Posted in: AI

MIT and IBM: putting our “minds and hands” together to create the future of AI

As an MIT graduate and senior leader within IBM Research, I have always felt a close kinship between these two institutions. Both are renowned for their technical excellence, and both are strongly committed to pushing the frontiers of science and technology to solve problems that matter to the world. MIT’s motto “Mens et Manus” (Latin for mind and hand) echoes our values here at IBM, to leverage the talent we have and create real technology with impact. That is why I am so excited about the creation of the MIT-IBM Watson AI Lab, the partnership we have formed spanning the next 10 years, through which we will achieve fundamental scientific breakthroughs at the heart of AI today and into the future.

The official seal of MIT, depicting its motto “Mens et Manus,” which is Latin for mind and hand.

Together with our fellow scientists at MIT, we selected four key pillars for our collaboration: core algorithmic advancements that enable learning and reasoning to broaden what AI systems can do, computational innovations tailored to AI and achieved through a mastery of physics, applications of AI to important domains like healthcare and cybersecurity, and achieving shared prosperity through AI technology. Each one of these areas touches upon our fundamental beliefs about the future of AI and where we believe it can and should go. And all four leverage technical strengths shared across MIT and IBM.

AI technology today has been extraordinarily successful at performing individual tasks effectively, though with considerable effort and oversight by the people who train the models. Through this collaboration, we will target innovations that will move us beyond specialized tasks to more general approaches to solving more complex problems, with the added capability of robust, continuous learning.  We will explore not only how to best leverage big data when available, but also learn from limited data to personalize and augment human intelligence.

Today, it takes an enormous amount of time to train high-performing AI models to sufficient accuracy. For very large models, it can be upwards of weeks of compute time on GPU-enabled clusters. That’s why we selected the Physics of AI as a critical area to explore in partnership with MIT. Our teams will explore new materials, devices and architectures for analog AI computation, as well as the intersection of quantum computing and machine learning. The latter involves both using AI to help characterize and improve quantum devices, and also researching the use of quantum computing to optimize and speed up machine-learning algorithms and other AI applications.

While a significant amount of the work we plan to do together is focused on achieving fundamental scientific breakthroughs, we are also extremely committed to leading in the application of AI to solve crucial problems in healthcare and security. And given the value of face-to-face collaboration, the fact that Watson Health and IBM Security headquarters are in Kendall Square, just a few blocks from MIT, puts us in a great position to make strides in these areas.

Finally, through our partnership with MIT, we will pursue an agenda of advancing shared prosperity through AI to achieve broad economic and societal benefits for the largest number of people. Some of the questions we’ll address together include the creation of AI systems that can detect and mitigate human biases, building trustworthiness and explainability into AI systems, ensuring that AI systems complement worker skills that might be in short supply and exploring how productivity gains will be distributed across firms, workers and consumers. We intend to build upon the Principles for the Cognitive Era that we laid out earlier this year, as well as our work as a founding member of the Partnership on AI, a consortium that focuses on guiding the development of AI to the benefit of society. As the creators of this technology, we take responsibility for ensuring that it is developed the right way and for the right reasons.

To add to our enthusiasm around each one of these pillars individually is our belief that bringing them all together will offer profound opportunities. We intend to develop the next generation of AI algorithms using accelerated computation, achieved through our mastery of physics, which we apply ethically and in pursuit of shared prosperity, e.g. for better health outcomes and a safer future. It is an exciting time to be involved in AI research, and an even more exciting time for MIT and IBM. I know we will build amazing things together, and I sincerely look forward to sharing them with the world.

Save

Save

Save

Save

Save

Comments

  1. Sam Rai says:

    Great initiative, excited to hear more about the developments

  2. Marilyn James-Kracke says:

    IBM Watson would more effectively provide medical diagnoses if the many and common symptoms of nutritional deficiencies were programmed into it. Mineral, vitamin and essential oil deficiencies have become very common. A recent study demonstrated that 85% of addicts to opioid pain medications are very low on Vitamin C. Their back pain is due to weak collagen in cartilage in joints and muscle causing tissue tears. Scurvy causes bleeding into joints and muscles which causes pain. Scurvy responds quickly to Vitamin C supplementation especially if the diet is changed to include fruits and vegetables. There are many diagnoses that just describe a symptom, like erythema multiforme, but this does not say what caused it. Diagnoses need to include causes so these can be fixed quickly and inexpensively. Big medical terms do nothing to define or correct the problem. No drug precisely fixes a nutritional deficiency; drugs usually only mask symptoms leaving the problem to progress. Healing requires complete nutrition. The symptoms of nutritional deficiency are known. They can be complex when many nutrients are deficient. However, this is what computers are good at unscrambling. Lots of illnesses, including type 2 diabetes and hypertension would rapidly be fixed if IBM Watson helped physicians to diagnose the very common nutritional deficiencies (low Mg and B vitamins). Would physicians actually followed the recommendations to correct these? Malnutrition (excess calories and too few nutrients) is ignored in most medical school. Therefore it does not surprise me that a search for IBM Watson nutritional deficiencies returns no results for both terms. To gain the most value from IBM Watson AI, the scientific information in Watson should not ignore nutritional deficiencies or the diagnoses it provides will be artificial indeed. Artificial intelligence needs to be bases on complete scientific input in order for it to provide scientifically real conclusions.

    1. Edwin Pell says:

      Max Planck said “Science advance one funeral at a time.” It is seldom that people change their world view. No matter how good a cognitive application is it may only help move people already close to the truth (fence sitters). But even that will save many lives.

  3. Ashish Verma says:

    Great to see both of my worlds coming together as a graduate of MIT and an employee of IBM.

  4. Mark Joyella says:

    Congrats! This is very exciting and brings all of what IBM and Watson are doing to a place where brilliant young minds can learn from it–and help to guide it. Talk about a win-win.

    I can’t wait to see what those lucky enough to work in the MIT-IBM Watson AI Lab do.

  5. Edwin Pell says:

    Broader AI is a marathon not a sprint. Forming alliances to develop the technology, as we have long done in semiconductor fabrication, is a great idea. I hope to see the alliance grow.

Add Comment

Your email address will not be published. Required fields are marked *

AI

Dario Gil

Vice President of AI and IBM Q