Today, Microsoft announced a major AI infrastructure and skilling investment supported by a new partnership on security. It will help the UK seize the artificial intelligence (AI) opportunity and ensure that AI innovation and safety progress together while creating jobs, improving services, and protecting public security. Microsoft is committing to more than doubling its datacenter footprint in the UK, training more than one million people for the AI economy and supporting the UK’s growing AI safety and research efforts through partnerships with the government and leading universities.
The investment will cover three key areas:
Capacity: First, Microsoft will invest in the UK’s AI infrastructure. Over the next three years Microsoft will spend £2.5 billion ($3.2bn) to expand its next generation AI datacenter infrastructure, bringing more than 20,000 of the most advanced GPUs to the UK by 2026. The single largest investment in its 40-year history in the country, Microsoft will grow its datacenter footprint across sites in London and Cardiff and potential expansion into northern England.
This infrastructure investment will help to meet the exploding demand for efficient, scalable and sustainable AI specific compute power and the needs of the private and public sector waiting to take advantage of the latest cloud and AI breakthroughs.
To support research on AI, Microsoft will extend its Accelerating Foundation Models Research (AFMR) program to include prioritized access to GPUs for the UK’s science and research community. AFMR drives interdisciplinary research on AI alignment and safety, beneficial applications of AI, and AI-driven scientific discovery in the natural and life sciences. This new UK effort will aim at harnessing the power of AI to accelerate scientific discovery via multiscale multimodal data generation through prioritized access to Microsoft’s AI tools. This program includes researchers from the UK’s world leading participating universities including Cambridge, Oxford, Imperial College, UCL, Bath, and Nottingham.
Capability: Second, Microsoft will invest in broad-based AI talent and education programs. To support UK workers across the AI economy, Microsoft will make a multi-million-pound investment to train one million people with the skills they need to build and work with AI. This will include expanded training for people looking to start, or move into, a career in AI. Working in partnership with multiple learning and non-profit partners, the program will focus on building AI fluency, developing AI technical skills, supporting AI business transformation, and promoting safe and responsible AI development and use including the first Professional Certificate on Generative AI.
As part of this skilling commitment, Microsoft will also turn all the lessons it has learned in operationalizing responsible AI principles for its own AI engineers and developers, into learning modules for UK customers and partners. This training will help the UK’s AI developer ecosystem to embed safety and security measures into their own systems and processes and takes Microsoft’s training support for responsible AI beyond principles and embeds sound practice into the way AI is developed and integrated across the IT industry.
Finally, to help ensure Microsoft technical trainers are adhering to the ethics and principles of developing AI solutions responsibly they will all complete and attest to Microsoft’s “Responsible Generative AI” training.
Security: Third, Microsoft will invest in strong AI safety and security measures. These will cover both Microsoft’s own infrastructure and support for AI developers and customers deploying and using AI applications. Microsoft will operate its AI services and infrastructure in accordance with industry-leading responsible AI practices. It will integrate the adoption and use of responsible AI principles into its Partner Pledge for its 25,000 UK partners and it will collaborate with the UK Government and AI Safety Institute on the ongoing development of refinements and improvements in this field.
Tags: AI, data center, Responsible AI, United Kingdom