Who leads AI at Toyota?

AI at Toyota is led by Gill Pratt (Chief Scientist and Executive Fellow for Research, Toyota; CEO, Toyota Research Institute); Keiji Yamamoto (Chief Officer, Digital Information and Communications Group, Toyota Motor Corporation); Akihiro Sarada (President, Software Development Center, Toyota Motor Corporation). Ownership of AI on the org chart: No group AI chief; CTO Hiroki Nakajima, Chief Scientist Gill Pratt (TRI) and regional heads such as Brian Kursar (TMNA).

What AI platform does Toyota use?

Toyota runs no single AI platform: an in-house factory AI Platform serves plant staff in Japan, O-Beya serves powertrain engineers, ToyotaGPT serves Toyota Motor North America, and vista on Databricks became the unified data and AI hub in Jan 2026.

Which AI models does Toyota use?

O-Beya runs OpenAI's GPT-4o on Microsoft Azure OpenAI Service, Toyota built a code assistant on NVIDIA Megatron-LM, Woven by Toyota builds its own vision-language model and driving stack, and Toyota Research Institute trains large behavior models for robots.

How many Toyota employees use AI?

Toyota has published no company-wide AI user count; its factory AI Platform had nearly 1,200 active users across 10 plants in Japan (Dec 2024), about 800 powertrain engineers have had O-Beya since Jan 2024, and Toyota Motor North America runs more than 50 agents (Aug 2026).

How far has AI spread at Toyota?

Toyota sits at Piloting: sanctioned AI tools are live for selected groups, such as a factory AI Platform with nearly 1,200 active users in its 10 Japanese car and unit plants (Dec 2024), O-Beya for about 800 powertrain engineers, and ToyotaGPT at Toyota Motor North America with more than 50 agents in production (Aug 2026). Toyota falls short of Scaling because it has published no AI tool open to most of its 390,927 employees, no group-level AI owner who reports results, and no business return from AI.

Verified Nov 1, 2026

Toyota

Piloting

Automotive
390,927 employees
¥1.52T R&D (FY2026)
No company-wide AI platform
¥500B AI platform with NTT by 2030

Toyota deploys AI through its business units and research arms rather than one company-wide platform. Plant staff in Japan build their own models on an in-house factory AI Platform running on Google Cloud and on-premises servers, powertrain engineers use O-Beya agents on Microsoft Azure OpenAI, and Toyota Motor North America runs more than 50 agents in production. Toyota Research Institute and Woven by Toyota build the robot and driving AI, and CTO Hiroki Nakajima told investors in Sep 2026 that Toyota plans about 400,000 AI-driven factory robots from 2028.

global.toyota

Their approach to AI

The platform: Toyota runs no single AI platform: an in-house factory AI Platform serves plant staff in Japan, O-Beya serves powertrain engineers, ToyotaGPT serves Toyota Motor North America, and vista on Databricks became the unified data and AI hub in Jan 2026.
The models: O-Beya runs OpenAI's GPT-4o on Microsoft Azure OpenAI Service, Toyota built a code assistant on NVIDIA Megatron-LM, Woven by Toyota builds its own vision-language model and driving stack, and Toyota Research Institute trains large behavior models for robots.
The reach: Toyota has published no company-wide AI user count; its factory AI Platform had nearly 1,200 active users across 10 plants in Japan (Dec 2024), about 800 powertrain engineers have had O-Beya since Jan 2024, and Toyota Motor North America runs more than 50 agents (Aug 2026).
The catch: Toyota reports no business return from AI, and its Jun 2026 annual report commits only to step up AI use for routine work, with no target or figure attached.

Snapshot

Employees: 390,927; Consolidated at Mar 31, 2026 (Form 20-F); Hino Motors left consolidation on Apr 1, 2026
R&D spending: ¥1.52T (FY2026); ¥1.33T in FY2025 and ¥1.20T in FY2024; no AI budget disclosed
AI/ML headcount: Not disclosed; Toyota Motor North America's enterprise AI team is about 35 people (Aug 2026)
Who owns AI: No group AI chief; CTO Hiroki Nakajima presented the robot AI plan; Gill Pratt is Chief Scientist and TRI CEO; Brian Kursar heads AI at Toyota Motor North America
Platform: Several, by unit; Factory AI Platform (plants, Japan), O-Beya (powertrain), ToyotaGPT (North America), vista on Databricks (data and AI hub)
Models: OpenAI GPT-4o via Azure, plus in-house; A code assistant on NVIDIA Megatron-LM; Woven by Toyota's vision-language model; TRI's large behavior models for robots
Reach: Not disclosed company-wide; Nearly 1,200 factory AI Platform users (Dec 2024); about 800 powertrain engineers on O-Beya since Jan 2024
Focus areas: 11 GAIA categories; From automated driving and robotics to office productivity (May 2025)

Key findings

Toyota has put no company-wide AI platform on the public record: its in-house factory AI Platform had nearly 1,200 active users in all 10 of its car and unit plants in Japan in Dec 2024, O-Beya has served about 800 powertrain engineers since Jan 2024, and Toyota Motor North America's AI team said in Aug 2026 that its ToyotaGPT platform runs more than 50 agents in production.
Toyota and NTT agreed on Oct 31, 2024 to invest a combined ¥500 billion by 2030 in a Mobility AI Platform of distributed data centers, communications and AI infrastructure, with development from 2025 and wider use planned from 2030.
Toyota launched the Global AI Accelerator (GAIA) on May 22, 2025 to expand AI research and development across 11 areas, from automated driving to office productivity, alongside a Toyota Software Academy of about 100 courses run with AISIN, DENSO, Toyota Tsusho and Woven by Toyota.
Kenta Kon, a former CFO of both Toyota and Woven by Toyota, became President and CEO on Apr 1, 2026, and Toyota's FY2026 Form 20-F (Jun 2026) commits to step up AI use for routine and low-value-added work so staff can focus on work only people can do, with no figure attached.
CTO Hiroki Nakajima told investors in Sep 2026 that from 2028 Toyota plans to spend about ¥1 trillion a year renewing about 60 plants and about 400,000 robots across Toyota, its group companies and suppliers, with robots running Toyota Research Institute's large behavior models, and that a new Toyota Robotics organization would lead the work; Automotive World reported that Toyota had not confirmed the full spending would go ahead.

