The UK’s competition watchdog has paved the way for a formal investigation into the partnership between Microsoft and the ChatGPT developer OpenAI by asking for comments on the arrangement.
The Competition and Markets Authority made the announcement on Friday after a bout of leadership and boardroom turmoil at OpenAI, an artificial intelligence (AI) company based in San Francisco.
The company was established as a non-profit entity whose board controls a commercial unit, in which Microsoft is the biggest investor.
The CMA said “recent developments” had prompted the organisation to review whether the partnership had resulted in “an acquisition of control”. Last month, OpenAI’s board fired and then reappointed its chief executive, Sam Altman, and announced the formation of a new board. Microsoft now has a non-voting observer seat on the OpenAI board.
“What is interesting here is that the CMA is referencing ‘recent developments’ in the Microsoft/Open AI partnership – presumably the fallout from the recent Sam Altman affair,” said Alex Haffner, a partner at Fladgate, a London law firm. “Essentially the CMA wants to find out what changes were made to the partnership and, in the wider context of CMA/UK government considering the regulation of the AI sector, whether there is anything worth delving further into.”
The Microsoft chief executive, Satya Nadella, said in the wake of the 17 November firing that his company had not been consulted on the decision. Speaking shortly before an agreement was announced for Altman’s reinstatement, Nadella added: “There is no OpenAI without, sort of, Microsoft leaning in, in a deep way, to partner with this company on their mission.”
Microsoft, a $2.8tn (£2.2tn) company, has invested $13bn in OpenAI.
Within days of Altman’s firing, Nadella had also announced that Altman and OpenAI’s ousted chair, Greg Brockman, had been hired to lead a new artificial intelligence unit at Microsoft, an arrangement that never came to pass as Altman negotiated his return.
The CMA said it wanted to review whether the partnership had resulted in an acquisition of control, whether a de facto merger had taken place and if this could have an impact on competition. It said the invitation to comment was in advance of “any launch of a formal phase one investigation”.
A phase one investigation would investigate whether Microsoft’s partnership with Open AI creates any competition concerns, specifically any risk of a substantial lessening of competition, said Fladgate’s Haffner.
The regulator, run by lawyer-turned-watchdog Sarah Cardell, said the partnership, which involves multi-year, multibillion-dollar investment and collaboration in technology development, represented a close, multi-faceted relationship between companies with significant activities in AI foundation models.
The move suggests the CMA is gearing up for a second battle with the tech company, after initially blocking Microsoft’s $69bn deal to buy Activision Blizzard, the maker of games including Call of Duty, only relenting after squeezing concessions from it. Cardell’s intervention in April provoked a war of words with Microsoft, with its president, Brad Smith, calling it the “darkest day in our four decades in Britain”
Foundation models are the technology that underpins chatbots such as ChatGPT, and OpenAI’s models are deployed by Microsoft in its products including its Bing search engine. As part of Microsoft’s investment, the company has given OpenAI access to the computing power it needs to train and operate its models.
The CMA is monitoring the area of foundation models closely for potential competition or consumer protection issues. The OpenAI partnership strategy has been replicated elsewhere with Anthropic, the developer of the Claude chatbot, receiving investment from Amazon and Google.
In September the CMA gave notice of its interest in OpenAI-style investments in a report in which it said would monitor closely the impact of investments by big tech firms in AI developers. It added that it was “essential” that the AI market did not fall into the hands of a small number of companies.
The watchdog said it would review any comments it received from interested parties and could launch an investigation into the partnership as a result if it felt it was necessary.
Sorcha O’Carroll, the senior director for mergers at the CMA, said: “The invitation to comment is the first part of the CMA’s information-gathering process and comes in advance of launching any phase one investigation, which would only happen once the CMA has received the information it needs from the partnership parties.”
Smith said in a statement on Friday that the OpenAI partnership, which began in 2019, had fostered “more AI innovation and competition” while preserving independence for both companies.
He added: “The only thing that has changed is that Microsoft will now have a non-voting observer on OpenAI’s board, which is very different from an acquisition such as Google’s purchase of DeepMind in the UK.”
An OpenAI spokesperson said the partnership allowed the company to develop “safe and beneficial AI tools for everyone” while remaining independent, adding that Microsoft’s board observer “does not provide them with governing authority or control over OpenAI’s operations”.
