Verified
Johnson & Johnson
Scaling- Pharma and MedTech
- 140,800 employees (2025)
- CIO-led AI program
- ~900 gen AI use cases reviewed
- 56,000+ trained in gen AI (2025)
Johnson & Johnson deploys AI through a program run by CIO Jim Swanson, an Executive Committee member: staff use an internal generative AI application after a required course, language models sit in secure company enclaves, and in 2025 J&J narrowed about 900 generative AI use cases to priorities in drug development, supply chain, sales and employee services. J&J MedTech builds AI into surgical devices and its Polyphonic digital surgery ecosystem with Nvidia, Microsoft and AWS. Johnson & Johnson has published no dollar return on AI.
jnj.comTheir approach to AI
- The platform
- An internal generative AI application is the sanctioned tool for Johnson & Johnson staff, and the company built secure enclaves for its large and small language models (Mar 2025); JAIDA, a generative AI assistant, helps employees navigate internal systems and policies (Aug 2025).
- The models
- Johnson & Johnson has not named the language models behind its internal tools; Nvidia IGX, Holoscan and RTX computing run its surgical AI, and Nvidia Isaac for Healthcare and Cosmos models support robot simulation in development (Mar 2024 to Oct 2025).
- The reach
- More than 56,000 Johnson & Johnson employees had completed the generative AI course required to use the tools by Mar 2025, and staff logged nearly 120,000 AI learning completions in 2025; J&J has not published active usage.
- The catch
- Johnson & Johnson has put no dollar figure on AI's return: CFO Joe Wolk said in Sep 2026 that the company may get some benefit from technology and AI in its cost review and will see how it plays out, and on Oct 13, 2026 Wolk left margin detail to the Dec 8, 2026 Enterprise Business Review.
Snapshot
- Employees
- 140,800
- Dec 28, 2025; about 138,200 full-time equivalents (2025 Form 10-K)
- Technology budget
- Not disclosed
- R&D investment was $14.7B in 2025 on sales of $94.2B
- AI/ML headcount
- ~7,000est.
- Digital specialists and data scientists (Sep 2026); 4,000+ in J&J Technology
- Who owns AI
- Jim Swanson, EVP and CIO
- Executive Committee member since 2022; an AI Council and a Data Management Council steer the stack
- Platform
- Internal gen AI application
- Secure enclaves for large and small language models (Mar 2025); JAIDA for employee questions
- Models
- Not disclosed
- Nvidia platforms power surgical and robotics AI
- Reach
- 56,000+ trained
- Required gen AI course completed by Mar 2025; active usage not published
- Use cases
- ~900 reviewed
- Narrowed in 2025 to drug discovery, trials, supply chain and sales
Key findings
Johnson & Johnson reviewed nearly 900 generative AI use cases and, after finding that 10% to 15% of them delivered about 80% of the value, moved in 2025 from broad experimentation to a short list of priorities in drug discovery, clinical trials, supply chain and sales, with the business units taking over delivery.
CIO Jim Swanson, a member of Johnson & Johnson's Executive Committee since 2022, leads a Technology organization of more than 4,000 people and said in Apr 2026 that AI had cut lead optimization time in drug discovery in half and sped up two compounds, one in oncology and one in immunology.
Johnson & Johnson requires a generative AI course before staff use its internal generative AI application; more than 56,000 employees had completed it by Mar 2025, and employees logged nearly 120,000 AI learning completions in 2025.
Johnson & Johnson MedTech builds AI into surgery through Polyphonic, an open digital surgery ecosystem announced in Jun 2024 and in early access at select hospitals as of Oct 2025, with Nvidia edge computing since Mar 2024, Microsoft as preferred cloud for digital surgery since Jan 2022, and Nvidia and AWS backing its Polyphonic AI Fund for Surgery since Jun 2025; Abu Dhabi had connected 60 operating rooms to Polyphonic by Oct 2026.
Johnson & Johnson has published no dollar return on AI: CFO Joe Wolk said in May 2025 that AI was helping forecasting and receivables work within a planned operating margin gain, and in Sep 2026 that the cost review around the orthopaedics separation may benefit from AI.
Where J&J sits
- Stage
- Scaling
Johnson & Johnson sits at Scaling: staff have a sanctioned internal generative AI application, more than 56,000 employees had completed the course required to use it by Mar 2025, AI runs in production across drug discovery, clinical trial site selection, supply chain risk, sales, employee services and surgical products, and CIO Jim Swanson, an Executive Committee member, owns the enterprise program. Johnson & Johnson falls short of Embedded on two tests: it has published no weekly or daily usage figure, and it has reported no financial return credited to AI.
How the stages are defined - Ownership
- EVP and CIO Jim Swanson, on the Executive Committee
- Platform
- Hybrid: Internal gen AI app built in-house; Nvidia and cloud platforms bought
- Models
- Undisclosed for internal tools; Nvidia platforms in surgical AI
- Hosting
- Secure company enclaves for language models (cloud not disclosed); Microsoft Azure named digital surgery cloud (2022); Nvidia edge hardware in ORs and devices
- Reach
- 56,000+ employees trained to use gen AI (Mar 2025)
How they deploy
From about 900 pilots to a short list
Johnson & Johnson ran as many as 900 generative AI experiments under a central governing board, then found that 10% to 15% of the use cases delivered about 80% of the value, according to The Batch (May 2025). Johnson & Johnson then handed AI projects to the departments that own them and chose work by ease of implementation, usefulness across the company and business benefit. Swanson described the shift as moving from a thousand flowers to a prioritized focus (CIO.inc, May 2025).
