A new non-profit dedicated to preventing artificial intelligence from becoming a tool of corporate or governmental control was announced on Friday, backed by Silicon Valley giants like Elon Musk, former CTO of Stripe Greg Brockman, and Google’s Ilya Sutskever.
OpenAI will be based in San Francisco. So far, funders including Musk, Amazon Web Services and Y Combinator, have committed $1 billion. However, they say that only “a tiny fraction” of that funding will be spend on the establishment of the organization over the next few years.
“Because of AI’s surprising history, it’s hard to predict when human-level AI might come within reach,” the team wrote in their initial press release. “When it does, it’ll be important to have a leading research institution which can prioritize a good outcome for all over its own self-interest.”
The groups aims to continue the study of artificial intelligence in a way neither encouraged nor constrained by the “need to generate financial return.”
Musk and other researchers, including Stephen Hawking are known to have spoken out against human-level intelligence in AI, citing the dangers of autonomous, intelligent weapons and an AI arms race.
They point to deep learning, pattern recognition, and art programs as evidence of the development of artificial intelligence to a potentially dangerous degree. AI solutions are “less general” than expected, with “impressive but narrow capabilities.” The group envisions a future in which human-level AI are used in a variety of tasks, contributing to many segments of society.
The complete group of founding members are Ilya Sutskever, Greg Brockman (Open AI CTO), inventor Trevor Blackwell of Y Combinator, Vicki Cheung of Duolingo, Google Deep Learning researcher Andrej Karpathy, Deep learning researcher Durk Kingma, PhD candidate in Computer Science John Schulman, AI researcher for Facebook Pamela Vagata, and Facebook AI Research intern Wojciech Zaremba. President of Y Combinator Sam Altman and Elon Musk will serve as co-chairs.
The group will publish papers in research journals, as blog posts, or code, and will collaborate with other institutions and companies. If their plan pans out, OpenAI will become the leading research institution for artificial intelligence – and keep it safe.
