Microsoft turns up the AI heat on tech, ethics and saving Planet Earth


Microsof is upping its AI game with a new research lab and a drive to pursue a wider societal mission.

Harry Shum

If anyone still doubts that Artificial Intelligence (AI) is the hottest tech sweet spot of 2017, look no further than the way Microsoft just upped its skin in the game.

The firm has opened a new research lab at its Redmond headquarters with a team of over 100 scientists working on AI, from various fields including natural language processing, learning and perception systems. It’s inevitably going to be compared to Google’s Deep Mind outfit.

Microsoft had last year formed Microsoft AI and Research, with more than 7,000 computer scientists, engineers, and researchers from Microsoft’s various labs. According to Harry Shum, Executive Vice President for that divison, one reason for the latest moves is to bring various diverse AI research initiatives, such as such as machine learning, perception and natural language processing:

This integrated approach will allow us to develop sophisticated understandings and tools that can help people do complex, multifaceted tasks. For example, using this approach we can create methods and systems that understand language and take action based on that understanding.

Take machine reading. It’s an emerging field that has incredible potential for situations such helping a doctor quickly find important information amid thousands of documents, saving time for higher-value and potentially life-saving work.

To do machine reading well requires combining AI disciplines such as natural language processing and deep learning. In fact, Microsoft’s integrated approach is currently leading a competition in the field.

But there’s also a societal mission statmement for the new research lab, he adds:

The intersection of AI with people and society presents us with incredible opportunity and demanding challenges. As AI breakthroughs abound, we as a technology company have the chance to channel all those innovations into tools that help people do their jobs better and more efficiently, and that solve some of the world’s toughest challenges. Using AI, we are already finding better ways to do things as seemingly narrow as determining what email warrants your attention and as astonishingly complex as discovering a personalized cancer treatment.

We are responsible for building AI advances that amplify human ingenuity, and also that reflect our shared societal values and expectations. The AI tools and services we create must assist humanity and augment our capabilities. The people who use those tools should be able to understand how they work and what data they rely on. AI can be more useful if the people who created and use the tools can explain how they work and why decisions are made.

Despite all the progress we’ve made so far, we know that we are still in the nascent stage of development of AI tools and technology. Technology that uses AI can perform specific tasks well, such as correctly identifying an image or recognizing the words in a conversation, but it cannot yet begin to compete with a child’s ability to understand and interact with the world around her using senses such as touch, sight and smell.

Saving the planet

In a clear sign of the social potential around AI, Microsoft also announced a new project called AI for Earth that will provide $2 million in cloud computing credits and training in machine learning techniques to scientists doing research on sustainability issues related to agriculture, water, biodiversity and climate change.

According to MIcrosoft President and Chief Legal Officer Brad Smith:

Our goal is to empower others in new and more impactful ways to help create a more sustainable future. This program expands our commitments to democratizing AI and advancing sustainability around the globe. And it builds upon our experience in accelerating the pace of innovation bringing together philanthropic work, connectivity advances and more accessible technology around the world.

There is no shortage of work to be done to chart a healthier, more sustainable future for our planet. Time is not our friend on this endeavor, and the stakes are high. We know that AI for Earth is just a first step in a longer journey, but what we’ve seen so far gives us optimism for the future.

AI for Earth is built around three foundational pillars:

  1. Access: New grants will be on offer to researchers and organizations to provide access to cloud and AI computing resources, including Azure compute time and data science virtual machine offerings on Azure.
  2. Education: New training and educational schemes will be delivered to help researchers and organizations know what AI tools are available, how to use these tools and how the tools can help meet their specific needs.
  3. Innovation: Microsoft will partner with third parties on “lighthouse projects that demonstrate how AI can deliver results more rapidly, accurately and efficiently”. There are three of these already underway – one enabling land cover mapping to aid precision conservation; another to enable smart agriculture through sensors, drones, data and broadband connectivity; and a third to test the viability of using smart mosquito traps to remotely track and monitor species health.

Both new initiatives come on the heels of an alliance with Facebook, Google, Amazon and IBM “to study and formulate best practices on AI technologies, to advance the public’s understanding of AI, and to serve as an open platform for discussion and engagement about AI and its influences on people and society”.

My take

This is Microsoft in catch-up mode to some degree, albeit perhaps more from a PR and perception point of view than a technology one.

The popular AI imagination has been owned by Google Deep Mind – not necessarly always for the better – and IBM Watson winning TV quiz shows in the US and beating chess grandmasters. In more recent months we’ve seen enterprise business software providers pitching the integration of AI with their core offerings, such as Salesforce with Einstein or this week Infor with Coleman.

We can expect to see a lot more noise coming out of Microsoft in respect of its own AI ambitions and capabilities in the coming months. There’s still everything to play for with no clear long-term winners or losers yet in the AI market. And lets never forget, once upon a time Microsoft was ready to dismiss the internet until one day Bill Gates fired off a memo and put the firm on war footing. Ask Netscape what happened after that…

Image credit - Microsoft

Disclosure - At time of writing, Infor and Salesforce are premier partners of diginomica.