Three layers of AI
Jim Swanson said in Mar 2025 that Johnson & Johnson applies AI in three layers: enterprise productivity, end-to-end processes such as drug discovery, clinical operations and customer engagement, and AI embedded in products and services. Reuters reported in Apr 2026 that Johnson & Johnson focuses its AI on AI-enabled products, drug development and supply chain optimization.
AI in research and clinical operations
Joaquin Duato said in May 2026 that Johnson & Johnson uses AI in two areas of drug development: molecular design, where AI shortens the time from hit to lead and helps predict toxicology and efficacy, and clinical operations, including protocol design, patient selection and choosing trial sites. Duato said several Johnson & Johnson oral medicines reached human testing faster than before AI, without naming them.
Skills before tools
Johnson & Johnson wants a bilingual workforce that pairs domain expertise with technology skills, and requires a generative AI course before staff use generative AI tools (Fortune, Apr 2024). Jim Swanson said in Apr 2026 that AI is an added skill for employees, not a replacement, calling these "'and' skills, not 'or' skills."
An open ecosystem in surgery
Johnson & Johnson MedTech is building Polyphonic as an open, secure ecosystem that takes surgical data from any source, not only its own devices. Hani Abouhalka, Company Group Chairman of Surgery, said in Mar 2026 that operating rooms average more than seven software solutions and five data streams, and that Johnson & Johnson is working with technology partners, academia and hospitals to build structured surgical datasets on which others can build AI algorithms.
- Deployment model
- Johnson & Johnson runs enterprise AI through J&J Technology, an organization of more than 4,000 people in 50 countries led by EVP and CIO Jim Swanson, with an AI Council and a Data Management Council steering the technology stack and use-case choices, and a review group and security and architecture experts checking every proposed generative AI use case for ethical, regulatory and compliance risk (Mar 2025). After the 2025 reprioritization, the business units own delivery: Innovative Medicine R&D runs in silico discovery and clinical trial AI, Innovative Medicine's commercial organization runs Engagement.ai, supply chain teams run risk sensing, Global Services runs employee assistants, and MedTech builds AI into devices, robots and the Polyphonic ecosystem.
- Model providers
- Johnson & Johnson has not named the large language models behind its internal generative AI application; Jim Swanson said in Mar 2025 that the company built secure enclaves for its large and small language models and trains them on vetted, curated data. In MedTech, Nvidia supplies the AI computing stack: IGX and Holoscan for edge AI in the operating room (Mar 2024), RTX-based computing for MONARCH QUEST's AI navigation (Mar 2025), and Isaac for Healthcare, Omniverse libraries and Cosmos world foundation models for robot simulation (Oct 2025). Swanson said in Mar 2025 that Johnson & Johnson's quantum computing work on protein folding will need third-party partners.
- Data and infrastructure
- Johnson & Johnson keeps generative AI inside secure company enclaves and has employee policies on the risks and restrictions of public generative AI tools (Mar 2025); the company has not said which cloud hosts those enclaves. Microsoft became the preferred cloud provider for Johnson & Johnson's digital surgery solutions in Jan 2022, with Azure, Azure IoT and edge computing, AI and machine learning, Microsoft 365 and Dynamics 365 in scope. Nvidia IGX and Holoscan let J&J MedTech analyze operating room data at the edge, which reduces the need to move sensitive data (Mar 2024). In the intelligent operating room network that Johnson & Johnson and the Department of Health Abu Dhabi began building on Polyphonic in May 2026, AWS, Nvidia and Core42 are named technology partners, but neither party has said where the surgical data is hosted. Johnson & Johnson's R&D models draw on de-identified electronic health records, lab results, genetic sequencing data and partner biobanks such as UK Biobank (Oct 2026).
Platforms and internal tools
Internal generative AI applicationBy Mar 2025
Johnson & Johnson's internal generative AI application is the tool staff are directed to instead of public generative AI services; Jim Swanson said in Mar 2025 that the company had built secure enclaves for its large and small language models and uses vetted, curated data. Employees must complete a generative AI course before using generative AI tools (Fortune, Apr 2024).
- Users
- 56,000+ employees trained to use it (Mar 2025)
JAIDABy Aug 2025
JAIDA is a generative AI tool from Johnson & Johnson's Global Services organization that helps employees navigate complex internal systems and policies, according to the Aug 2025 Knowledge at Wharton episode with Ajay Anand, senior vice president of Global Services, Strategy and Business Services. The episode summary says tools like JAIDA help 138,000 employees, a figure that matches Johnson & Johnson's total headcount rather than a measured user count. Jim Swanson had described AI-powered agents that answer employees' service questions and send links to the right documents in Mar 2025; Johnson & Johnson has not said whether they are the same tool.
