Mitsubishi Electric and Inria Commence Joint Technology Development to Ensure AI Trustworthiness Using Formal Methods
Will deliver next-generation AI technology and help to realize a society where AI can be used with peace of mind
With the rapid development and global spread of AI, companies that develop and provide AI systems are required to manage risks appropriately. In particular, systems that require a high level of safety, such as infrastructure systems and cybersecurity systems, can cause significant damage in the event of any malfunction, making it extremely important to guarantee the reliability of AI output. In theory, AI reliability assessment requires exhaustive testing, thoroughly verifying output results for all possible inputs. However, it is a challenging task that takes an enormous amount of time and incurs huge cost.
Mitsubishi Electric’s European subsidiary, Mitsubishi Electric R&D Centre Europe, and Inria have been collaborating through joint research on advanced verification technologies utilizing Formal Methods since 2015. This cutting-edge initiative is aimed at developing a technology that theoretically verifies the reliability of AI output by combining Formal Methods technologies with AI technologies such as large-scale language models, leveraging the expertise of both parties and building on their long-standing track record and trust-based relationship. The goal is to further expand the scope of AI application towards systems that require high levels of trustworthiness.
For the full text, please visit: www.MitsubishiElectric.com/news/
View source version on businesswire.com: https://www.businesswire.com/news/home/20250917231505/en/
prd.gnews@nk.MitsubishiElectric.co.jp
nathalie.lacaux@inria.fr
Customer Inquiries
Mitsubishi Electric R&D Centre Europe
1 Allée de Beaulieu, 35700
info@fr.merce.mee.com
nathalie.lacaux@inria.fr
Source: