Microsoft has announced a partnership with Elon Musk, Peter Thiel, and Sam Altman's non-profit Open AI to further research into deep learning and neural networks.
Open AI is a non-profit organization designed to promote artificial intelligence that "helps" rather than hurt humanity. The three entrepreneurs started the venture in 2015 with a $1 billion investment. Now, the non-profit plans to use "thousands" of Microsoft's virtual machines.
According to tOpen AI's blog, Microsoft Azure will be a crucial part of the non-profit's mission. Going open source will be a good way to protect against the possibility of a centralized, monopolized power over AI. Altman said artificial intelligence has the power to harm humanity if used irresponsibly.
The blog post added that more and faster computers can greatly accelerate the non-profit's plans to research on emerging technologies such as reinforcement learning and generative models. It noted that Microsoft Azure has hardware configurations that are optimized for deep learning, such as K80 GPUs with InfiniBand interconnects.
They plan on releasing the research results with everyone. They are also planning to release open-source software to make it easier for people to run large-scale AI workloads on the cloud.
The non-profit also praised Microsoft's stance toward "democratizing AI." It said in a blog post that Microsoft will be harnessing AI to "fundamentally change" how people interact with computers, to infuse every application with intelligence capabilities, and to provide developers with the same materials. They are also planning to build the world's most "powerful AI supercomputer" and make it available to everyone.
It can be remembered that Elon Musk has spoken about his fear that artificial intelligence may even be "potentially dangerous as nukes." He even told in a tweet that hopefully, we are "not just the biological boot leader for digital superintelligence."