Engagement.aiBy Mar 2025
Engagement.ai is Johnson & Johnson Innovative Medicine's suite for guiding outreach to healthcare providers and patients, trained on de-identified datasets; Jim Swanson said in Mar 2025 that the company had consolidated all its provider interactions into the tool in more than 14 countries. Ajit Menon, Chief Global Commercial Data and AI Officer, said in Oct 2026 that Johnson & Johnson uses generative AI and AI agents to turn its insights into action for providers and patients.
AI Advisor and AskJIA2025 pilot
AI Advisor is a chatbot that gives Johnson & Johnson commercial and sales teams compliant answers to customer questions, piloted with more than 1,000 users in 2025. AskJIA, built from Johnson & Johnson's medically validated and legally reviewed product content, was being deployed to train sales representatives, Jim Swanson said in a Mar 2025 Greylock interview.
- Users
- 1,000+ in the AI Advisor pilot (2025)
PolyphonicJun 2024
Polyphonic is Johnson & Johnson MedTech's open digital ecosystem for surgery, connecting surgical video, device data and AI insights for case planning, telepresence and post-case review, with Polyphonic Studio and Polyphonic Store announced for later releases. Polyphonic remained in early access at select hospitals in Oct 2025, connects to the OTTAVA robot (Aug 2026) and the MONARCH platform (Aug 2026), and anchors an intelligent operating room network with the Department of Health in Abu Dhabi that began in May 2026 and had connected 60 operating rooms by Oct 2026, according to Khaleej Times.
Supply risk sensing platformBy Mar 2025
Johnson & Johnson runs an AI-enabled platform that predicts and flags supply risks such as a fire at a supplier site, mapping the effect on materials and products (Mar 2025), and uses machine learning on data such as demand swings and supplier performance to predict the impact of real-time events that may disrupt supply (Oct 2026).
Vendors and partners
NvidiaMar 2024
AI computing for surgical devices, the operating room edge and robot simulation; Polyphonic AI Fund coalition member · Johnson & Johnson MedTech signed a memorandum of understanding in Mar 2024 to use Nvidia IGX and Holoscan in its digital surgery ecosystem, runs MONARCH QUEST navigation on Nvidia computing (Mar 2025), uses Nvidia Isaac for Healthcare in developing the MONARCH Platform for Urology (Oct 2025), and named Nvidia a technology partner in the Abu Dhabi operating room network (May 2026).
MicrosoftJan 2022
Preferred cloud provider for digital surgery solutions · The Jan 2022 agreement covered Azure, Azure IoT and edge computing, AI and machine learning, Microsoft 365 and Dynamics 365 for the digital surgery platform and device connectivity.
Amazon Web ServicesJun 2025
Polyphonic AI Fund for Surgery coalition member; technology partner in the Abu Dhabi operating room network · AWS joined the committee that evaluates and selects Polyphonic AI Fund awardees (Jun 2025) and adds cloud computing capabilities to the fund (MedTech Dive, Jun 2025). Johnson & Johnson named AWS a technology partner in the Abu Dhabi network in May 2026 without describing its role.
Department of Health, Abu DhabiMay 2026
Building an intelligent operating room network on Polyphonic · The network launched on May 21, 2026 with Cleveland Clinic Abu Dhabi, PureHealth, Mediclinic Group and NMC Healthcare deploying Polyphonic, and 60 operating rooms were connected by Oct 2026 (Khaleej Times). Johnson & Johnson said in Oct 2026 that the Department is standardizing operating rooms on Polyphonic.
Core42May 2026
Technology partner in the Abu Dhabi operating room network · The Abu Dhabi Media Office named Core42 alongside AWS and Nvidia as an existing technology partner of the network; no party has described its role.
Google.orgApr 2026
Matching funder for AI tools for U.S. rural health workers · The J&J Foundation committed $5 million and Google.org matched it, for $10 million over three years.
Leadership
AI, data and platform leaders
Executives
AI org map
CIO-led enterprise AI with central councils; delivery owned by the business units and MedTech product teams
Johnson & Johnson has no chief AI officer: EVP and CIO Jim Swanson, an Executive Committee member since 2022, leads J&J Technology and the enterprise AI program, with an AI Council and a Data Management Council guiding the stack and use-case choices (Mar 2025). After the 2025 reprioritization, responsibility for AI projects moved to the departments that own them, so Innovative Medicine, its R&D organization, supply chain, Global Services and MedTech each run their own AI teams. The Board oversees AI use through its committees, with the Science & Technology Committee covering emerging technologies such as AI (2026 proxy).
Johnson & Johnson says it employs nearly 7,000 digital specialists and data scientists worldwide (Sep 2026), and J&J Technology, the central IT organization under CIO Jim Swanson, has more than 4,000 staff in 50 countries; Johnson & Johnson has not said how the two counts overlap. Data science and AI teams also sit inside the businesses: in silico discovery and AI/ML and computer vision groups in Innovative Medicine R&D, a commercial data and AI office in Innovative Medicine, supply chain analytics, and MedTech's robotics and digital R&D and Polyphonic units.