Where Toyota sits

Stage: Piloting
Toyota sits at Piloting: sanctioned AI tools are live for selected groups, such as a factory AI Platform with nearly 1,200 active users in its 10 Japanese car and unit plants (Dec 2024), O-Beya for about 800 powertrain engineers, and ToyotaGPT at Toyota Motor North America with more than 50 agents in production (Aug 2026). Toyota falls short of Scaling because it has published no AI tool open to most of its 390,927 employees, no group-level AI owner who reports results, and no business return from AI.
How the stages are defined
Ownership: No group AI chief; CTO Hiroki Nakajima, Chief Scientist Gill Pratt (TRI) and regional heads such as Brian Kursar (TMNA)
Platform: Hybrid: Factory AI Platform, O-Beya, ToyotaGPT and vista built in-house on bought cloud, models and data tools
Models: OpenAI GPT-4o via Azure, Microsoft and NVIDIA models, in-house vision and robot models
Hosting: Azure (O-Beya), Google Cloud plus on-premises plant servers (factory AI Platform), Databricks (vista), Woven multi-cloud; NTT data centers planned
Reach: Not disclosed company-wide; ~1,200 factory AI Platform users (Dec 2024), ~800 on O-Beya

How they deploy

Invest in people across the group

Toyota launched the Global AI Accelerator (GAIA) on May 22, 2025 to significantly expand investment in AI research, development, people and implementation in 11 initial categories: automated driving and driver assistance, business software development, customer relations, knowledge retention and transfer, in-vehicle agents, material discovery, manufacturing, novel mobility, office productivity, robotics and vehicle engineering. Toyota bases GAIA on jidoka, automation with a human touch, and paired it with the Toyota Software Academy, about 100 courses in AI, programming, data security and vehicle regulations run with AISIN, DENSO, Toyota Tsusho and Woven by Toyota. A Lean Enterprise Institute analysis in Aug 2026 noted that the program centers on people rather than a flagship system or one vendor.

Build the tools, buy the models and data platform

Toyota builds its AI applications in-house on bought models and platforms. Toyota's Production Digital Transformation Office built a factory AI Platform from early 2022 so plant staff without AI expertise can train their own models, running it on on-premises servers linked to Google Cloud; O-Beya, built by powertrain engineers in Toyota City, runs OpenAI's GPT-4o through Microsoft Azure OpenAI Service; Woven by Toyota built MISRA Copilot with Microsoft; and Toyota Motor Corporation adopted Databricks in Jan 2026 for vista, its unified data and AI hub. Brian Kursar, head of AI at Toyota Motor North America, said in Apr 2025 that Toyota was evaluating whether to build its own generative AI models and had not decided. Toyota's D-ROOM factory AI group showed the in-house habit early: plant staff built a project for under ¥10,000 that an outside vendor had quoted at over ¥100 million (Lean Enterprise Institute, Aug 2026).

AI in the car and on the road

Toyota's Jun 2026 Form 20-F ties AI to software-defined vehicles: AI agents that support drivers and grow with customers, Arene as the platform for safe, reliable vehicle software with full over-the-air updates, and data from the 150 million Toyota vehicles in operation. Kenta Kon said in Jul 2026 that Toyota Safety Sense already includes technology developed for end-to-end, AI-based autonomous driving, and Toyota said in Aug 2026 that it will bring Level 2++ driver assistance, built on in-house end-to-end AI with a rule-based guardrail, to passenger cars from 2028. Woven by Toyota builds one driving stack for assisted and automated driving that learns from fleet data and ships models over the air, NVIDIA said in Jul 2026 that Toyota's Level 2++ vehicles will run on its DRIVE AGX computers and DriveOS, and Toyota is partnering with Waymo on a robotaxi fleet in Japan from 2027 and with NTT on a Mobility AI Platform for road safety.

Robots and research

Toyota established Toyota Research Institute (TRI) in Jan 2016 to accelerate AI research and development, and TRI now works on robots meant to work alongside people rather than replace them. TRI and Boston Dynamics agreed in Oct 2024 to run TRI's large behavior models on the Atlas humanoid; Walden Robotics, a TRI spinout led by Russ Tedrake, formerly TRI's senior vice president for large behavior models, emerged in Jul 2026 with a $300 million round that Toyota co-led; and Toyota Times reported in Oct 2026, from an interview with Gill Pratt, TRI CEO and Toyota's Chief Scientist, that TRI is concentrating on reasoning AI. Hiroki Nakajima told investors in Sep 2026 that Toyota plans to put robots running TRI's large behavior models into its plants and those of its group companies and suppliers from 2028, under a new Toyota Robotics organization led by Chief Production Officer Takefumi Shiga.

AI for routine work, gated by return

Toyota's FY2026 Form 20-F says the company will step up its use of AI for routine tasks and low-value-added work so staff can focus on tasks only people can perform, raising what Toyota calls its rate of value-added work. Toyota Motor North America sets a harder gate: each AI project must clear a six-to-seven-figure annual return threshold, according to its AI engineering team (Aug 2026), and Brian Kursar said in Apr 2025 that projects are judged on cost savings, revenue, customer satisfaction, scalability and total cost of ownership.

Deployment model: Toyota runs AI through its units, with no central AI office on the public record. In Japan, Toyota Motor Corporation's Production Digital Transformation Office runs the factory AI Platform for plant staff; the Digital Transformation Promotion organization chose Databricks for vista and absorbed the plant-floor D-ROOM AI community in 2022, which now feeds an AI Center of Excellence responsible for AI implementation and governance across the company; the Software Development Center builds vehicle software; and engineering groups such as powertrain development build their own tools, such as O-Beya. Toyota Research Institute runs AI and robotics research and, per Brian Kursar's company biography, the Global AI Accelerator program, while a new Toyota Robotics organization under Chief Production Officer Takefumi Shiga leads the rollout of AI-driven factory robots (Sep 2026). Woven by Toyota builds the driving AI, the Arene software platform, Woven City's AI systems and a shared Cloud & AI foundation. Toyota Motor North America has its own Enterprise AI function of about 35 people under Brian Kursar that builds agents for manufacturing, supply chain and R&D (Aug 2026).
Model providers: OpenAI's GPT-4o, reached through Microsoft Azure OpenAI Service, powers O-Beya (Nov 2024). Woven by Toyota's MISRA Copilot started on Azure OpenAI and uses Microsoft's language models (Nov 2025), and NVIDIA said in Jul 2026 that Toyota built a MISRA-compliant code assistant trained with NVIDIA Megatron-LM on data that includes NVIDIA Nemotron. Toyota Motor North America does not name its model providers, but its GearPal diagnostics tool routes through an LLM gateway that falls back to another provider when one goes down (Aug 2026), and Brian Kursar said in Apr 2025 that Toyota was taking a wait-and-see approach to DeepSeek. Woven by Toyota built its own Woven City AI Vision Engine, a vision-language model built with NVIDIA H100 GPUs (Apr and Jul 2026), Toyota Research Institute and its spinout Walden Robotics train large behavior models for robots, and the planned robotaxi fleet will combine Waymo's autonomous driving technology with Toyota vehicles.
Data and infrastructure: Toyota's factory AI Platform runs on a hybrid of on-premises servers in its plants and Google Cloud: an on-premises GKE Enterprise cluster is linked to Google Kubernetes Engine in the cloud, and plant staff train models on NVIDIA T4 and A100 GPUs, so each plant needs one local server instead of its own GPU servers (Dec 2024). Toyota started a Data Library Project in 2021 and adopted Databricks in Jan 2026 to power vista, a hub that unifies corporate, connected and customer data under Unity Catalog permissions; Toyota is building use cases with Databricks Genie and Databricks Apps and plans to adopt Agent Bricks, and Databricks did not name the cloud underneath. O-Beya runs on Microsoft Azure, with Azure Functions linking Azure OpenAI Service to Azure Cosmos DB, which holds vector search, conversation histories and expert reviews of answers. Woven by Toyota's Cloud & AI foundation provides shared workspaces and AI infrastructure across private and public clouds without naming providers. The Mobility AI Platform with NTT plans distributed AI data centers linked by NTT's IOWN optical network in areas rich in renewable energy.

