OpenAI’s Altman Courted Investors For AI Chip, Device

Sam Altman was seeking to raise billions from some of the world’s biggest investors in the Middle East and elsewhere for an artificial intelligence chip venture in the weeks before his shock ouster from OpenAI on Friday, according to multiple reports.

Altman pitched AI-related projects to investors during a fund-raising trip to the Middle East in October, including a plan to develop custom chips for training AI models that would compete with chips from market leader Nvidia, the New York Times reported.

Altman sought billions of dollars for the project, which was code-named Tigris, Bloomberg reported.

Altman and OpenAI co-founder and former president Greg Brockman, who quit on Friday in solidarity with Altman, have now both joined Microsoft to head a new AI research team, Microsoft chief executive Satya Nadella said on Monday morning.

Custom AI device

Microsoft only last week at its Ignite conference said it had developed a custom AI chip called Maia, along with an ARM-based CPU called Cobalt for general-purpose cloud workloads.

The AI chip is designed to reduce the company’s reliance on Nvidia, although Microsoft was quick to downplay the chip’s impact on its partnership with Nvidia.

Microsoft said it is already developing a second-generation Maia chip.

Altman has also reportedly been looking to raise funds for an AI-focused hardware device he has been developing with former Apple designer Jony Ive.

The former OpenAI chief has been in talks around both of these ventures with Japan’s SoftBank Group, Saudi Arabia’s Public Investment Fund (PIF) Mubadala Investment Co and others, Bloomberg reported.

‘Not consistently candid’

His fundraising efforts came at a moment when OpenAI has reportedly been working to finalise a tender offer, led by Thrive Capital, that would allow employees to sell their shares at an $86 billion (£69bn) valuation.

SoftBank and others had hoped to be part of the tender offer, but were put on a waiting list, with Altman asking them to consider his two other ventures in the interim.

OpenAI said on Friday that Altman had been fired after an internal review found he “was not consistently candid in his communications with the board”.

The Information said employees  had been told that Altman’s interactions with the board had undermined its ability to supervise AI development.

Commercialisation

The four-person not-for-profit board and Altman had differences of opinion on AI safety, the speed of development of the technology and the commercialisation of the company, Bloomberg reported.

The news service said Altman’s ambitions and side ventures added complexity to an already strained relationship with the board.

But as he took over as interim chief executive on Sunday night, former Twitch chief executive and co-founder Emmett Shear denied Altman had been removed over “any specific disagreement on safety”.

Shear added that he was “not crazy enough to take this job without board support for commercialising our awesome models”.

Shear promised to launch an independent inquiry into the circumstances of Altman’s removal and to “drive changes in the organisation – up to and including pushing strongly for significant governance changes if necessary”.