Reporting lines
Chairman and CEO
J&J Technology
Jim Swanson, EVP and CIO
More than 4,000 staff; Executive Committee member; the line to the CEO is inferred from Executive Committee membership, not published
AI Council and Data Management Council
Steer the technology stack and use-case selection (Mar 2025); membership and chairs not published, and the line to J&J Technology is inferred
Innovative Medicine R&D
John Reed, EVP
In Silico Discovery
Justin Scheer, VP
Innovative Medicine
Tom Cavanaugh, EVP, Worldwide Chairman
Cavanaugh succeeded Jennifer Taubert on Sep 1, 2026
Global Commercial Data and AI
Ajit Menon
Part of the Innovative Medicine Global Commercial Strategy Organization; runs Engagement.ai
Technical Operations and Risk
Kathryn Wengel, EVP
Supply chain standards, quality and compliance, global security
Finance
Joe Wolk, EVP and CFO
MedTech
Tim Schmid, EVP, Worldwide Chairman
Surgery
Hani Abouhalka, Company Group Chairman
Polyphonic
Shan Jegatheeswaran, President
Reporting line within MedTech not published
Robotics and Digital R&D
Neda Cvijetic, SVP
Global Services
Ajay Anand, SVP
Shared services across HR, finance and procurement; runs JAIDA; reporting line not published
Johnson & Johnson's leadership page, checked Nov 1, 2026, lists ten Executive Committee members, with Jim Swanson the only technology executive and no chief data or AI officer. Johnson & Johnson does not publish reporting lines below the Executive Committee, so the placement of Polyphonic, Global Services and the AI councils is inferred from titles and press releases.
Where the AI workforce sits
J&J Technology (central IT, data and AI platforms)
Rough estimate: more than 4,000 Technology staff against the nearly 7,000 digital specialists and data scientists Johnson & Johnson cites (Sep 2026); the counts may define roles differently or overlap.
~55-60%Digital and data science teams in Innovative Medicine and MedTech
Rough estimate: the remainder of the nearly 7,000 figure, including R&D data science, commercial data and AI, and MedTech robotics and digital teams.
~40-45%
Named people
Use cases
Johnson & Johnson reviewed nearly 900 generative AI use cases and in 2025 narrowed them to a short list in drug discovery, clinical trials, supply chain, sales and employee services, alongside AI built into MedTech products such as CARTO 3, MONARCH QUEST, VIRTUGUIDE and VELYS Hip Navigation. Johnson & Johnson has published no business-level return credited to AI for any use case: the figures it cites are speedups, such as lead optimization time cut in half and a clinical report going from 700 hours to about 15 minutes (Apr 2026), higher enrollment at AI-recommended trial sites, or results for hospitals and surgeons. Johnson & Johnson has put no firm-level dollar figure on AI.
| Use case | Area | Status | ROI |
|---|---|---|---|
| AI-assisted molecular design Johnson & Johnson uses AI to screen chemical compounds and biologics and to optimize drug candidates, in-house and with partners; Jim Swanson said AI cut lead optimization time in half and sped up two compounds, one in oncology and one in immunology (Apr 2026). | Innovative Medicine R&D | In production | |
| Clinical trial site selection Johnson & Johnson uses AI and machine learning to find trial sites with patients who could benefit from the medicines studied and to optimize inclusion criteria (Oct 2026); Jim Swanson said in Mar 2025 that sites recommended by its AI models enrolled up to 2.6 times more patients. | Innovative Medicine R&D | In production | |
| Clinical and regulatory document drafting Johnson & Johnson uses AI to prepare documents for regulators; Jim Swanson said a clinical trial report that took 700 to 900 hours now takes about 15 minutes (Apr 2026). | Innovative Medicine R&D | In production | |
| AI biomarker test for bladder cancer An AI-assisted test reads digitized biopsy images to detect likely FGFR alterations in bladder cancer; Johnson & Johnson has used it in prior screening and is evaluating it for trials and routine care (Oct 2026). | Innovative Medicine R&D | Pilot | |
| Pharmacovigilance Johnson & Johnson uses generative and agentic AI to strengthen pharmacovigilance, improve data quality and update product safety profiles faster (2025 Health for Humanity Report). | Medical safety | In production | |
| Supply risk sensing An AI-enabled platform predicts supplier, material and product risks (Mar 2025), and machine learning on data such as demand swings and supplier performance predicts the impact of real-time events that may disrupt supply (Oct 2026). | Supply chain and manufacturing | In production | |
| Chemical process design An AI system helps Johnson & Johnson design chemical manufacturing processes, such as when to add a solvent and at what temperature to turn a liquid into a solid. | Supply chain and manufacturing | In production | |
| Engagement.ai Engagement.ai guides Johnson & Johnson's outreach to healthcare providers and patients using generative AI and AI agents (Oct 2026); Jim Swanson said in Mar 2025 that it held the company's provider interactions in more than 14 countries. | Innovative Medicine commercial | In production | |
| Sales copilot A sales copilot built into the customer relationship management system of a cancer-treatment unit gives representatives medically validated, legally reviewed product and customer information, and is being adapted for MedTech sales of robotics and hip implants (May 2025). | Innovative Medicine commercial | In production | |
| AI Advisor AI Advisor is a chatbot that answers customer questions for commercial and sales teams with compliant responses, piloted with more than 1,000 users in 2025. | Commercial | Pilot | |
