Why is the little ol’ Competition & Markets Authority, a UK regulator, inserting itself into the entertaining and important – but distant – drama at San Francisco-based OpenAI? Even if the CMA finds eventually that Microsoft, another US company, is pulling the strings at Sam Altman’s show, what could it actually do? Doesn’t it all paint the UK as an unfriendly place for tech investment, notwithstanding Rishi Sunak’s eagerness to host AI summits and conduct cosy chats with Elon Musk?
All fair questions, and the CMA should brace for more in that vein. It is indeed slightly odd that the UK regulator is the first out of traps in wondering, albeit in a preliminary manner, if Microsoft has gained effective control over OpenAI and, if it has, whether that amounts to a problem. But there is another way to look at developments: thank goodness a regulator somewhere is seeking clarity about what just occurred at OpenAI.
The company is the leading player in an artificial intelligence market that is clearly going to be enormous. If the current Big Tech crew have the whole thing sewn up before the game has started in earnest, that is surely suitable territory for regulatory inquiry. Watchdogs globally were asleep at the wheel a generation ago when the leading social media companies erected fortresses around their businesses and then acquired any upstart that looked remotely threatening. Best not to repeat the mistake.
And, actually, this is probably not a case of the CMA going out on a limb. The US regulator the Federal Trade Commission was fretting in the summer about whether AI firms’ “open first, closed later” tactics – which may be what’s happened at OpenAI – could “undermine long-term competition”. Note, too, that German regulators explored Microsoft’s influence over OpenAI this year and reserved the right to look again should it increase.
Last month’s upheaval at OpenAI saw Altman fired by the controlling not-for-profit board and then rehired within days. The governance arrangements under a new and expanded board therefore look to be crucial to OpenAI’s true degree of independence from Microsoft, which had already invested $13bn (about £10.3bn) in OpenAI’s profit-seeking unit.
“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 [in 2014],” argued Brad Smith, Microsoft vice-chair and president, on Friday, playing it all down.
Well, maybe, but that wasn’t quite what Satya Nadella, Microsoft’s own chief executive, was saying at the height of the turmoil. “There is no OpenAI without, sort of, Microsoft leaning in, in a deep way, to partner with this company on their mission,” he said a couple of weeks ago, emphasising the computing and cloud capabilities that his firm brings.
The CMA’s move is minor at this stage. The watchdog is merely seeking views on whether “recent developments” have created “a relevant merger” situation. It is entitled to do so because the two companies’ turnover in the UK is the relevant consideration – not where they come from.
Since OpenAI’s ChatGPT product is already huge everywhere, and Microsoft is clearly big in the UK, this is fair game. Yes, remedies would be another matter – and it’s hard to see what could be achieved without US counterparts playing a role. But be glad the CMA is awake. The safety of AI products is the top worry, which is not what’s under consideration here. But competition also matters.