Platforms and internal tools

Factory AI PlatformEarly 2022
Toyota Motor Corporation's AI Platform, built in-house by a small team in the Production Digital Transformation Office, lets plant staff with no AI expertise train image and data models through a web application and deploy them on the line, for tasks such as inspecting parts, checking adhesive on back-door glass and spotting faults in bumper molding machines at the Takaoka plant. The AI Platform was in use at all 10 of Toyota's car and unit plants in Dec 2024, when 10,000 models had been created that year, up from 8,000 in 2023, and more than 400 employees a year took in-house training.
Users
Nearly 1,200 active users (Dec 2024)
Models
Models trained by plant staff on NVIDIA T4 and A100 GPUs
Source
O-BeyaJan 2024
O-Beya, named after Toyota's big-room team practice, is a multi-agent system for Toyota Motor Corporation's powertrain development department: users pick agents such as a Vibration Agent or a Fuel Consumption Agent, and O-Beya merges their answers into one reply grounded in past design reports, current regulations and handwritten notes from veteran engineers. O-Beya had nine agents and was used hundreds of times a month in Nov 2024, and Toyota planned a Consumer Voice Agent to surface the most common complaints about a model.
Users
About 800 powertrain engineers (since Jan 2024)
Models
OpenAI GPT-4o on Microsoft Azure OpenAI Service
Source
ToyotaGPT
ToyotaGPT is Toyota Motor North America's internal AI platform, which answers employees' questions from internal data with access limited to the datasets each user may see and carries more than 50 agents in production. Toyota Motor North America builds the agents with LangChain's Deep Agents, LangGraph and LangSmith and a shared library of skills for manufacturing, supply chain, R&D and branding (Aug 2026).
Users
Employees across departments; no count published
Source
GearPal
GearPal lets Toyota Motor North America plant staff diagnose manufacturing robots and line equipment in natural language, through an LLM gateway that switches providers during outages. Toyota's AI team says a diagnosis that took five to six hours now takes two to three minutes (Aug 2026).
Source
vistaJan 2026
vista is Toyota Motor Corporation's unified data and AI hub on Databricks, joining corporate, connected and customer data with central governance under Unity Catalog so employees can reach data on their own and teams can build and run AI models; Toyota is developing natural-language analysis with Databricks Genie and plans to adopt Agent Bricks for agents.
Source
Woven City AI Vision Engine and Integrated ANZEN SystemApr 2026
Woven by Toyota's AI Vision Engine is a vision-language foundation model that combines camera feeds, mobility and environmental data in Woven City to detect risks in real time; the Integrated ANZEN System pairs it with Behavior AI and Drive Sync Assist to warn drivers and pedestrians. Woven by Toyota plans version 2.0 by the end of 2026 and will test ANZEN in Shizuoka Prefecture government vehicles within fiscal 2026.
Models
In-house vision-language model, built with NVIDIA H100 GPUs
Source
Woven Cloud & AI
Woven by Toyota's Cloud & AI foundation gives automakers, suppliers and partners multi-tenant workspaces to build vehicle software together and a shared environment to prepare data and train, validate and deploy AI, with auto-labeling, automated driving data pipelines and LLM application tools, across private and public clouds.
Source
Arene
Arene is Toyota's software development platform for software-defined vehicles, made up of Arene SDK, Arene Tools for virtual testing and Arene Data for collecting and analyzing vehicle data; Toyota used Arene for the cockpit software of the new RAV4 (Form 20-F, Jun 2026).
Source
MISRA CopilotNov 2025 (proof of concept)
MISRA Copilot is a Woven by Toyota tool built with Microsoft in which three AI agents propose, review and check fixes for coding-standard violations in embedded vehicle software, using retrieval over past fixes. MISRA Copilot resolved 81.5% of violations in a proof of concept on internal driver-assistance code and was under further evaluation for deployment in Nov 2025.
Models
Microsoft language models; Azure OpenAI in early tests
Source
Mobility AI PlatformDevelopment from 2025
The Mobility AI Platform is a joint Toyota and NTT project of three layers: distributed computing in data centers, intelligent communications, and AI that learns from mobility data, aimed at preventing collisions at blind spots and supporting automated driving, with partner rollout from about 2028.
Source

Vendors and partners

MicrosoftBy Jan 2024
Azure OpenAI Service and Azure cloud behind O-Beya; co-developer of MISRA Copilot · O-Beya runs on Azure OpenAI, Azure Functions and Azure Cosmos DB, and Woven by Toyota said in Nov 2025 that it was exploring further work with Microsoft.
Google Cloud2022
Cloud half of the hybrid infrastructure behind the factory AI Platform · Toyota runs the AI Platform on Google Kubernetes Engine and Google Cloud GPU virtual machines linked to on-premises servers in its plants (Dec 2024).
NVIDIAExpanded Jul 2026
Driving computers, AI models and simulation tools · NVIDIA said on Jul 16, 2026 that Toyota's Level 2++ vehicles will use DRIVE AGX and DriveOS, that Toyota built a code assistant with Megatron-LM, that Toyota uses Omniverse and Isaac Sim for digital twins of production lines, and that Woven by Toyota built its vision-language model with H100 GPUs.
DatabricksJan 2026
Data and AI platform under vista · Toyota Motor Corporation chose Databricks to power vista; Kento Izumi said Databricks was the best option to meet Toyota's requirements.
NTTOct 2024
Co-developer of the Mobility AI Platform · Toyota and NTT plan a combined ¥500 billion investment by 2030, with NTT contributing communications and IOWN-linked data centers.
WaymoApr 2025
Autonomous driving partner · A preliminary agreement in Apr 2025 became a plan in Oct 2026 for an autonomous ride-hailing fleet in Japan from 2027, with Woven by Toyota exploring road safety work drawing on Waymo's operations.
Walden RoboticsJan 2026
Toyota Research Institute spinout building robots on large behavior models · Toyota co-led Walden's $300 million seed round in Jul 2026, and Walden's robots have worked in production at a Toyota plant in North America since Feb 2026.
Boston DynamicsOct 2024
Robotics research partner of Toyota Research Institute · TRI's large behavior models run on the Atlas humanoid robot.
LangChain
Agent frameworks (Deep Agents, LangGraph, LangSmith) at Toyota Motor North America · LangChain published the ToyotaGPT case study in Aug 2026.
AISIN, DENSO and Toyota TsushoMay 2025
Group companies running the Toyota Software Academy with Toyota and Woven by Toyota · The academy adopted DENSO's SOMRIE skill certification as the group standard.

Leadership

AI, data and platform leaders

Gill Pratt

Chief Scientist and Executive Fellow for Research, Toyota; CEO, Toyota Research Institute

Gill Pratt became the founding CEO of Toyota Research Institute in Jan 2016 and guides research strategy for Toyota and the Toyota Group as Chief Scientist; Pratt previously ran robotics research programs at DARPA. Pratt said in May 2025 that GAIA aims to put people at the heart of technological change, and Toyota Times reported from an Oct 2026 interview with Pratt that TRI is concentrating on reasoning AI.