| Employee service assistants and JAIDA AI agents answer employees' questions on benefits, policies and procedures with links to the relevant documents, and the JAIDA generative AI tool helps staff navigate internal systems and policies. | Global Services | In production | |
| Forecasting and receivables Johnson & Johnson Finance uses AI to make forecasting more predictable and, with Global Services, to process accounts receivable and disputes faster (May 2025); the 2026 proxy cites AI-enabled tools for forecasting and risk management. | Finance | In production | |
| AI-enabled quality management An AI-enabled Quality Management System digitized quality workflows and added a global product complaint intake portal to speed investigations and standardize corrective actions (2025). | MedTech | In production | |
| Electrophysiology mapping Version 8 software for the CARTO 3 System uses deep learning in heart mapping for atrial fibrillation ablation, and CARTOSOUND SONATA added AI-powered imaging and mapping in the second quarter of 2026. | MedTech | In production | |
| Robotic bronchoscopy navigation MONARCH QUEST, cleared in Mar 2025, runs AI-powered lung navigation on Nvidia accelerated computing, and MONARCH QUEST 3, cleared in Aug 2026, adds AI-powered nodule segmentation for procedure planning and connects cases to Polyphonic. | MedTech | In production | |
| Orthopaedic planning and navigation The VIRTUGUIDE AI-powered patient-matched system plans bunion surgery (U.S. launch 2025), and VELYS Hip Navigation with AI automates image interpretation and landmark identification while surgeons keep control of final landmark placement (U.S. launch Jun 2026). | MedTech | In production | |
| Polyphonic surgical video and insights Johnson & Johnson is developing AI tools that cut surgical video into highlight reels in minutes instead of hours or days (Oct 2026); Polyphonic connects surgical data, video, devices and AI insights and supports telepresence and post-case review, in early access at select hospitals and in the Abu Dhabi operating room network. | MedTech | Pilot | |
| Robot simulation and digital twins Johnson & Johnson uses AI-driven simulation in developing the MONARCH Platform for Urology, building digital twins with Nvidia Isaac for Healthcare and generating synthetic data with Cosmos models, with the digital assets also meant to support training for clinical teams (Oct 2025). | MedTech | Pilot |
Adoption
Johnson & Johnson has used traditional AI for nearly a decade in surgical guidance software, drug discovery and manufacturing inventory, according to Business Insider (Mar 2025), and gated generative AI behind training: about 20,000 employees had taken the required generative AI course by Apr 2024 and more than 56,000 of about 138,000 by Mar 2025, while Jim Swanson cited 47,000 in an interview published the same month. Johnson & Johnson employees logged nearly 120,000 completions and nearly 80,000 hours of AI learning in 2025, more than 14,000 had completed deeper digital boot camps by Mar 2025, and more than 2,500 joined a six-week digital immersion program in 2024. Johnson & Johnson has not published how many employees use its generative AI tools each week; an Aug 2025 description of JAIDA as helping 138,000 employees matches total headcount rather than measured use, and the AI Advisor sales chatbot was piloted with more than 1,000 users in 2025.
By the numbers
Rollout, first pilot to today
Jan 2022
The Johnson & Johnson Medical Devices Companies made Microsoft the preferred cloud provider for their digital surgery solutions, with Azure, AI and machine learning in scope.
2022
Jim Swanson, CIO since 2019, became an Executive Vice President and joined Johnson & Johnson's Executive Committee.
Mar 2024
Johnson & Johnson MedTech signed a memorandum of understanding with Nvidia to bring IGX and Holoscan edge AI into its digital surgery ecosystem.
Apr 2024
About 20,000 Johnson & Johnson employees had taken the generative AI course required to use generative AI tools.
Jun 2024
Johnson & Johnson MedTech announced Polyphonic, an open digital ecosystem for surgery, in early access at select hospitals.
Mar 2025
Jim Swanson said Johnson & Johnson had nearly 900 generative AI use cases under review, secure enclaves for its language models and an internal generative AI application for staff, and that 47,000 employees had taken the required generative AI course; Business Insider put the figure above 56,000 later that month.
May 2025
Johnson & Johnson's shift from about 900 generative AI experiments to a short list of priority use cases, owned by the departments, was reported.
Jun 2025
Johnson & Johnson launched the Polyphonic AI Fund for Surgery with Nvidia and AWS.
Aug 2025
Ajay Anand described JAIDA, a generative AI tool helping Johnson & Johnson employees navigate internal systems and policies.
Oct 2025
Johnson & Johnson named the first Polyphonic AI Fund awardees, Mayo Clinic and QAS.AI, and described robot simulation with Nvidia Isaac for Healthcare.
Apr 2026
Jim Swanson said AI had halved lead optimization time in Johnson & Johnson's drug discovery and sped up two compounds.
Jun 2026
Johnson & Johnson reported nearly 120,000 AI learning completions in 2025 and a pilot of the AI Advisor sales chatbot with more than 1,000 users.
Economics & controls
Spend and ROI
Technology and R&D spending
Johnson & Johnson invested $14.7 billion in research and development in 2025 on worldwide sales of $94.2 billion, according to its 2025 Form 10-K, and the 2026 proxy credits John Reed with deepening investment in data science and digital health. Johnson & Johnson does not disclose a technology budget or a figure for AI spending.
