After observing success and failure in other organizations, ITER has developed and deployed a first set of AI-based tools that deliver real value to a large user base. Near the end of 2023, the IT Project Tools Section within the Central Integration Division began to look at how it might apply artificial intelligence (AI) to improve ITER systems engineering processes. One use case that very quickly rose to the top of the list was facilitating information retrieval from the ITER Document Management (IDM) system which, some 20 years after it was first deployed, is widely used across the project.While the IDM system has been generally well received, a recurring issue for users has been the difficulty in finding documents among approximately 1.5 million stored in the system. “We had made several attempts in the past to change the way we do searches on IDM, but that proved very challenging,” says Jean-Daniel Delaplagne, Section Leader for the IT Project Tools Section.After a little brainstorming, Delaplagne’s team concluded that AI could solve the problem. They built a proof of concept (PoC) in Q1 2024—and improved it in an iterative fashion, through testing, until it was ready for the first production release in Q3 2024. In response to user feedback on the first version, a second major version was deployed in early January 2025, including a fundamental shift in architecture from standalone to multi-agent.The new set of tools uses AI to summarize each document within the IDM system and the results are used to index all the information in a big vector database. The team developed an application that uses natural language processing to send queries to the vector database and return one or more responses using a popular technique called retrieval-augmented generation (RAG), which instead of just matching keywords like traditional search engines, allows users to ask questions and find semantically meaningful information. ITER's new chatbot that uses natural language processing to send queries to the vector database and return one or more responses using a technique called retrieval-augmented generation which, instead of just matching keywords like traditional search engines, allows users to ask questions and find semantically meaningful information. A good illustration of the way vector databases work is to imagine a document management system with information on cats, dogs, and birds. If a user asks a question about “animals” using a traditional keyword search, no matches are found. But with a vector database, the query would return all documents about cats, dogs, birds—and any other member of the semantic category “animal.”“For the core AI backend capabilities, we decided to use the latest AI models from OpenAI,” says Junmin Yang, AI Project Lead in the IT Project Tools Section. “Users can ask questions in all the different languages supported by OpenAI—including Mandarin, Korean, Japanese, Russian and Hindi. But the answers are always in English to standardize the output and facilitate performance monitoring as English is the official language in the organization.”The AI tool has been used by about 120 partners—ITER and Domestic Agency employees, but also contractors with access to IDM. According to Yang, nearly a thousand people have used it so far, posting a total of 20,000 queries. The license fees paid to OpenAI average approximately €200 per month for that level of activity. Delaplagne says that since the first production release, users have sent in about 600 pieces of feedback, providing guidance to the team on how to tune and improve the RAG* model and other parts of the system. Approximately 1.5 million documents are stored in the ITER Document Management (IDM) system; now, there is an easy way for ITER users (including Domestic Agency colleagues and contractors) to find what they are looking for. Some of the comments after the first release were about the hundreds of acronyms used inside ITER. In response, the team created a specific agent inside the chatbot dedicated to answering questions on this topic, which was included in the January release. This solution was a fundamental shift to a more agent-based technical stack, which has implications beyond deciphering acryonyms. “Now we can create specific agents for different queries and get quicker results,” says Delaplagne. “We are now developing agents tailored to fetch information from sources outside of the IDM system to satisfy requests we’ve received from users to extend the search to other databases. For example, we recently added a new agent to answer questions about HOPs [hand over packages, which are sets of documents handed off from construction to commissioning].”“We’re also planning to integrate the chatbot into applications people use at work on a daily basis,” says Yang. “This includes products in the Microsoft suite, like Teams and Microsoft Copilot.”To allow users to make local choices about the large language models (LLMs) they use, IT plans to interface with other LLMs, such as Llama, DeepSeek and Mistral. According to Delaplagne, the new tools will also be made available to other organizations. “The IDM system is used in other fusion institutions, and we are already in contact with IPP [The Max Planck Institute for Plasma Physics] to deploy it there,” he explains.The IT Project Tools Section is proud of having managed a project that went so quickly from a PoC to a production system used by a thousand people. “A lot of organizations have tried implementing chatbots, but not many have been able to go live and have it used by so many people on a day-to-day basis,” says Delaplagne. “It was a challenging project that we built with industrial partners like Microsoft, and we had good interns to do the coding to tailor the tools to our needs. It has been satisfying to deliver a tool that makes a real difference for IDM users.”* RAG (retrieval-augmented generation) is a technique for enhancing the accuracy and reliability of generative AI models with information retrieved from external data sources.

