The UK government has announced a new strategic partnership with OpenAI that could lead the company to “expand AI security research collaborations, explore investing in UK AI infrastructure like data centers, and find new ways for taxpayer funded services” to use AI. The move follows the introduction of the AI Action Plan in January, which fast-tracks the construction of data centers in certain regions of the UK.
In the (entirely voluntary) partnership agreement — technically a Memorandum of Understanding — OpenAI and the Department for Science, Innovation and Technology (DSIT) agree to tackle positive-sounding, but ultimately vague tasks things like finding ways for “advanced AI models” to be used in both the public and private sectors and sharing information around the security risks of AI. OpenAI is also supposed to help DSIT identify ways it can deliver on the infrastructure goals of the AI Action Plan, and possibly explore building in one of the UK’s new data center-friendly “AI Growth Zones.”
All of this sounds nebulous and non-committal because the memorandum OpenAI signed is not at all legally-binding. The partnership sounds nice for elected officials eager to prove the UK is competing in AI, but it doesn’t tie anyone down, including the UK government: If Anthropic offers a deal on Claude, they can take it.
OpenAI already has offices in London, so deepening its investment doesn’t seem out of the question. Signing the memorandum is also consistent with OpenAI’s growing interest in working with governments desperate for the high-tech gloss of the AI industry. The logic follows that if OpenAI can get regulators dependent on its tools — say, a ChatGPT Gov specifically designed for government agencies — they’ll be more inclined to favor the company in policy decisions. Or at the very least, making a show of collaborating early could win the company a sweeter deal down the road.