What AI itself costs
Johnson & Johnson has not disclosed what its AI programs cost. Its published AI commitments are small: up to $100,000 in grant funding through the Polyphonic AI Fund for Surgery (MedTech Dive, Jun 2025) and a $5 million J&J Foundation commitment, matched by Google.org, for AI tools for U.S. rural health workers over three years (Apr 2026).
Reported ROI
Johnson & Johnson has published no dollar, cost or headcount return credited to AI. The figures it cites are value concentration (10% to 15% of generative AI use cases delivering about 80% of the value, 2025), speedups (lead optimization time cut in half and a clinical report going from 700 hours to about 15 minutes, Apr 2026) and higher enrollment at AI-recommended trial sites (up to 2.6 times, Mar 2025). On the Q3 2026 earnings call on Oct 13, 2026, Joe Wolk tied future margin gains to technology advances without naming AI and said more detail would come at the Dec 8, 2026 Enterprise Business Review.
Margins and the cost review
Joe Wolk said in May 2025 that about 150 basis points of Johnson & Johnson's planned 300 basis point operating margin improvement for 2025 would come from operational improvement, some of it from technological advances, with AI making forecasting more predictable and speeding receivables and dispute processing in global services. Wolk said in Mar 2026 that the orthopaedics separation lets Johnson & Johnson review its cost structure and use new capabilities such as AI to get leaner, and in Sep 2026 that the company may benefit from technology and AI in that review.
Headcount
Johnson & Johnson had about 140,800 employees at Dec 28, 2025, up from 139,800 a year earlier, or about 138,200 full-time equivalents. Jim Swanson said in Apr 2026 that about 4,000 of them work in information technology and that AI is an added skill for employees rather than a replacement, and Johnson & Johnson has not tied any job reductions to AI.
Governance and risk
Board oversight
Johnson & Johnson's Board and its committees oversee the company's use of AI and receive regular updates on AI developments, with the Science & Technology Committee covering emerging technologies such as AI and the Regulatory Compliance & Sustainability Committee overseeing cybersecurity risk, according to the 2026 proxy. Artificial intelligence was among the topics investors raised with Johnson & Johnson in 2025.
Principles and councils
Johnson & Johnson says fairness, privacy, security, responsibility and transparency guide its work with AI (2026 proxy). Jim Swanson said in Mar 2025 that an AI Council and a Data Management Council steer the technology stack, that a group reviews every generative AI use case for ethical, regulatory and compliance risk, and that security and architecture experts review each proposal.
Data boundary and training gate
Johnson & Johnson keeps language models in secure company enclaves, trains them on vetted and curated data, and has policies on the risks and restrictions of public generative AI tools that point staff to the internal application (Mar 2025). Employees must complete a generative AI course before using generative AI tools (Fortune, Apr 2024), and Johnson & Johnson added healthcare compliance guidance on generative AI and scientific independence in 2025.
AI in medical devices
Johnson & Johnson MedTech states that Polyphonic is not intended to direct surgery or to aid diagnosis or treatment (Jun 2024), and VELYS Hip Navigation with AI uses a human-in-the-loop design in which the surgeon makes the final landmark placement (Jun 2026). Nvidia edge computing lets J&J MedTech analyze operating room data on site instead of moving sensitive data (Mar 2024).
Disclosed risks
Johnson & Johnson's 2025 Form 10-K warns that adopting AI may bring unintended consequences, that AI investments may not deliver the expected benefits or may become obsolete early, that new laws such as the EU AI Act may raise costs or limit use, and that wider use of AI raises cybersecurity risk. The CIO and the Chief Information Security Officer share responsibility for managing cybersecurity risk.
Recent signals
Oct 22, 2026
Khaleej Times reported that Abu Dhabi had connected 60 operating rooms to Johnson & Johnson's Polyphonic platform, with Tim Schmid calling the network a first for the company.
Oct 20, 2026
Johnson & Johnson said the Department of Health in Abu Dhabi is standardizing operating rooms on Polyphonic to build a foundation for surgical intelligence, and announced plans to work with Hub71 on surgical startups.
Oct 6, 2026
Joaquin Duato called AI a force multiplier that has to be embraced, in a Bloomberg interview recorded in June.
Sep 17, 2026
Joe Wolk said Johnson & Johnson is reviewing its infrastructure in a deep exercise around the orthopaedics separation and may benefit from technology and AI, with any savings aimed at margins and R&D.
Sep 4, 2026
Johnson & Johnson said it employs nearly 7,000 digital specialists and data scientists and is supporting New Jersey's effort to build an AI hub.
Aug 18, 2026
Johnson & Johnson published a Future of Digital Surgery report that defines surgical intelligence, drawing on interviews with more than 40 senior leaders.
Aug 17, 2026
Johnson & Johnson received U.S. 510(k) clearance for MONARCH QUEST 3, robotic bronchoscopy software that adds AI-powered nodule segmentation for procedure planning and connects to Polyphonic.
Aug 3, 2026
On an investor call after FDA authorization of the OTTAVA robot, Neda Cvijetic said Polyphonic for OTTAVA captures surgical data and brings data-driven insights to OTTAVA teams.