Keiji Yamamoto

Chief Officer, Digital Information and Communications Group, Toyota Motor Corporation

Keiji Yamamoto joined Toyota in 1987, led the Connected Company from 2016, and in Jan 2021 became Chief Information & Security Officer and Chief Software Officer and took charge of the Digital Transformation Promotion Department. Yamamoto took part in the Oct 2024 announcement of the Mobility AI Platform with NTT as a Senior Fellow, and Toyota's Jul 2026 executive structure lists him as chief officer of the Digital Information and Communications Group, home of its cybersecurity team.

Akihiro Sarada

President, Software Development Center, Toyota Motor Corporation

Akihiro Sarada, an engineer who led development of the latest Crown range, became president of Toyota's Software Development Center when it was formed in Oct 2023 to bring together software work spread across other units (Reuters). Sarada said in Aug 2026 that Toyota will develop Level 4 driving technology but offer it commercially as Level 2++ (The Next Web).

Hajime Kumabe

CEO, Woven by Toyota

Hajime Kumabe became Representative Director and CEO of Woven by Toyota on Oct 1, 2023, moving from CEO of J-QuAD DYNAMICS. Kumabe leads the unit that builds Toyota's driving AI, the Arene platform and Woven City.

John Absmeier

Chief Technology Officer, Woven by Toyota

John Absmeier has been CTO of Woven by Toyota since before Oct 2023 and sits on its board.

Jack Yan

Vice President and Head of Cloud & AI, Woven by Toyota

Jack Yan is responsible for Woven by Toyota's global technology infrastructure, the shared cloud and AI foundation for vehicle software and AI development, and sits on the Toyota Technology Steering Committee, which drives digital transformation at Toyota Motor Corporation. Yan joined TRI-AD, Woven's predecessor, in 2019 after serving as CIO of Nomura IT Shanghai.

Dushyant Wadivkar

Vice President, Head of Global AD/ADAS, Woven by Toyota

Dushyant Wadivkar leads automated driving and driver assistance at Woven by Toyota and spoke on machine learning in autonomy at KAKEZAN 2026 in Apr 2026.

Jean-François Campeau

Vice President and Head of Arene, Woven by Toyota

Jean-François Campeau leads Arene, Toyota's software development platform for software-defined vehicles, and wrote in Jun 2026 on the responsibility behind each over-the-air update.

Brian Kursar

Group Vice President, Head of AI, Toyota Motor North America

Brian Kursar joined Toyota in 2002, left for Warner Bros in 2015, returned in 2018 as chief data scientist of Toyota Connected North America, and was CTO and chief data officer before becoming Toyota Motor North America's first head of Enterprise AI at the end of 2024. Kursar leads AI strategy, governance, delivery and adoption for the region, advises TRI's Global AI Accelerator and co-leads in-vehicle AI agent development.

Kento Izumi

Head of the Digital Transformation Promotion Office, Toyota Motor Corporation

Kento Izumi spoke for Toyota Motor Corporation when it adopted Databricks for its vista data and AI hub in Jan 2026, saying Toyota needed a platform that enables company-wide AI use.

Executives

Kenta Kon

President and Chief Executive Officer

Kenta Kon joined Toyota in 1991, became chief officer of the Accounting Group in 2019 and CFO in 2020, served as CFO of Woven by Toyota from 2023, took charge of the Mobility 3.0 Office in Apr 2025 and the CFO role again in Jul 2025, and became President and CEO on Apr 1, 2026. Kon said in Jul 2026 that AI and robots will inevitably play a major role in Toyota's plants and that in some cases AI will do the teaching.

Akio Toyoda

Chairman of the Board

Akio Toyoda joined Toyota in 1984, was President from 2009 to 2023 and has chaired the board since 2023. Toyoda is the Master Weaver of Woven City and contributed to the Akio Toyoda AI, a model of his approach to leadership that Toyota said in Apr 2026 was in development to promote AI adoption across the Toyota Group.

Koji Sato

Vice Chairman and Chief Industry Officer

Koji Sato joined Toyota in 1992, was President and CEO from Apr 2023 to Mar 2026 and became Vice Chairman and Chief Industry Officer on Apr 1, 2026, a role focused on the wider auto industry; Sato chairs the Japan Automobile Manufacturers Association and took part in the joint announcement of the Mobility AI Platform with NTT in Oct 2024.

Hiroki Nakajima

Executive Vice President and Chief Technology Officer, Member of the Board

Hiroki Nakajima joined Toyota in 1987, led its Mid-size Vehicle and CV companies, and has been Executive Vice President and Chief Technology Officer since Apr 2023. Nakajima spoke for Toyota when it announced the Waymo partnership in Apr 2025 and told investors in Sep 2026 that Toyota plans about 400,000 robots running large behavior models from 2028, saying Toyota aims for robots that coexist with people rather than replace them.

Yoichi Miyazaki

Executive Vice President and Chief Financial Officer

Yoichi Miyazaki joined Toyota in 1986, became Executive Vice President and CFO in Apr 2023, and took the CFO role again on Apr 1, 2026, when Kenta Kon became CEO.

Takefumi Shiga

Operating Officer and Chief Production Officer; leader of Toyota Robotics

Takefumi Shiga joined Toyota in 1992, was an Executive Vice President of Toyota Motor North America from 2023, became Operating Officer and Chief Production Officer in Apr 2026 and chief officer of the new Production Engineering Group in Jul 2026. Hiroki Nakajima told investors in Sep 2026 that Shiga leads Toyota Robotics, a new organization set up to bring AI-driven robots into Toyota's plants (Car Watch).

Tetsuo Ogawa

Operating Officer; President and CEO, Toyota Motor North America

Tetsuo Ogawa joined Toyota in 1984, has been chief executive of Toyota's North America Region since Apr 2020 and President and CEO of Toyota Motor North America since Jun 2023. Ogawa heads the regional company in which Brian Kursar's Enterprise AI group builds ToyotaGPT and its agents.

AI org map

No central AI office; AI built in research subsidiaries, a software subsidiary, Japanese digital, production and engineering teams, and regional AI groups

Toyota's Chief Technology Officer is Executive Vice President Hiroki Nakajima, and Toyota names no chief AI, data or information officer among its executives. Toyota Research Institute, led by Chief Scientist Gill Pratt, runs AI research and the Global AI Accelerator; Woven by Toyota, led by Hajime Kumabe, builds driving AI, Arene and Woven City's AI; Toyota Motor Corporation's digital organizations run the factory AI Platform, the vista data hub and an AI Center of Excellence; Toyota Robotics, under Chief Production Officer Takefumi Shiga, leads factory robots; and Toyota Motor North America has its own Enterprise AI group under Brian Kursar. Kenta Kon, President and CEO since Apr 1, 2026, sits above all of them.

