Mastercard has announced a partnership with the US Department of Commerce to help create the US AI Safety Institute Consortium (AISIC). In essence, AISIC aims to create new AI standards focused on trustworthiness.

The consortium will comprise AI developers, users, and researchers from Fortune 500 companies, academic institutions, nonprofits, and other US government agencies. The primary objective of AISIC will be to conduct research and offer guidance on AI safety issues, informing future standards and policies. Members will facilitate collaboration between the Commerce Department's National Institute of Standards and Technology and the broader AI safety community to establish best practices, share insights, and track new developments.

Regarding these new developments, representatives from Mastercard emphasised the importance of building trust in AI technology, stating that establishing common and meaningful standards is important for protecting users and promoting inclusive innovation. They also highlighted the significance of the public-private partnership facilitated by AISIC in advancing responsible AI practices. According to the official press release, Mastercard aims to underscore its commitment to respecting individual rights and societal values in technology development.

The company has developed a set of principles for data and tech responsibility and aims to collaborate similarly within AISIC, sharing its experiences with partners, regulators, and policymakers. What else has been going on with Mastercard in the field of AI? In February 2024, Mastercard developed its own proprietary generative artificial intelligence model designed to enhance banks' capabilities in detecting and preventing fraudulent transactions. The newly introduced AI model, known as Decision Intelligence Pro, enables banks to improve their assessment of suspicious transactions on Mastercard's network in real time, distinguishing between legitimate and fraudulent activities.

According to company officials, this advanced AI solution is based on a proprietary recurrent neural network, a fundamental component of generative AI, developed entirely by Mastercard's cybersecurity and anti-fraud teams. Mastercard's algorithm is trained using data sourced from approximately 125 billion transactions processed through the company's card network annually. This data aids the AI in comprehending the relationships between merchants, focusing on transaction patterns rather than textual inputs, such as those utilized by large language models such as OpenAI's GPT-4 and Google's Gemini.

This approach allows Mastercard's AI to predict and identify fraudulent transactions more effectively. The payments company revealed that it invested more than USD 7 billion in cybersecurity and AI technologies since 2019. That includes a number of acquisitions, including its March 2023 deal to buy Swedish cybersecurity firm Baffin Bay Networks.

.