Jul 23, 2026
Johnson & Johnson named three 2026 Polyphonic AI Fund awardees: Curium Life, C1DO1 and iQ Surgical.
Jul 15, 2026
Joaquin Duato reported the debut of CARTOSOUND SONATA, which brings AI-powered imaging and mapping to Johnson & Johnson's electrophysiology business.
Jun 9, 2026
DePuy Synthes launched VELYS Hip Navigation with AI in the U.S., software that automates landmark identification while surgeons keep control of final placement.
May 27, 2026
Joaquin Duato said Johnson & Johnson uses AI in molecular design and clinical operations and that several of its oral medicines reached human testing faster with AI.
May 21, 2026
Johnson & Johnson and the Department of Health Abu Dhabi launched an intelligent operating room network on Polyphonic, with Cleveland Clinic Abu Dhabi, PureHealth, Mediclinic Group and NMC Healthcare connecting operating rooms and AWS and Nvidia named as technology partners.
Apr 27, 2026
Jim Swanson told a Reuters event that AI had cut Johnson & Johnson's lead optimization time in half and that a clinical report taking 700 hours now takes about 15 minutes.
Mar 11, 2026
Johnson & Johnson's 2026 proxy said the Board and its committees oversee the company's use of AI and that AI was among the topics investors raised in 2025.
Quick answers
- How does Johnson & Johnson deploy AI?
- Johnson & Johnson deploys AI through a program run by CIO Jim Swanson, an Executive Committee member: staff use an internal generative AI application after a required course, language models sit in secure company enclaves, and in 2025 J&J narrowed about 900 generative AI use cases to priorities in drug development, supply chain, sales and employee services. J&J MedTech builds AI into surgical devices and its Polyphonic digital surgery ecosystem with Nvidia, Microsoft and AWS. Johnson & Johnson has published no dollar return on AI.
- What does Johnson & Johnson use AI for?
- Johnson & Johnson reviewed nearly 900 generative AI use cases and in 2025 narrowed them to a short list in drug discovery, clinical trials, supply chain, sales and employee services, alongside AI built into MedTech products such as CARTO 3, MONARCH QUEST, VIRTUGUIDE and VELYS Hip Navigation. Johnson & Johnson has published no business-level return credited to AI for any use case: the figures it cites are speedups, such as lead optimization time cut in half and a clinical report going from 700 hours to about 15 minutes (Apr 2026), higher enrollment at AI-recommended trial sites, or results for hospitals and surgeons. Johnson & Johnson has put no firm-level dollar figure on AI.
- Who leads AI at Johnson & Johnson?
- AI at Johnson & Johnson is led by Jim Swanson (Executive Vice President and Chief Information Officer); Shan Jegatheeswaran (President, Polyphonic, MedTech); Neda Cvijetic (Senior Vice President, Global Head of Robotics and Digital R&D, MedTech). Ownership of AI on the org chart: EVP and CIO Jim Swanson, on the Executive Committee.
- What AI platform does Johnson & Johnson use?
- An internal generative AI application is the sanctioned tool for Johnson & Johnson staff, and the company built secure enclaves for its large and small language models (Mar 2025); JAIDA, a generative AI assistant, helps employees navigate internal systems and policies (Aug 2025).
- Which AI models does Johnson & Johnson use?
- Johnson & Johnson has not named the language models behind its internal tools; Nvidia IGX, Holoscan and RTX computing run its surgical AI, and Nvidia Isaac for Healthcare and Cosmos models support robot simulation in development (Mar 2024 to Oct 2025).
- How many J&J employees use AI?
- More than 56,000 Johnson & Johnson employees had completed the generative AI course required to use the tools by Mar 2025, and staff logged nearly 120,000 AI learning completions in 2025; J&J has not published active usage.
- How far has AI spread at Johnson & Johnson?
- Johnson & Johnson sits at Scaling: staff have a sanctioned internal generative AI application, more than 56,000 employees had completed the course required to use it by Mar 2025, AI runs in production across drug discovery, clinical trial site selection, supply chain risk, sales, employee services and surgical products, and CIO Jim Swanson, an Executive Committee member, owns the enterprise program. Johnson & Johnson falls short of Embedded on two tests: it has published no weekly or daily usage figure, and it has reported no financial return credited to AI.
Open questions
- Which large language models power Johnson & Johnson's internal generative AI application and JAIDA, and on which cloud do the secure enclaves run?
- How many Johnson & Johnson employees use its generative AI tools each week or each day, as opposed to completing the required training?
- What cost, revenue or headcount return do Johnson & Johnson's priority AI use cases deliver, given that the company cites only speedups and enrollment gains?
- How much of Johnson & Johnson's 2025 operating margin improvement, and of the cost review around the orthopaedics separation, comes from AI?
- How many of the nearly 900 generative AI use cases Johnson & Johnson reviewed remain in production after the 2025 narrowing?
- Who sits on Johnson & Johnson's AI Council and Data Management Council, and does a single executive lead data science across Innovative Medicine R&D?
- Is AI Advisor the same tool as the sales copilot described in 2025, and will it move beyond its pilot of more than 1,000 users?