Toyota does not publish a count of its AI workforce. Toyota's AI work sits in at least five places: Toyota Research Institute in the United States, Woven by Toyota in Tokyo, the digital, production and engineering organizations of Toyota Motor Corporation in Japan (where a team of six developers built the factory AI Platform), the new Toyota Robotics organization for factory robots, and regional teams such as Toyota Motor North America's Enterprise AI group of about 35 people (Aug 2026). The Toyota Software Academy, with about 100 courses, trains AI and software staff across five group companies, and Toyota's D-ROOM community of plant staff building their own AI grew to about 500 members across 18 sites before joining the digital transformation organization in 2022.

Reporting lines

President and CEO
Kenta Kon
- Chief Technology Officer
  Hiroki Nakajima, Executive Vice President
  Presented the plan for AI-driven factory robots to investors in Sep 2026
- Software Development Center
  Akihiro Sarada, President
  In-house company formed in Oct 2023 to consolidate software work; reporting line not published
- Digital Information and Communications Group
  Keiji Yamamoto, chief officer
  Houses Toyota's cybersecurity team (Form 20-F); reporting line not published
  - Cyber Security Office
    Its chief officer reports material risks to the board through the Chief Risk Officer (Form 20-F)
- Digital Transformation Promotion
  Kento Izumi, head of the DX Promotion Office
  Runs vista on Databricks; reporting line not published
  - AI Center of Excellence
    Manabu Okuyama, group manager
    Grew out of D-ROOM, which joined the Digital Transformation Promotion Division in 2022; line inferred
- Production Digital Transformation Office
  Part of the Production Group; built and runs the factory AI Platform; line inferred
- Toyota Robotics
  Takefumi Shiga, Chief Production Officer
  Set up by Sep 2026 to bring AI-driven robots into plants; reporting line not published
- Toyota Research Institute
  Gill Pratt, CEO and Toyota Chief Scientist
  Founded Jan 2016 for AI research; line inferred from ownership
  - Global AI Accelerator (GAIA)
    Leader not published; Brian Kursar advises on infrastructure, data sharing and integration
- Woven by Toyota
  Hajime Kumabe, CEO
  Wholly owned subsidiary; line inferred from ownership
  - Woven Cloud & AI
    Jack Yan, VP
  - Woven AD/ADAS
    Dushyant Wadivkar, VP
  - Arene
    Jean-François Campeau, VP
- Toyota Motor North America
  Tetsuo Ogawa, President and CEO
  Regional subsidiary
  - TMNA Enterprise AI
    Brian Kursar, Group VP and Head of AI
    About 35 people (Aug 2026)

Toyota's executive page, checked Nov 1, 2026, names a President, two Executive Vice Presidents and five other operating officers, with Gill Pratt among the Executive Fellows, and Toyota's Jul 1, 2026 executive structure names Hiroki Nakajima as Chief Technology Officer and Takefumi Shiga as Chief Production Officer but no chief AI, data, digital or information officer. The lines from Toyota Research Institute, Woven by Toyota, Toyota Robotics and the digital organizations to the President are inferred from ownership and are not published, and Toyota has not said who leads the Global AI Accelerator.

Where the AI workforce sits

Research: Toyota Research Institute
TRI runs AI, robotics and human-centered AI research; no headcount published.
Not disclosed
Software and driving AI: Woven by Toyota
Woven by Toyota builds the driving stack, Arene, Cloud & AI and Woven City's AI systems.
Not disclosed
Toyota Motor Corporation digital, production, engineering and plant teams
Includes the factory AI Platform team, the AI Center of Excellence, Toyota Robotics and engineering groups such as the powertrain team behind O-Beya.
Not disclosed
Regional enterprise AI teams
Toyota Motor North America's Enterprise AI group is about 35 people (Aug 2026).
Not disclosed

Named people

Group leadership

Kenta Kon
President and CEO (since Apr 1, 2026)
Hiroki Nakajima
Executive Vice President and CTO
Yoichi Miyazaki
Executive Vice President and CFO
Takefumi Shiga
Chief Production Officer; leader of Toyota Robotics

Toyota Research Institute

TRI runs AI, robotics and human-centered AI research from Los Altos and Cambridge; Russ Tedrake, who led its large behavior model research, now runs Walden Robotics, a TRI spinout.

Gill Pratt
CEO; Toyota Chief Scientist

Woven by Toyota

Hajime Kumabe
CEO
John Absmeier
CTO
Jack Yan
VP and Head of Cloud & AI
Dushyant Wadivkar
VP, Head of Global AD/ADAS
Jean-François Campeau
VP and Head of Arene

Toyota Motor Corporation digital and engineering AI

Keiji Yamamoto
Chief Officer, Digital Information and Communications Group
Akihiro Sarada
President, Software Development Center
Kento Izumi
Head of the Digital Transformation Promotion Office
Manabu Okuyama
Group manager, AI Center of Excellence · Co-founded D-ROOM at the Miyoshi Plant in 2018 (Lean Enterprise Institute, Aug 2026)
Kenji Onishi
AI project lead, powertrain development · Leads the O-Beya generative AI project (Nov 2024)

Toyota Motor North America Enterprise AI

About 35 people (Aug 2026)

Brian Kursar
Group Vice President, Head of AI
Kordel France
Director of AI Engineering
Ravi Chandu Ummadisetti
Director of Agentic AI and Product Research
Dave Tsai
CTO, Toyota Connected North America (named Dec 2024) · Previously vice president of Enterprise AI at Toyota Motor North America; current role not confirmed

Use cases

Toyota runs AI in production in plant quality and maintenance (a factory AI Platform in all 10 of its car and unit plants in Japan, and Walden Robotics' robots in a North American plant), engineering knowledge (O-Beya for powertrain engineers), North American plants and offices (ToyotaGPT, GearPal and other agents, more than 50 in production by Aug 2026), software testing and in-vehicle safety systems, with pilots in Woven City, robotics and code compliance and announced plans with Waymo, NTT and about 400,000 AI-driven factory robots. Toyota has published no business-level return for any AI use case and no firm-level return from AI; the figures on record are task speedups, an hours-saved figure, projections and investment thresholds.

