Building artificial general intelligence is OpenAI’s ambitious goal
By James Vincent Jul 22, 2019, 10:08am EDT
Microsoft is investing $1 billion in OpenAI, a San Francisco-based research lab founded by Silicon Valley luminaries, including Elon Musk and Sam Altman, that’s dedicated to creating artificial general intelligence (AGI).
The investment will make Microsoft the “exclusive” provider of cloud computing services to OpenAI, and the two companies will work together to develop new technologies. OpenAI will also license some of its tech to Microsoft to commercialize, though when this may happen and what tech will be involved has yet to be announced.
OpenAI began as a nonprofit research lab in 2015 and was intended to match the high-tech R&D of companies like Google and Amazon while focusing on developing AI in a safe and democratic fashion. But earlier this year, OpenAI said it needed more money to continue this work, and it set up a new for-profit firm to seek outside investment.
To attract backers, OpenAI has made outrageous promises about the potential of its technology. Altman, who became CEO of the new for-profit OpenAI, has said that if the lab does manage to create artificial general intelligence, it could “maybe capture the light cone of all future value in the universe.”
In order to restrain the greed of investors, OpenAI operates as a capped-profit entity, meaning anyone who puts in money can only expect returns 100 times their investment. That’s not offering “all future value” in the universe, but it’s not pocket change either. Exactly what terms Microsoft and OpenAI have agreed on with this $1 billion investment isn’t clear.
Altman’s confidence in OpenAI comes from his belief in AGI. This has been the holy grail of the field of artificial intelligence for decades, and it refers to an AI system that is as flexible and generally intelligent as a human being.
Right now, researchers are able to create AI software that is superhuman in specific domains, like playing certain board games or analyzing medical scans, but these systems are unable to transfer that competency from one task to another. As Facebook AI chief Yann LeCun put it: when it comes to general intelligence, we can’t even build something as smart as a rat.
When exactly researchers might be able to create AGI — and whether it’s even possible — is a topic of lively debate in the community. In a recent survey of some of the field’s leading experts, the average estimate was that there was a 50 percent chance of creating AGI by the year 2099.
To date, OpenAI has certainly impressed the AI world with its research. It has set new benchmarks for robot dexterity; its gaming bots have flattened human champions at Dota 2; and it’s designed remarkably flexible text-generation systems, which can write anything from convincing song lyrics to fake news articles and short stories.
But none of these projects look like the huge step-change that a leap to AGI entails. Meanwhile, other researchers have bemoaned OpenAI’s hyperbole and questioned whether its switch to for-profit research undermines its claims to be “democratizing” AI.
In a statement, Microsoft CEO Satya Nadella said the partnership would keep “AI safety front and center” so that “everyone can benefit.” Altman stressed that safety and spreading the “economic benefits” were key concerns.
“The creation of AGI will be the most important technological development in human history, with the potential to shape the trajectory of humanity,” said Altman. “Our mission is to ensure that AGI technology benefits all of humanity, and we’re working with Microsoft to build the supercomputing foundation on which we’ll build AGI.”