Microsoft maneuvers to a prime role with banks in OpenAI upheaval

HDM109 Microsoft.jpg
Hiring the creators of ChatGPT during a leadership crisis at OpenAI gives Microsoft an edge in becoming the leading provider of generative AI for financial institutions, experts say.

The chaos caused by the sudden defection of ChatGPT founder Sam Altman from the company he founded to Microsoft ultimately may prove to bring much-needed clarity for financial institutions seeking an AI road map.

When OpenAI's board unexpectedly parted ways with CEO Altman on Friday, Nov. 17, citing his "lack of transparency," it briefly sent a signal of uncertainty about the trajectory of generative AI, whose potential has sparked both optimism about its power to improve operations, and fear about its lack of controls and regulations.

For a period of time over the weekend, Altman seemed to suggest he might return to OpenAI, and nearly 600 of OpenAI's employees signed a letter threatening to quit if Altman didn't return to his original post. 

But Microsoft's swift decision to hire Altman and OpenAI's president and co-founder, Greg Brockman, to lead an independent research lab at the Redmond, Washington-based firm, may be the stabilizing influence banks have been looking for, according to various experts.

While OpenAI on Monday named Emmett Shear, who previously headed Twitch, as OpenAI's interim CEO — after briefly naming Mira Murati, the firm's chief technical officer, as interim CEO — Microsoft is perceived to have already absorbed most of OpenAI's former momentum. 

"I wouldn't be surprised if Microsoft starts building their own hardware around AI and possibly even providing their own chips to help with the computing power problem with AI, in order to try to control the entire AI stack and be as vertically integrated as possible," said Alex Johnson, an independent fintech analyst.

The drama apparently was caused by a philosophical split within OpenAI over the firm's mission, according to Johnson.

"The central tension seems to have been that some employees — led by Altman — wanted to move quickly to roll out generative AI-based products for commercial use (like ChatGPT), while another group of more cautious employees was concerned that OpenAI was moving too fast and being irresponsible in its direction," he said.

The move gives Microsoft — which had already invested $13 billion in OpenAI — a distinct edge as a trusted supplier of generative AI for corporations, including financial institutions, because of its deep pockets and reputation for developing reliable products that pass muster with regulators, Johnson said.

Microsoft's CEO Satya Nadella on Monday said on X, formally known as Twitter, that Microsoft looks forward to moving quickly to provide Altman's group with resources they need to succeed. Because of this, Johnson speculates that "a large number" of OpenAI employees will eventually migrate to Microsoft.

For financial institutions and banking core providers that already use Microsoft's Azure cloud computing platform, the fact that Altman is now under its umbrella should be reassuring, said Richard Crone, a partner with Crone Consulting LLC. 

"Microsoft's investment and involvement with OpenAI strengthens its position as a preferred vendor to financial institutions over competitive cloud-based platforms like Amazon, Oracle, IBM and the like because AI functionality is integrated into the Azure platform and possesses the talent for large language models in-house," Crone said.

One core processor, Finastra, primarily uses Azure for its entire cloud-based operation, according to Crone. "Others like FIS, Fiserv and Jack Henry & Associates, who are minimally invested in gen AI, must weigh the long-term impact of OpenAI's leadership changes on their tech evolution," he said.

While many financial institutions are eager to use generative AI to harness troves of data to improve their products, operations and customer service, there's also great anxiety about whether the resources they adopt will be safe and secure, according to Crone.

The MeridianLink ransomware breach spotlighted the risks for banks of relying on fintechs that haven't been fully vetted by accepted standards and protocols, he noted.

"Operational continuity is a regulated mandate in the vendor due diligence process for any third-party processor, and Microsoft's fusion with OpenAI and commitment to absorbing its talent solidifies Azure as a leader in AI and cloud services," he said.

OpenAI's leadership upheaval is a further sign of the transformative effect generative AI is expected to have in many industries and the role corporations will play in shaping its impact on society, said Hugh Tallents, a senior partner at management consulting firm cg42.

"There's been a sort of public arms race over the last 12 months to establish AI, but this has been building in the tech community over the last five years, with people now talking about ethics around AI and worried about the lack of oversight and regulation in this area," he said.

In his opinion, OpenAI's board recently grew uneasy about its inability to control the direction and use cases for generative AI and acted too hastily to oust Altman.

"They panicked when Altman started talking about different types of oversight that might create the equivalent of something like blockchain, or distributed ledgers, to manage AI development, and that makes board members squirrely," Tallents said.

Altman's unexpected exit is more illustrative of poor communication and planning than evidence of uncertainty around generative AI's trajectory, he said.

"Microsoft has what OpenAI does not, which is access to Washington, and the capacity to help educate the regulators who are going to have to do their job in this environment over the next two years, putting a framework in place to ensure that what's created is useful and not destructive," Tallents said.

Many banks may be more likely to lean on Microsoft because of its sudden expansion of AI talent, but other big tech firms investing in AI — including Google, Amazon, IBM, Oracle and others — will now need to make their own bigger plays within AI or risk falling behind in the fast-changing environment, he said.

"Even if many banks rely on the same provider for generative AI models, succeeding in this area will come down to having the skills and foresight to use it in a way that benefits their unique strengths," Tallents said.

For reprint and licensing requests for this article, click here.
Artificial intelligence Technology
MORE FROM AMERICAN BANKER