FUTURE DECODED: Microsoft describes how ‘respectful’, ‘trusted’ AI and cloud infrastructure can change every business responsibly
Microsoft has put artificial intelligence (AI) at the heart of its vision for digital transformation and stressed it has plotted a sustainable future for the development of its new technology and use of cloud.
“Microsoft is not immune from the waves shaping our business,” Cindy Rose, Microsoft’s UK CEO told Future Decoded in London. “Whether it’s cloud or AI, Microsoft is seeking to not only embrace but lead these waves of change and take full advantage of these opportunities.
“It’s safe to say at Microsoft we have learned about ourselves. Our view is that a digital business is a business that embraces technology to effectively engage customers, empower employees, and transform products and services in a way that turns disruption into opportunity.
“We are on the cusp of a fourth industrial revolution.”
Role of AI
And it’s clear on how AI will accelerate this process. Chris Bishop, laboratory director at Microsoft Research Cambridge, talked about his teams efforts in AI and how machine learning was powering the advent of intelligent computing.
“[Our vision of] artificial intelligence is to empower every person and organisation on the planet to achieve more,” he said. “AI is a very powerful and very broadly applicable tech. Over the years and decades to come it will transform every aspect of our lives.”
Many leading figures of the technology and scientific communities have expressed concerns about the threat advanced AI could pose to the human race, but although Bishop acknowledged such fears, he does not hold them.
He said AI should “augment” human ability and experiences, be trustworthy, inclusive and respectful, while it was Microsoft’s mission to democratise the technology.
“Microsoft’s approach is not to replace people with machines but to power them to achieve more,” he said.
Two examples shown on stage were a research project called InnerEye that is using AI to help treat cancer and SwiftKey, which used AI to create predictive emoji texting.
“Our goal is not to replace the doctor,” said Bishop. “Machines are complementary to the capabilities of people. Where the AI is powerful is it is precise, quantitative and fast.”
“Machine Learning was critical for this to work,” added SwiftKey. “Emojis are such a new language and doesn’t follow the same rules [as verbal languages].”
Machine learning is the critical component of this vision, as is the cloud infrastructure that supports it. To do this, Microsoft requires algorithms,
“Microsoft is transforming the infrastructure for AI,” explained Bishop. “Modern AI is based not on the idea of making a computer to be intelligent, that was tried in the 60s and didn’t work very well, but instead on machine learning. The computer is not trained to know the problem but to learn using large amounts of data.”
To illustrate, Bishop described an example closer to home. An AI chatbot has been deployed on some Microsoft’s own customer service webpages with the ultimate goal of revolutionising remote support. The bot promises a “frictionless” human-like conversation and if it cannot resolve the customer’s query, it is referred to one of 2,000 agents. Either way, the AI learns from the interaction.