Thales speeds up its development of AI for defence

Thales has announced the creation of cortAIx, which brings together the company’s AI capabilities in research, sensors and systems. cortAIx is the AI accelerator that will equip armed forces, aircraft manufacturers and all critical infrastructure operators with highly secure solutions to provide them with more efficient data analysis and decision support, whilst taking into account specific constraints such as cybersecurity, embeddability and frugality that are related to critical environments.

With over 600 AI experts, around 100 doctoral students in AI and a top-tier network of industry, start-up and academic partners, Thales has been a major player in trusted, transparent, explainable and ethical AI for the past decade. The company is Europe’s top patent applicant in the field of AI for critical applications, and has incorporated AI into over a hundred of its products and services.

The defence sector is undergoing major changes related to the proliferation of data on the battlefield. And it is taking advantage of new usage in the civilian world, such as personal assistants, video analysis and noise reduction.

On the back of the investments it has made in the field of AI, Thales has been incorporating AI into its critical systems for the last ten years, in order to enable its customers to gain an edge and to become more resilient in the face of a deluge of data, threats and information.

Thales Chairman and CEO Patrice Caine, said “Our company, which is a European leader in the fields of civil and defence high-tech, is able to draw on its mastery of AI and on the assets that are its strength: its technological expertise, its knowledge of critical domains – defence and aerospace, space, cybersecurity, digital identity – and its management of constrained environments. Thales is already developing embeddable, frugal, trusted, explainable and secure end-to-end solutions, and today it is moving to the next level by acquiring major AI capabilities, that meet the security and sovereignty needs of our customers.”

cortAIx, the trusted AI accelerator for research, sensors and systems

On the occasion of fourth edition of the Thales Media Day, structured around AI, Thales is introducing cortAIx, the AI accelerator designed to expand the integration of AI technology into all of the company’s sectors of activity (defence, space, aerospace, cybersecurity and digital identity). cortAIx brings together:

• cortAIx Lab – the most powerful integrated lab for critical AI in Europe, based in Saclay, at the heart of the European innovation ecosystem.
• cortAIx Factory – Thales’s AI technology factory, designed to speed up the classification and industrialisation of AI development tools and use cases for system data. Thales already equips its systems with AI, and continues to identify new use cases to increase performance, such as, for instance, mission planning, air traffic management, and steering drones and robots. AI enables situations to be analysed and processed quicker and systems to be used more intuitively, in order to speed up decision-making and reduce operators’ mental workload.
• cortAIx Sensors – this represents all the AI capabilities in the field of sensors within the company’s defence businesses. The Thales sensors (sonar, radar, radio and optronics) that incorporate AI offer hugely increased capabilities – both more precise and more efficient – in terms of sighting and identifying threats and targets, and they meet frugality requirements in terms of size, weight, power, as well as being resilient in extreme environments. 

This trusted AI serves its end-users, who remain the final decision-makers at all times. 

Cybersecure AI 

In the face of the security threats of AI, Thales, which is a pioneer in the field, has developed specific assessment methods at its Toulouse-based CESTI laboratory (Centre d’Évaluation de la Sécurité des Technologies de l’Information – Centre for the Assessment of the Security of Information Technology), which is certified by ANSSI (the French Agency for Information System Security). It draws on the expertise of more than 5,800 cybersecurity experts across the company to provide tailor-made solutions for the vulnerabilities that have been detected.

Trusted AI in Thales’s critical systems 

Sonar, radar, equipment on board maritime patrol aircraft and fighter jets… Thales’ solutions use the best sensor and system technologies, which cover the whole range of intelligence needs in the land, aerospace and space domains. Trusted AI, which is incorporated into these sensor technologies, lightens operators’ workloads and speeds up the detection, identification and classification of objects of interest and target scenes:

• The Talios pod is designed for air reconnaissance and targeting missions, and is to be found on board Dassault Aviation’s Rafale fighter jet. Until now, the images collected by the pod were analysed on the ground, but thanks to the Thales Neural Processor, the optronics images captured in flight will be analysed in real time, and 100 times quicker, by embedded AI.
• With its Friendly Hacking unit, Thales provides its development teams with the ability – hitherto unseen in the industry – to submit their AI-based solutions to a cyber crash-test, with the aim of strengthening considerably the cybersecurity of systems that implement AI.
• AI on board maritime patrol aircraft’s AirMaster surveillance radars enables ever-greater quantities of data to be analysed, and targets to be classified in a few tenths of a second.
• Thales’ air defence radars perform remarkably well in detection at very low altitude. AI will provide significant help to operators in identifying small targets.
• AI in air traffic control systems will enable the optimisation of the sequencing of aircraft on the approach paths to airports, thus contributing to a reduction in their carbon footprint.
• AI will make the coordination of multi-drone and multi-robot systems easier. The controlled and secure autonomy of these systems will significantly reduce pressure on operators.
• Using an innovative human-machine interface coupled with AI, helicopter pilots will be able to not only prepare tactical missions much quicker, but also to change them in flight, which had until now been impossible. Pilots will remain in charge of the choice of trajectory depending on the mission objective, whilst minimising their cognitive workload.
• AI in radios for armed forces improves sound quality in noisy environments. It has been designed to be embedded in a miniature chip that ensures that the autonomy of equipped devices is maintained.