Overview of the OpenAI company

- To understand what motivated the formation of Open AI, the San Francisco based company behind the GPT-3. I need to briefly discuss a core debate, within the AI community. It's this debate, which makes open AI notable in terms of its Genesis as an AI development organization. This debate asks whether advances in AI present on one side of the argument, a better more prosperous future, or on the other side, a future where AI is to be feared. Those who say AI is a positive force, such as Ray Kurzweil, the American Inventor, Futurist and Machine Learning Director Google, argue that it will free up humans from more leisure time and enable us to focus on creative pursuits. Those on the other, concern side of the argument, such as Bill Gates, the former CEO of Microsoft, worry that AI is both amazing and dangerous. He fears, for example, AI being applied a warfare to manage autonomous weapons, those that search for and destroy targets without human intervention. Elon Musk, the CEO of Tesla and SpaceX, has been outspoken about his concerns. Much like the late Stephen Hawking, the notable theoretical physicist and cosmologist. They both characterize some forms of future AI as potentially humanity's greatest existential threat. A concerning view, indeed. Responding to this concern, in October 2015, Elon Musk, along with Sam Oltman, former president of startup accelerator Y Combinator and other investors such as co-founder of LinkedIn Reid Hoffman and Peter Thiel co-founder of PayPal created open AI. Corporations, such as Microsoft and Infosys were also founding members. By acknowledging their concerns about the risks of future AI and committing to actions to prevent such risks. These partners formed Open AI to promote and develop friendly AI in a manner that would benefit humanity. To demonstrate their commitment, together the founding team pledged over $1 billion to create the organization. Their mission was clear. Open AI would be focused on ensuring that artificial general intelligence or AGI would benefit the world. They would build safe AGI and we'll consider their mission fulfilled if their work aided others in achieving the same outcome. The organization began as a nonprofit in early 2016 with nine researchers. By April of that year, they released their first product called Open AI Gym. A toolkit for developing and comparing reinforcement learning algorithms. Reinforcement learning is when a computer employees trial and error to come up with a solution to a problem. The AI receives either rewards or penalties for the actions it performs . In December 2016, Open AI released universe. A platform for enabling AI software to interact with a computer such as viewing what's on the screen and using a virtual keyboard and mouse. These solutions were followed by several more in the machine learning space. And in 2018, Open AI published a paper on generative pre-training and what would soon become the first versions of GPT software. I discuss the details of GPT and generative modeling in a later video. Much of the code behind the AI so far produced by open AI, has been open sourced and is freely available. This is to encourage transparency, learning that dissemination of benefits and for the advancement of the code. However, the latest version of GPT version three, the subject of this course has not yet been open sourced. The reasons Open AI provide are mixed, including the size of the code and the opportunity for monetization. They also fear the challenges it could create, given its power to create problematic output that could be offensive or subject users to litigation. More on that later . In 2019 ,Open AI dropped its designation as a pure non-profit and became a for-profit enterprise. This was done to attract further investment by allowing investors to get returns, if open AI became profitable. Microsoft stepped forward and became a significant investor by pumping $1 billion into the new commercial entity. Further, Microsoft became Open AI's preferred partner to license its technology and integrated into their existing as your powered AI platforms. And into their new products and services. Elon Musk's involvement in Open AI decreased significantly in 2018. After resigning his board seat. He cited a potential conflict of interest with his work at Tesla in developing AI for self-driving cars. He does remain engaged as a donor. Today ,with the remarkable initial success of GPT-3. And with the weight of Microsoft behind it. Open AI is well positioned to produce further AI breakthroughs and have a discernible impact on all of our futures.