The mission of the IAEA, the International Atomic Energy Agency, is to work with its member states and many partners around the world to promote safe, secure and peaceful nuclear technologies. Convinced of the ability of artificial intelligence in the nuclear field to help solve some of humanity’s great challenges, she presented in an article seven areas where she is already improving the peaceful uses of nuclear technology.
The genesis of the IAEA dates back to the “Atoms for Peace” speech delivered by President Dwight D. Eisenhower to the United Nations General Assembly on December 8, 1953. In it, he proposed the creation of the International Atomic Energy Agency. , whose status was approved by 81 countries in October 1956, and on July 29, 1957, the agency was officially born. As of March last year, it had 175 member countries, its headquarters are in Vienna, Austria, and since December 2019 it has been led by Rafael Mariano Grossi.
The IAEA also has four offices: 2 regional offices located in Toronto (Canada) and Tokyo (Japan) and two liaison offices in New York. The agency operates laboratories specializing in nuclear technology in Vienna and Seibersdorf (Austria) and, since 2002, a laboratory in Monaco to investigate the effects of radioactivity on the marine environment.
In 2005, the IAEA and its Director General, Mohamed ElBaradei, received the Nobel Peace Prize. for their efforts to prevent nuclear energy from being used for military purposes and to ensure that nuclear energy for peaceful purposes is used in the safest possible way”. Subsequently, Mr. Mohamed ElBaradei was honored for his tolerance, his humanity and his freedom in many countries India, Austria, Italy, Bolivia…)
In addition, the agency has created several specialized programs, particularly on cancer treatment, nuclear safety and security, innovative nuclear reactors and fuel cycles.
AI and human health
In the field of health, the purpose of the IAEA is to help Member States establish high-quality health care. The use of nuclear techniques in medicine and nutrition has become one of the most widespread peaceful uses of nuclear energy.
The IAEA has collaborated with the WHO for six decades in the fight against cancer. It has also established the “Human Health Programme” to help member states use nuclear techniques to prevent, diagnose and treat cancer and cardiovascular disease. In the context of prevention, she is particularly concerned with nutrition and food quality.
As a peaceful application in the field of health, she mentions the fight against cancer, and especially its screening. In fact, the IAEA launched the “Beams of Hope” initiative last February with the aim of supporting the establishment and expansion of radiotherapy services, especially in developing countries that do not have them. It has also just entered into a partnership in this context with 11 global cancer companies.
On the other hand, since 2005 it has conducted more than 100 imPACT reviews to analyze the capacity and needs of national health systems in cancer prevention and control. The last one took place last November in Colombia.
In terms of cancer screening, last June it launched a coordinated research project aimed at studying the potential of AI to improve contouring skills in radiotherapy, with a particular focus on increasing the accuracy of the delineation of organs at risk in the head and throat cancer.
As the Covid-19 pandemic raged, the IAEA launched the Integrated Zoonose Action (ZODIAC) initiative in June 2020 to help countries prevent outbreaks caused by bacteria, viruses, fungi or parasites from animals that can be transmitted to humans.
AI for food and agriculture
Farmers are facing new problems: degradation of the environment, soil, climate change… certain start-ups like ITK are supporting farmers in this transition to agro-ecology thanks to the modeling of living things (plants and animals) to achieve optimal productivity.
Combined with nuclear technologies, AI can provide solutions to fight hunger and malnutrition, improve environmental sustainability and ensure food security. The IAEA and the Food and Agriculture Organization of the United Nations (FAO) work in partnership to support these goals. 30 coordinated research projects are being carried out in this area, involving more than 400 research institutions and experimental stations in the Member States.
AI, water and the environment
AI can analyze lots of isotope data (isotopes are atoms that have the same number of electrons and therefore protons to remain neutral, but a different number of neutrons). Among these data we find those from GNIP, the Global Network for the Measurement of Hydrogen and Oxygen Isotopes in Precipitation (including oxygen-18 and deuterium), created in 1960 by the IAEA and the World Meteorological Organization (WMO), which works in collaboration with many institutions in the Member States.
The analysis of this data allows scientists to better understand climate change and its impact on the environment, especially the availability of fresh water.
AI, nuclear science and fusion research
In nuclear science, AI is used for data analysis, theoretical modeling, and simulations, promoting basic research and technological innovation.
In particular, it plays an important role in nuclear fusion research, at the heart of the work of researchers around the world, especially at CEA in France, because it has the potential to provide clean and renewable electricity and plasma physics.
Last June, the IAEA invited its interested partners to participate in a coordinated research project to accelerate fusion R&D using artificial intelligence through the creation of a platform and a network between communities for innovation and partnership.
AI and nuclear power
In light of climate change, the need to make greater use of low-carbon energy sources has become obvious.
Nuclear power is one of the solutions, it is the second source of clean energy in the world. It currently accounts for around 11% of global electricity generation and emits virtually no greenhouse gases or air pollutants.
AI is widely used in this sector, ML thus makes it possible to automate tasks, ensure the reliability of processes and detect irregularities. For their part, AI systems and simulations speed up the creation and optimization processes, reducing maintenance costs.
AI, nuclear safety and radiation protection
The IAEA works to promote a strong and sustainable global framework for nuclear safety and security in member states, one of its roles is to protect people and the environment from the harmful effects of ionizing radiation.
It has launched the “IAEA Nuclear Safety Action Plan”, approved by member states in September 2011, to strengthen the global nuclear safety framework, following the March 2011 accident at TEPCO’s Fukushima Daiichi nuclear power plant.
AI has the potential to improve nuclear safety in several ways. It can be used to process data from radiation detection systems to refine the detection and identification of nuclear and other radioactive materials, to analyze data from physical protection systems to improve the detection of intrusions and irregularities that may result from cyber attacks on nuclear facility.
In the radiation protection sector, software related to safety standards that integrate AI have the capacity to strengthen the protection of workers exposed to occupational exposure.
As part of France Relance, the government has launchedcall for projects to support investment and modernization of the nuclear industry and selected the ARDNA project (AI Research on Data for Nuclear Application) in late 2021. This project aims to establish an AI-augmented control system and was led by the company Aquila Data Enabler, in collaboration with Andra (National Agency for Radioactive Waste Management) and the startup Spotlight-Earth. This technology may relate to seismic monitoring of Cigéo repository structures, the geological repository project for the most radioactive waste.
AI and IAEA Safeguards
Through a series of technical measures, called “safeguards”, the IAEA verifies that states comply with their international legal obligations and use nuclear materials and technology only for peaceful purposes, preventing the proliferation of nuclear weapons.
To detect misuse of nuclear materials or techniques at an early stage, it uses data from satellite images, environmental samples, gamma spectroscopy and video surveillance. It uses ML to identify abnormal points in these large amounts of data, help verify spent fuel and analyze monitoring records. AI also improves the enforcement of safety measures by saving inspectors from a lot of repetitive tasks