- When will Polyphonic move beyond early access, how many hospitals use it outside Abu Dhabi, and which cloud hosts its surgical data, given that AWS and Core42 are named partners in the Abu Dhabi network without a stated role?
- Will Johnson & Johnson quantify AI's contribution at its Enterprise Business Review on Dec 8, 2026?
- What has Johnson & Johnson's quantum computing work on protein folding produced, and with which partners?
Sources
Johnson & Johnson
- Leadership and Executive Committee
- Jim Swanson, EVP and CIO, biography
- Joaquin Duato, Chairman and CEO, biography
- Joe Wolk, EVP and CFO, biography
- Tom Cavanaugh, EVP, Worldwide Chairman, Innovative Medicine, biography
- Kathryn Wengel, EVP, Chief Technical Operations and Risk Officer, biography
- Tim Schmid, EVP, Worldwide Chairman, MedTech, biography
- John Reed, EVP, Innovative Medicine R&D, biography
- 6 ways Johnson & Johnson is using AI (updated Oct 26, 2026)
- Artificial intelligence advancing healthcare from New Jersey to the world (Sep 2026)
- J&J MedTech working with Nvidia to scale AI for surgery (Mar 18, 2024)
- Advancing robotics development with Nvidia Isaac for Healthcare (Oct 28, 2025)
- J&J MedTech announces the Polyphonic digital ecosystem (Jun 20, 2024)
- Strategic partnership with Microsoft for digital surgery (Jan 10, 2022)
- Polyphonic AI Fund for Surgery launch (Jun 25, 2025)
- Polyphonic AI Fund first awardees (Oct 27, 2025)
- Polyphonic AI Fund 2026 awardees (Jul 23, 2026)
- Future of Digital Surgery report on surgical intelligence (Aug 18, 2026)
- Surgical intelligence at the Abu Dhabi Future Health Summit (Oct 20, 2026)
- VELYS Hip Navigation with AI launch (Jun 9, 2026)
- MONARCH QUEST clearance (Mar 12, 2025)
- MONARCH QUEST 3 clearance (Aug 17, 2026)
- Johnson & Johnson MedTech: global intelligent OR network with the Department of Health Abu Dhabi (May 21, 2026)
- J&J Foundation commits $5M for AI tools for rural health workers (Apr 14, 2026)
Filings, reports and investor materials
- Johnson & Johnson 2025 Form 10-K (filed Feb 11, 2026)
- Johnson & Johnson 2026 proxy statement (filed Mar 11, 2026)
- Form 8-K on Jennifer Taubert's retirement (Aug 4, 2026)
- Press release: Tom Cavanaugh appointed Worldwide Chairman, Innovative Medicine (Aug 4, 2026)
- 2025 Health for Humanity Report (published Jun 2026)
- Enterprise Business Review announcement (Oct 27, 2026)
- Q3 2026 earnings call transcript (Oct 13, 2026)
- Deutsche Bank Healthcare Summit transcript, Joe Wolk (Sep 17, 2026)
- OTTAVA investor call transcript (Aug 3, 2026)
- Q2 2026 earnings call transcript (Jul 15, 2026)
- Bernstein Strategic Decisions Conference transcript, Joaquin Duato (May 27, 2026)
- Barclays Global Healthcare Conference transcript, Hani Abouhalka (Mar 10, 2026)
- TD Cowen Health Care Conference transcript, Joe Wolk (Mar 3, 2026)
- Bernstein Healthcare Forum transcript, Tom Cavanaugh (Sep 24, 2025)
- Wells Fargo Healthcare Conference transcript, Tim Schmid (Sep 3, 2025)
- Bernstein Strategic Decisions Conference transcript, Joe Wolk (May 28, 2025)
News and interviews
- Reuters via Rappler: J&J sees AI halving the time to generate drug development leads (Apr 27, 2026)
- Khaleej Times: Abu Dhabi links operating rooms to AI platform (Oct 22, 2026)
- Abu Dhabi Media Office: Department of Health launches intelligent surgical network with Johnson & Johnson (May 21, 2026)
- MT Newswires via MarketScreener: Johnson & Johnson CEO says AI has to be embraced (Oct 6, 2026)
- CDO Magazine: Johnson & Johnson promotes Ajit Menon (May 2026)
- Greylock: Gen AI present and future, a conversation with Jim Swanson (Mar 13, 2025)
- The Batch: Johnson & Johnson reveals its revised AI strategy (May 7, 2025)
- CIO.inc: Johnson & Johnson taps AI to advance surgery, drug discovery (May 23, 2025)
- Fortune: Johnson & Johnson CIO says employee upskilling is key (Apr 3, 2024)
- Business Insider Japan, translating Business Insider: pharma companies train employees in AI (Mar 25, 2025)
- Knowledge at Wharton: Johnson & Johnson's AI journey, with Ajay Anand (Aug 7, 2025)
- MedTech Dive: J&J, Nvidia and Amazon fund AI surgery solutions (Jun 26, 2025)
Data confidence
- Confirmed
- Stated in a primary or reputable public source.
- Estimated
- Derived or approximated from public reporting.
- Unconfirmed
- Reported, but not yet corroborated.
Johnson & Johnson profile · Enterprise AI diffusion report by Imagine · facts checked against their sources on Nov 1, 2026