The United States Domestic Agency, US ITER, has completed delivery of all components for the support structure of the central solenoid—the 60-foot-tall superconducting magnet that is the “heart” of the ITER machine. Comprised of more than 9,000 individual parts, the support structure was manufactured by eight US suppliers across six states. With the final delivery crossing the Atlantic in January, this milestone was the culmination of more than a decade of work by US ITER staff and US suppliers. Contributors included Kamatics in Connecticut; Major Tool & Machine in Indiana; Keller Technology in New York; Hamill Manufacturing, Precision Custom Components, and Superbolt in Pennsylvania; Petersen in Utah; and Robatel Technologies in Virginia.“Designing and manufacturing the first-of-a kind superconducting central solenoid is a major engineering challenge,” says David Vandergriff, a senior project engineer who has been part of the central solenoid team since joining US ITER in 2016. “But this magnet can’t deliver for ITER without the support structure.”The support structure can be described as an exoskeleton, or a cage, that surrounds the central solenoid at the center of the ITER machine. Composed of a stack of six individual magnet modules, each weighing 121 tonnes, the central solenoid induces the majority of the magnetic flux change needed to initiate the plasma, generate the plasma current, and maintain this current during the burn time. The last shipment of components of the support structure for the central solenoid delivery was prepared in Pennsylvania during January 2025. Credit: US ITER “The first role of the structure is to hold the six central solenoid modules in position within strict tolerances measured in millimetres,” says Kevin Freudenberg, US ITER’s engineering technical director. “Then the real challenge occurs during operation. At key times, the resultant vertical force on the module stack is up to 60 meganewtons—more than twice the force of a space rocket at blast off.”    To achieve a structure that can withstand the levels of extreme force that the central solenoid will generate, the team had to overcome numerous challenges. One example was the development of the 27 connectors, called tie plates, that form the vertical bars of the support structure.  The tie plates connect the lower key blocks on the solenoid’s foundation platform to the upper key blocks at the top of the structure, forming a cage around the central solenoid.“When we began the initial design work, there was significant concern over whether we could make each tie plate as a single piece,” Freudenberg said.  â€œThese pieces are 50 feet long and must be incredibly straight, meeting strict tolerances.  We worked with forging shops and specialized engineering companies to achieve this, changing our initial design from a two-piece concept to a single piece.” Eight companies across the United States fabricated components for the support structure, including Superbolt in Pennsylvania, whose technology is key to holding the structure together against extreme forces. Credit: Superbolt Another challenge was fastening the structure components together sufficiently to withstand the extreme force that the central solenoid will generate. This led the team to turn to a US company that had developed what is known as Superbolt® technology.With four of the central solenoid modules now in place, and the final two expected to be stacked this year, the team is looking forward to the next phase of construction of the support cage at ITER.“The real reward for the team will be watching the structure be assembled around the central solenoid, seeing their more than a decade of work coming to fruition and knowing we are one step closer to ITER operations,” said Freudenberg. 

On 30-31 January, the ITER In-Vessel Assembly Project conducted a comprehensive review of the assembly and installation processes for key in-vessel components, including the in-vessel coils and the blanket and divertor systems. The initiative aimed to review the installation activities by optimizing the sequence, tools, and logic underpinning these critical installation activities, which are projected to last a little over two years.  The integration of a large number of complex components in the constrained environment of the vacuum vessel will require careful planning and execution. The review in late January, chaired by Doug Loesser from the Princeton Plasma Physics Laboratory (PPPL), was organized to carefully examine the proposed assembly sequences, to ensure logical interdependencies and efficient workflows. Specialized tools—such as the in-vessel tower crane, the blanket assembly transporter, the divertor assembly transporter and other bespoke handling devices—underwent detailed assessments to confirm their suitability for the assembly process, with potential refinements discussed to maximize efficiency. Strategies were identified to enhance installation operations while also addressing potential bottlenecks and constraints.A key aspect of the review meeting was to align the installation schedule with overall project timelines. Identifying potential misalignments and mitigating technical challenges through rigorous risk assessment are both crucial aspects to ensuring that the schedule remains realistic and achievable and preventing delays and cost overruns.  Participants to the assembly and installation review organized in January for key in-vessel components. Since 2015, the tools needed for in-vessel assembly have been under development and testing. Several handling facilities, including a Trial, Test, and Training Facility, will soon be available for use on site, providing a valuable resource for process qualification and personnel training. Lessons learned from recent trials indicate that some upgrades are necessary to enhance functionality, particularly for the blanket assembly transporter, which will play a crucial role in the installation strategy we are planning. Unlike the tower crane, which operates on rails in the lower part of the vessel and will be used to install the divertor, the blanket assembly transporter is port-based, allowing for the concurrent installation of blanket and divertor components, significantly improving efficiency.Several refinements have also been proposed to simplify divertor installation, which are being organized through the Domestic Agencies that are providing divertor components.ITER intends to collaborate with an industrial partner for the complete development of those assembly and installation processes and for the execution of onsite works.By integrating lessons from previous trials, refining tool functionality, and fostering collaboration among stakeholders, the in-vessel assembly strategy is evolving into a more efficient, cost-conscious, and schedule-aligned approach. 

From 13 April to 13 October 2025, the World EXPO will open its gates in Osaka, Japan. As part of the theme “Designing Future Society for Our Lives,” 158 countries and organizations will showcase—in spectacular fashion—their cultural heritage and forward-thinking projects.As part of the EXPO2025, ITER will have a prominent presence in the International Organizations Pavilion, where we will present the promise of fusion energy and the global collaboration that drives our pursuit of this transformative energy source.Half-way through the six-month exhibition, on 13 July, ITER will host a special FUSION DAY in the National Day Hall with presentations, animations and video screenings.If you are interested to visit the ITER exhibit or if you would like to attend the Special Fusion Day, please write to EXPO2025@iter.org.We very much look forward to seeing you in Osaka.