Use case	Area	Status
Factory AI Platform for plant-built models Plant staff train their own inspection and anomaly-detection models on Toyota's in-house AI Platform, in use at all 10 car and unit plants in Japan with nearly 1,200 active users; Toyota's AI Group wrote in Dec 2024 that the platform had cut more than 10,000 hours of manufacturing work a year.	Manufacturing (Japan)	In production
O-Beya engineering knowledge agents O-Beya lets Toyota powertrain engineers query agents on vibration, fuel consumption, engines and regulations and get one combined answer drawn from design reports, regulations and veteran engineers' notes.	Powertrain engineering	In production
ToyotaGPT enterprise assistant and agents ToyotaGPT answers employees' questions from internal data, with access limited to each user's permitted datasets, and hosts more than 50 agents in production across departments.	Toyota Motor North America	In production
GearPal equipment diagnostics GearPal lets plant staff diagnose robots and line equipment in natural language; Toyota's AI team says diagnosis time fell from five to six hours to two to three minutes.	Manufacturing (North America)	In production
Knowledge bots for plant workers Toyota Motor North America built AI knowledge tools from repair manuals, trouble tickets and video documentation for manufacturing workers, which Brian Kursar described as in use in Apr 2025.	Manufacturing (North America)	In production
Walden Robotics robots on the line Robots from Walden Robotics, a Toyota Research Institute spinout that learns tasks through large behavior models, have loaded and unloaded parts, cleaned machinery and kitted parts in production at a Toyota plant in North America since Feb 2026.	Manufacturing (North America)	In production
R&D GPT R&D GPT is an agent that searches Toyota Motor North America's internal research on paint, corrosion, seat materials and leathers.	R&D (North America)	In production
AI-generated unit tests Toyota Motor North America developers use AI to generate unit tests; Brian Kursar cited a productivity gain of at least 20% from an internal study (Apr 2025).	Software development	In production
AI in Toyota Safety Sense and the driving stack Toyota Safety Sense, Toyota's driver assistance suite, includes technology developed for end-to-end, AI-based autonomous driving, Kenta Kon said in Jul 2026.	Vehicles	In production
MISRA Copilot MISRA Copilot uses three AI agents to propose fixes for coding-standard violations in embedded vehicle software as pull requests that engineers accept or reject; the proof of concept resolved 81.5% of violations.	Software quality	Pilot
D-ROOM plant-built AI Quality control staff Masaya Furutani and Manabu Okuyama began building factory-floor AI models at the Miyoshi Plant in 2018 on Raspberry Pi computers and USB cameras, and the D-ROOM community spread to 18 sites before joining Toyota's digital transformation organization in 2022.	Plant quality control	Pilot
vista natural-language analysis Toyota Motor Corporation is building use cases on vista with Databricks Genie for natural-language data analysis and Databricks Apps, and plans agents with Agent Bricks.	Data and analytics	Pilot
AI Vision Engine and Integrated ANZEN System Woven by Toyota's vision-language model watches Woven City through cameras and sensors to flag risks and warn drivers and pedestrians, and UCC Japan uses it in a proof of concept.	Woven City	Pilot
Large behavior models for humanoid robots Toyota Research Institute's large behavior models control Boston Dynamics' Atlas humanoid in joint research, and a single model drove Atlas through a sequence of humanlike movements in 2025.	Robotics	Pilot
ELEY robot that learns from workers ELEY, Toyota's two-armed robot on a wheeled base, learns tasks by watching workers who wear jigs shaped like its hands, and folded T-shirts with near-perfect accuracy after 1,500 practice sessions; Toyota plans to share centrally learned data across its factories.	Manufacturing (Japan)	Pilot
Akio Toyoda AI The Akio Toyoda AI models Chairman Akio Toyoda's approach to leadership and decisions; Toyota said in Apr 2026 that it is in development to promote AI adoption across the Toyota Group.	Group-wide	Announced
AI-driven robots across Toyota, group and supplier plants Toyota plans to renew about 60 plants and about 400,000 robots across Toyota, its group companies and suppliers from 2028, with robots running Toyota Research Institute's large behavior models, led by the new Toyota Robotics organization.	Manufacturing (group-wide)	Announced
Level 2++ driver assistance on end-to-end AI Toyota plans to bring Level 2++ driver assistance to passenger cars from 2028, with in-house end-to-end AI making driving decisions and a rule-based guardrail blocking unexpected behavior, while the driver stays responsible.	Vehicles	Announced
AI for aerodynamic design Toyota plans to use AI on its next-generation battery electric vehicles to secure aerodynamic performance and double the time designers spend on styling (Form 20-F, Jun 2026).	Vehicle design	Announced
In-vehicle AI agents Toyota plans AI agents that support drivers and grow alongside customers in its software-defined vehicles, and Brian Kursar co-leads in-vehicle agent development.	Vehicles	Announced
AI for routine and low-value-added tasks Toyota's FY2026 Form 20-F commits to step up AI use for routine tasks so employees can focus on work only people can do, as part of company-wide measures to lower its break-even volume.	Office work	Announced
Robotaxi fleet with Waymo Toyota and Waymo will begin initial operations of an autonomous ride-hailing fleet in Japan in 2027 on a purpose-built Toyota bZ4X, with commercial service to follow.	Mobility services	Announced
Mobility AI Platform with NTT Toyota and NTT are building a platform that learns from mobility data to prevent collisions at blind intersections and support automated driving, with partner rollout planned from about 2028.	Road safety	Announced

Adoption

Toyota spread AI from the edges rather than through one rollout: Toyota Research Institute from 2016, plant staff building their own models through D-ROOM from 2018 and on the factory AI Platform from 2022, which had nearly 1,200 active users across 10 plants in Japan by Dec 2024, O-Beya for about 800 powertrain engineers from Jan 2024, and an Enterprise AI group at Toyota Motor North America that had more than 50 agents in production by Aug 2026. Toyota added group-wide AI training through the Toyota Software Academy and the Global AI Accelerator in May 2025 and a unified data and AI hub, vista, in Jan 2026. Toyota has published no count of employees who use AI weekly or have access to a general AI assistant.

By the numbers

Mobility AI Platform investment with NTT: ¥500B; Combined Toyota and NTT investment by 2030, announced Oct 31, 2024

AI agents in production at Toyota Motor North America: 50+; On ToyotaGPT, per Toyota's AI engineering team in a LangChain case study, Aug 2026

Active users of the factory AI Platform: ~1,200; In all 10 of Toyota's car and unit plants in Japan, per Toyota's AI Group on Google Cloud's blog, Dec 2024

Powertrain engineers with O-Beya access: ~800; Since Jan 2024; nine AI agents on Azure OpenAI (Microsoft, Nov 2024)

R&D expenditure (FY2026): ¥1.52T; Year to Mar 31, 2026, up from ¥1.33T; no separate AI budget disclosed (Form 20-F)

Employees: 390,927; Consolidated, Mar 31, 2026 (Form 20-F)

Rollout, first pilot to today

Jan 2016
Toyota established Toyota Research Institute to accelerate research and development of artificial intelligence.
Form 20-F
2018
Quality control staff at Toyota's Miyoshi Plant started D-ROOM to build AI for the factory floor, piloting three AI models in 2019.
Lean Enterprise Institute
2021
Toyota launched its Data Library Project to give staff timely access to data.
Databricks
2022
Toyota's Production Digital Transformation Office began building an in-house AI Platform for plant staff on on-premises servers and Google Cloud.
Google Cloud
2023
Toyota Motor North America formed a central Enterprise AI group with Toyota Connected North America, which co-leads a Responsible AI steering board.
Toyota Motor North America
Oct 2023
Hajime Kumabe became CEO of Woven by Toyota, Toyota's software and automated driving subsidiary.
Woven by Toyota
Jan 2024
Toyota opened O-Beya, a multi-agent system on Azure OpenAI, to about 800 powertrain engineers.
Microsoft
Oct 2024
Toyota Research Institute and Boston Dynamics agreed to put TRI's large behavior models on the Atlas humanoid.
Toyota Research Institute
Oct 2024
Toyota and NTT agreed to invest a combined ¥500 billion by 2030 in a Mobility AI Platform.
Toyota
Dec 2024
Toyota's factory AI Platform was in use at all 10 of its car and unit plants in Japan, with nearly 1,200 active users.
Google Cloud
Dec 2024
Toyota Motor North America named Brian Kursar its first head of AI from the end of 2024.
Automotive World
May 2025
Toyota launched the Global AI Accelerator across 11 areas and, with four group companies, the Toyota Software Academy of about 100 courses.
Toyota
Sep 2025
Woven City, Toyota's test town for AI and mobility, opened on Sep 25, 2025.
Form 20-F
Jan 2026
Toyota Motor Corporation adopted Databricks to power vista, its unified data and AI hub.
Databricks
Feb 2026
Robots from Walden Robotics, a Toyota Research Institute spinout, started production work at a Toyota plant in North America.
The Next Web
Apr 2026
Toyota and Woven by Toyota unveiled the Woven City AI Vision Engine and the Integrated ANZEN System and said an Akio Toyoda AI was in development.
Woven by Toyota
Aug 2026
Toyota Motor North America's AI team reported more than 50 agents in production on ToyotaGPT.
LangChain
Sep 2026
Toyota said it had set up Toyota Robotics, led by Chief Production Officer Takefumi Shiga, to bring AI-driven robots into its plants.
Car Watch

