Many setups in the San Francisco Bay Area boast that they are planning to change the world. However, OpenAI founded by Elon Musk has made a bigger promise than the rest: It wants to build artificial general intelligence (AGI), an AI system that like humans, can reason across many different domains and apply its skills to unfamiliar problems.
For this reason, it announced a billion-dollar partnership with Microsoft to fund its work. This hints that AGI research is leaving the field of science fiction and entering the territory of serious research.
“We believe that the creation of AGI will be the most important tech development in human history, with the potential to chaneg and shape the trajectory of humanity,” Greg Brockman, chief technology officer (CTO) of OpenAI, informed the press.
This investment will make Microsoft the exclusive cloud computing services provider for OpenAI, and the both companies will work together to create new technologies. OpenAI will also license some of its technology to Microsoft to commercialize, although when and how this may happen has yet to be announced.
To date, OpenAI has certainly impressed the AI world with its research. Its gaming bots have flattened human champions at Dota 2; it has set new standards for robot dexterity and it designed remarkable text-generation systems, which can write anything from fake news articles, convincing song lyrics to short stories.
Current AI systems outpace humans at lots of narrow tasks-be it chess, Go, Starcraft, or image generation. They are also catching up to humans at others like translation and news reporting. However, artificial general intelligence is said to be that one system with the capacity to beat us at all of those things.
Many enthusiasts believe that it would enable centuries of technological advances to arrive, effectively, all at once — transforming food production, medicine, green technologies, and almost everything else in sight.
While others warn that if poorly designed, it could be a menace for humans in a few different ways. Quite an advanced AI could pursue a goal that was never aimed for-a recipe for catastrophe. Once running, it could turn out unexpectedly impossible to correct. Otherwise, it could be used unkindly by a small number of people to harm others. Or it may just make the rich even richer and leave the rest of humanity even further in the dark.
It is a vital challenge ahead for humanity to get AGI. Microsoft’s billion-dollar investment has the potential to break the barriers for AI growth, but to get AGI technology right, investors have to be willing to prioritize all safety concerns that might slow its commercial development.
Breakthrough technology with pros and cons
Some analysts have compared the development of AGI akin to the development of electricity. However, transformation is just one side of the picture; on the other hand, it has brought numerous other changes in the way we live our lives.
With the benefits that come from this technology, there are chances that things could go wrong too. OpenAI’s team is vigorously working on the safety and policy implications of AGI has been unafraid to articulate debacle rather than a blessing.
“To accomplish our mission of ensuring that AGI benefits all of humanity,” Brockman says in the press release, “we will need to ensure that AGI is deployed securely and safely; that society is well-prepared for its implications; and that its economic upside is widely shared.”
In a statement, Microsoft CEO Satya Nadella ensured that the partnership would keep “AI safety front and center” so that “everyone can benefit.”
There are hard problems. Current AI systems are susceptible to adversarial examples — inputs designed to confuse them — and more advanced systems might be, too. Current systems loyally do what we tell them to do, even though if it’s not exactly what we meant them to do.
It is anticipated that there are some reasons which suggest that advanced systems will have problems that current systems don’t. Some researchers have argued that an AGI system that may appear to be performing well at a small scale might unexpectedly fail in bigger performance. This could be due to more resources available to it as the best route to achieve its goals changes. Imagine this by thinking about a company that follows the rules when it’s relatively small and inspected, but cheats or lobbies to get the rules changed once it has enough power to do so.
Even AGI’s most enthusiastic advocates think there’s a lot of potentials for things to go wrong, but they feel that the benefits of developing AGI are worth it and outweigh the risks. A success with AGI could let us address climate change, extreme poverty, pandemic diseases, and whatever new challenges are around the corner, by optimizing our power grid, identifying promising new drugs and speeding up the rate at which we develop new technologies.
When is AGI coming?
Here also, experts have a difference of opinion. Some estimate that we’re only ten years away while others point out that there’s been optimism that AGI is just around the corner for a long time and it has never arrived.
The disagreements are still in the grey area. Some academics, such as MIT’s Max Tegmark, are among those predicting AI soon, while some key figures in the industry, such as Facebook’s Yann LeCun, are among those who think it’s likely somewhat distant. One thing that they all agree on is that it is possible and will happen someday, and that makes it one of the huge, open challenges of this century. The median estimate is of a 50 percent chance of creating AGI by the year 2099.
From non-profit to a for-profit organization
OpenAI began as a non-profit lab in 2015 and was envisioned to match the high-tech R&D of companies like Amazon and Google while focusing on developing AI in an independent and safe fashion. Musk, one of its founders, left the board in 2018, quoting conflicts of interest with Tesla. 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. The new company is called Open AI LP where LP stands for “limited partnership.”
To attract backers, OpenAI has made audacious promises about the potential of its technology. Altman, who became CEO of the new for-profit OpenAI, has said that if the laboratory does manage to create artificial general intelligence, it could “maybe capture the light cone of all future value in the universe.”
Altman has shown immense trust in OpenAI, coming from his belief in AGI. This has been the pot of luck of the field of artificial intelligence for many decades, and it refers to an AI system that is generally intelligent and as flexible as a human being.
The experts criticized the transition from non-profit to for-profit firm and inferred it as a betrayal of the nonprofit’s democratic mission. So why the change? The OpenAI’s leadership team had realized that they require pumping of billions of dollars to stay on the forefront of the field and pave the way towards AGI, which is hard for a nonprofit to get.
However, once you have investors, you have responsibilities to maximize their profits by hook or crook, which is irreconcilable with ensuring that the benefits of AI are widely distributed.
OpenAI LP is meant to solve that problem. It operates as a capped profit entity to restrain the greed of investors. It’s a hybrid, OpenAI, says, of non-profit and for-profit, and the company promises to cap the pay of the shareholders to a return on their investment, upto 100 times of what they invest. Everything beyond that goes to the general public. The OpenAI nonprofit board still oversees everything.
Well, this seems quite absurd, doesn’t it? After all, how much can possibly be left over after paying investors 100 times what they paid? But the proposition can not be ignored entirely. Many early investors in many tech companies have made far more than 100 times what they invested. Jeff Bezos reportedly invested $250,000 in Google back in 1998; if he held onto those shares, they’d be worth more than $3 billion today. If Google had adopted OpenAI LP’s cap on returns, Bezos would’ve gotten $25 million — a fine return on his investment — and the rest would go to humankind.
If OpenAI hits the jackpot, Microsoft will no doubt profit immensely — but, they say, so will the rest of us. “Both companies have very aligned missions,” Brockman said: “Microsoft to empower every organisation and every person on the planet to achieve more; OpenAI to ensure that AGI benefits all of humanity.”
Many are eagerly waiting to see how the technology would bring change in the future – whether such partnerships can drive advances that are good for humans or put the brakes on advances that are bad.