.webp)
Microsoft and OpenAI support UK AISI project to ensure safe, reliable AI and promote trustworthy artificial intelligence innovation.
Microsoft has joined forces with OpenAI and the UK's AI Security Institute (AISI) to back global initiatives related to advanced AI alignment research and to stimulate innovations in trustworthy AI technology. The information was released by UK Deputy Prime Minister David Lammy and AI Minister Kanishka Narayan at the AI Impact Summit in India. As a result of this partnership, £27 million has been allocated through the Alignment Project fund to guarantee that AI technologies are used for their intended purposes and do not generate unintended consequences.
The contributions that Microsoft makes will bolster those made by OpenAI through the pledge of £5.6 million. The focus of this project is to address key challenges relating to the prediction of AI behaviors, the development of reliable and comprehensive safety frameworks for AI systems, and the creation of machine learning models that are aligned with human values. The first round of funding has already been distributed across a total of 60 projects in eight countries. The second round of funding will open this summer.
This program demonstrates Microsoft's ongoing commitment to mitigating corporate AI risks and emphasizes collaboration across sectors to create greater acceptance of AI by society. Through its support of standards for ethical governance of AI systems, and next-generation control methods for AI systems, Microsoft's partners share in the goal of developing safe and reliable AI technologies that will provide economic benefit for us all.
The AISI project also includes participation from governments, universities, and tech companies around the world. The coalition embodies the understanding that research on alignment and safety of AI systems must advance at least as quickly as AI technology is advancing. With Microsoft's engagement, the AISI project intends to further AI's positive impact on public services, healthcare delivery, and overall enterprise performance and ensures that innovations stemming from trustworthy AI systems will be advantageous for society, with reduced risk factors.
.webp)
.webp)