Economics & controls

Spend and ROI

R&D budget

Toyota spent ¥1,522.9 billion on research and development in the year to Mar 31, 2026, up about 15% from ¥1,326.5 billion a year earlier and ¥1,202.4 billion in FY2024, according to its Form 20-F. Toyota does not report a separate technology or AI budget.

What AI itself costs

Toyota has not disclosed what it spends on AI. Toyota and NTT plan a combined ¥500 billion by 2030 for the Mobility AI Platform (Oct 2024), and Toyota said in May 2025 that GAIA would significantly expand its AI research and development investment without giving an amount. Toyota Research Institute's venture funds have committed $100 million in 2017, $100 million in 2019, $150 million in 2021 and $150 million in Apr 2024 to startups in AI, robotics, autonomous mobility and data and cloud, Woven Capital has $800 million for growth-stage companies in autonomous driving, AI and smart cities, and Toyota co-led a $300 million round for Walden Robotics in Jul 2026. Separately, Hiroki Nakajima told investors in Sep 2026 that Toyota plans about ¥1 trillion a year from 2028 to renew about 60 plants and about 400,000 robots running large behavior models across Toyota, its group companies and suppliers; Automotive World reported that Toyota had not confirmed the full investment would go ahead, and Toyota has not said how much of it is AI.

Reported ROI

Toyota has reported no business-level return from AI. Toyota's AI Group wrote on Google Cloud's blog in Dec 2024 that the factory AI Platform had cut more than 10,000 hours of manufacturing work a year, a labor-hour figure Toyota has not converted into cost or headcount and that the same post also gives as a projection. Toyota Motor North America's AI team said in Aug 2026 that its manufacturing use cases are on track for at least six figures of annual savings per line, shop or plant and projected multimillion-dollar savings per facility, but those are projections in a vendor case study. Woven by Toyota projected that MISRA Copilot could save several hundred million yen if deployed (Nov 2025), also a projection from a proof of concept.

Productivity markers

Toyota's productivity figures for AI are task-level: Brian Kursar cited at least a 20% productivity gain from AI-generated unit tests (Apr 2025), Toyota Motor North America says building an agent now takes one engineer four days instead of six engineers six months (Aug 2026), and Toyota's Form 20-F expects AI in aerodynamic design to double the time designers spend on styling. Toyota's company-wide aim, set in Jun 2026, is to raise the rate of value-added work by using AI for routine tasks.

Headcount

Toyota had 390,927 consolidated employees at Mar 31, 2026, up from 383,853 a year earlier, and Hino Motors left consolidation on Apr 1, 2026. Kenta Kon described Toyota's future plants in Jul 2026 as people-centered, with AI and robots playing a major role, and said Toyota has not yet decided which processes will move to robots; in Aug 2026 Kon added that the value of what only people can do will increase. Toyota has announced no AI-related job cuts.

Governance and risk

People at the center

Toyota frames its AI program around jidoka, automation with a human touch: GAIA rests on that principle, Gill Pratt said in May 2025 that Toyota wants to put people at the heart of technological change, and Toyota Research Institute runs a human-centered AI division whose Human-AI Kaizen Initiative with four universities studies how people and AI systems grow together (Sep 2026).

Data access and privacy

Toyota's vista hub manages permissions and metadata centrally through Databricks Unity Catalog, and ToyotaGPT limits each user's answers to the datasets that user may see. Woven by Toyota says its fleet data pipelines enforce strict privacy and data-handling governance and that the Woven City Data Fabric supports data use while respecting individual preferences and privacy.

People approve AI output

Woven by Toyota's MISRA Copilot never changes code by itself: fixes arrive as pull requests that engineers accept or reject (Nov 2025). O-Beya stores human expert reviews of its answers to improve them, and Brian Kursar, who leads AI governance at Toyota Motor North America, said in Apr 2025 that guardrails must ensure the quality of AI output can always be measured.

Return thresholds

Toyota Motor North America requires each AI project to clear a six-to-seven-figure annual return threshold (Aug 2026) and weighs projects on cost savings, revenue, customer satisfaction, scalability and total cost of ownership, Brian Kursar said in Apr 2025.

Cybersecurity

Toyota Motor Corporation's cybersecurity team sits in its digital information and communication group under the chief officer of the cyber security office, who sits on a governance and risk subcommittee with board members and reports material risks to the board through the Chief Risk Officer. Toyota's Form 20-F (Jun 2026) says no material cybersecurity incident has occurred and that subsidiaries and Japanese dealers self-inspect against more than 100 items in the All Toyota Security Guidelines.

Vehicle AI rules

Toyota's Form 20-F notes that the UN's WP.29 vehicle regulation forum endorsed considerations on AI in road vehicles in Jun 2024 and is exploring instruments for AI used in vehicles. Woven by Toyota's driving stack pairs its AI decision layer with a separate safety framework that enforces validated constraints based on local driving rules and Toyota's safety principles, and Toyota said in Aug 2026 that its Level 2++ system will add a rule-based guardrail that blocks unexpected AI behavior against criteria people set in advance.

