working with In Silico Discovery and the Enterprise Generative AI team [>JNJ-R184, >R185, >R118]. Omics and screening roles are to integrate LIMS, ELN and lab automation [>JNJ-R160, >R232]. What is the remit of each unit for lab and scientific data systems? Who decides on new systems, and how are the interfaces between Research Systems, Business Technology, DDSAI and the discovery groups governed?
2. J&J has taken access to external AI discovery capability through a platform licence, a research collaboration, two federated consortium initiatives and two J&J Innovation challenges [>JNJ-E01, >E11, >E02, >E05, >E07, >E09]. A Senior ML Engineer is to operate models "developed by partner organizations" [>JNJ-R186], and the AI & Autonomous Discovery Systems role is to work "in collaboration with generic agentic system providers" [>JNJ-R086]. Who holds decision rights over taking in an external model or platform rather than building one, and does that run through J&J Innovation, In Silico Discovery, DDSAI or a procurement route? How is funding set for moving automated discovery workflows from prototype to "full portfolio deployment" [>JNJ-R190]?

Sources for the main claims

• JNJ-E01 XtalPi release, Ailux XtalFold licence (J&J Innovation copy): https://juniverseadmin.jnjinnovation.com/jji-rest-api/getNewsById/143409
• JNJ-E02 Apheris, OpenFold3 to be fine-tuned with pharma data: https://www.apheris.com/resources/alquraishi-lab-s-openfold3-to-be-fine-tuned-with-pharma-industry-data-in-a-secure-ai-collaboration
• JNJ-E07 J&J Innovation, CAT-PD QuickFire Challenge: https://juniverseadmin.jnjinnovation.com/jji-rest-api/getNewsById/144769
• JNJ-E09 J&J Innovation, Data & AI QuickFire Challenge: https://juniverseadmin.jnjinnovation.com/jji-rest-api/getNewsById/144935
• JNJ-E11 Isomorphic Labs and J&J research collaboration: https://storage.googleapis.com/isomorphiclabs-website-public-artifacts/Isomorphic_Labs_Enters_into_a_Research_Collaboration_with_Johnson_Johnson.pdf
• JNJ-E05 and E14 AISB Network, AISB Bind: https://www.apheris.com/resources/aisb-network-launches-aisb-bind-to-improve-prediction-of-how-molecules-bind-for-virtual-screening-and-lead-optimization
• JNJ-E04 AISB consortium results: https://www.apheris.com/resources/federated-training-dramatically-improves-the-accuracy-of-protein-ligand-co-folding-on-private-pharma-structures
• JNJ-E06 BigHat completion: https://www.bighatbio.com/news/bighat-announce-successful-completion-of-our-strategic-collaboration-with-johnson-johnson
• JNJ-E12 Reuters report (Yahoo Finance copy): https://finance.yahoo.com/sectors/healthcare/articles/j-j-sees-ai-halving-150312205.html
• Other event rows and all requisitions: URLs and archived copies in the J&J tables. No SRC source files are cited for J&J.

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.