Recent signals

Oct 30, 2026
Toyota and Waymo moved to the next phase of their collaboration, planning initial operations of an autonomous ride-hailing fleet in Japan in 2027 on a purpose-built Toyota bZ4X.
Woven by Toyota
Oct 21, 2026
Toyota Times reported, from an interview with Gill Pratt, that Toyota Research Institute is concentrating on reasoning AI that can work through situations with few precedents.
Toyota Times
Sep 25, 2026
Woven City marked its first year with about 30 resident-centered trials, AI Robot Association household robots in 11 homes, and AI Vision Engine 2.0 planned by the end of 2026.
Woven by Toyota
Sep 15, 2026
Toyota Research Institute selected four university projects for its Human-AI Kaizen Initiative from 194 proposals.
Toyota Research Institute
Sep 2026
At an investor briefing at Toyota Motor Europe, Hiroki Nakajima said Toyota plans about ¥1 trillion a year from 2028 to renew about 60 plants and about 400,000 robots running large behavior models, and that a new Toyota Robotics organization led by Takefumi Shiga will lead the rollout.
Car Watch
Aug 27, 2026
Toyota said it will bring Level 2++ driver assistance built on in-house end-to-end AI, with a rule-based guardrail, to passenger cars from 2028.
Car Watch
Aug 24, 2026
Toyota Motor North America's AI team said ToyotaGPT runs more than 50 agents in production and that each AI project must clear a six-to-seven-figure annual return threshold.
LangChain
Aug 13, 2026
Kenta Kon said Toyota will use AI and more robots in production and that the value of what only people can do will increase, as Toyota forecast its operating margin falling to 6.3%.
Kyodo via Bernama
Jul 27, 2026
Toyota Times published an interview in which Kenta Kon said AI and robots will inevitably play a major role in Toyota's plants and that Toyota Safety Sense includes technology developed for end-to-end, AI-based autonomous driving.
Toyota Times
Jul 16, 2026
NVIDIA expanded its partnership with Toyota: DRIVE AGX and DriveOS for Toyota's Level 2++ vehicles, a code assistant built on Megatron-LM, Omniverse and Isaac Sim for factory digital twins, and H100 GPUs for Woven by Toyota's vision-language model.
Car Watch
Jul 15, 2026
Walden Robotics, a Toyota Research Institute spinout run by Russ Tedrake, emerged with a $300 million round co-led by Toyota and robots already working in a Toyota plant in North America.
The Next Web
Jun 10, 2026
Toyota's FY2026 Form 20-F committed to step up AI use for routine and low-value-added work and reported R&D spending of ¥1.52 trillion.
Form 20-F
Apr 22, 2026
Toyota and Woven by Toyota unveiled the Woven City AI Vision Engine and the Integrated ANZEN System and said an Akio Toyoda AI meant to promote AI adoption across the Toyota Group was in development.
Woven by Toyota
Feb 6, 2026
Toyota named CFO Kenta Kon President and CEO from Apr 1, 2026, with Koji Sato becoming Vice Chairman and Chief Industry Officer.
Toyota
Jan 8, 2026
Toyota Motor Corporation adopted Databricks to power vista, a unified data and AI hub, and plans to add Databricks Agent Bricks.
Databricks

Quick answers

How does Toyota deploy AI?: Toyota deploys AI through its business units and research arms rather than one company-wide platform. Plant staff in Japan build their own models on an in-house factory AI Platform running on Google Cloud and on-premises servers, powertrain engineers use O-Beya agents on Microsoft Azure OpenAI, and Toyota Motor North America runs more than 50 agents in production. Toyota Research Institute and Woven by Toyota build the robot and driving AI, and CTO Hiroki Nakajima told investors in Sep 2026 that Toyota plans about 400,000 AI-driven factory robots from 2028.
What does Toyota use AI for?: Toyota runs AI in production in plant quality and maintenance (a factory AI Platform in all 10 of its car and unit plants in Japan, and Walden Robotics' robots in a North American plant), engineering knowledge (O-Beya for powertrain engineers), North American plants and offices (ToyotaGPT, GearPal and other agents, more than 50 in production by Aug 2026), software testing and in-vehicle safety systems, with pilots in Woven City, robotics and code compliance and announced plans with Waymo, NTT and about 400,000 AI-driven factory robots. Toyota has published no business-level return for any AI use case and no firm-level return from AI; the figures on record are task speedups, an hours-saved figure, projections and investment thresholds.
Who leads AI at Toyota?: AI at Toyota is led by Gill Pratt (Chief Scientist and Executive Fellow for Research, Toyota; CEO, Toyota Research Institute); Keiji Yamamoto (Chief Officer, Digital Information and Communications Group, Toyota Motor Corporation); Akihiro Sarada (President, Software Development Center, Toyota Motor Corporation). Ownership of AI on the org chart: No group AI chief; CTO Hiroki Nakajima, Chief Scientist Gill Pratt (TRI) and regional heads such as Brian Kursar (TMNA).
What AI platform does Toyota use?: Toyota runs no single AI platform: an in-house factory AI Platform serves plant staff in Japan, O-Beya serves powertrain engineers, ToyotaGPT serves Toyota Motor North America, and vista on Databricks became the unified data and AI hub in Jan 2026.
Which AI models does Toyota use?: O-Beya runs OpenAI's GPT-4o on Microsoft Azure OpenAI Service, Toyota built a code assistant on NVIDIA Megatron-LM, Woven by Toyota builds its own vision-language model and driving stack, and Toyota Research Institute trains large behavior models for robots.
How many Toyota employees use AI?: Toyota has published no company-wide AI user count; its factory AI Platform had nearly 1,200 active users across 10 plants in Japan (Dec 2024), about 800 powertrain engineers have had O-Beya since Jan 2024, and Toyota Motor North America runs more than 50 agents (Aug 2026).
How far has AI spread at Toyota?: Toyota sits at Piloting: sanctioned AI tools are live for selected groups, such as a factory AI Platform with nearly 1,200 active users in its 10 Japanese car and unit plants (Dec 2024), O-Beya for about 800 powertrain engineers, and ToyotaGPT at Toyota Motor North America with more than 50 agents in production (Aug 2026). Toyota falls short of Scaling because it has published no AI tool open to most of its 390,927 employees, no group-level AI owner who reports results, and no business return from AI.

Open questions

Does Toyota give its 390,927 employees a company-wide generative AI assistant, and how many use AI each week?
Who leads the Global AI Accelerator, how much does it invest, and who owns AI for the Toyota Group as a whole?
What does Toyota spend on AI each year, separate from its ¥1.52 trillion R&D budget, the ¥500 billion Mobility AI Platform plan with NTT and the ¥1 trillion a year it plans for plants and robots from 2028?
What business return do the factory AI Platform, O-Beya, ToyotaGPT, GearPal and the D-ROOM plant models deliver, given that Toyota has published only task speedups, an hours-saved figure and projections?
Which clouds host ToyotaGPT, vista and Woven by Toyota's Cloud & AI foundation, and which model providers sit behind Toyota Motor North America's LLM gateway?
How many people has the Toyota Software Academy trained since May 2025, and how many AI and machine learning specialists does Toyota employ?
Has Woven by Toyota deployed MISRA Copilot or the NVIDIA-based code assistant, and did the projected savings of several hundred million yen materialize?
Which models will power Toyota's in-vehicle AI agents, and will they run in the car, in Toyota's cloud or in the NTT data centers?
Who leads large behavior model research at Toyota Research Institute since Russ Tedrake left to run Walden Robotics, and how will robots from TRI, Walden and other suppliers share the 400,000-robot plan?
How will Toyota measure the rate of value-added work it expects from using AI for routine tasks?
Will Toyota's own driving stack and Waymo's technology both reach personally owned Toyota vehicles, and on what timeline?

Sources

Toyota

Woven by Toyota and Toyota Research Institute

Filings

Toyota Form 20-F for the year to Mar 31, 2026 (filed Jun 10, 2026)

Partners

News and analysis

Data confidence

Confirmed: Stated in a primary or reputable public source.
Estimated: Derived or approximated from public reporting.
Unconfirmed: Reported, but not yet corroborated.

Toyota profile · Enterprise AI diffusion report by Imagine · facts checked against their sources on Nov 1, 2026

More company profiles

All profiles