Over the course of the 70+-year history of computing, government, academia and industry have joined forces repeatedly to bring cutting-edge science to bear on the some of the major challenges of the day—and, in each case, computer science has played an important role. Among those remarkable efforts were the Manhattan Project, the Space Race, and the Human Genome Project.
Today, we’re at a turning point in the evolution of information technology. A new age is dawning: the era of cognitive computing. At IBM, we believe it will be as distinct from today’s era of programmable computing as this period was from the earlier tabulating era. IBM’s Watson, which defeated two former grand-champions on the TV quiz show Jeopardy, was a first step forward on a long journey. Over the coming years and decades, computers will learn, reason and partner with human beings to help them harvest the benefits of Big Data and better understand how the world works so people can make superior decisions and live more successful lives.
Yet, there’s no assurance that breakthroughs in cognitive computing will progress as quickly as their advocates would like. That’s why IBM is reaching out to partners in academia, industry and across our clients in an effort to set a shared agenda aimed at hastening and guiding progress.
To learn more about the new era, download a free chapter of the new book, Smart Machines: IBM’s Watson and the Era Cognitive Computing, by IBM Research Director John E. Kelly III, at http://cup.columbia.edu/static/cognitive.
Tomorrow marks a first step in the process of setting an agenda. We have invited scholars from a number of leading universities to a Cognitive Systems Colloquium at IBM Research headquarters in Yorktown Heights, New York, to begin a conversation about the new era of computing. (Additional colloquia will follow at other IBM Research labs around the world.) As a signal of our commitment to this cause, we’re creating a collaborative research initiative with four universities in order to advance the study of cognitive technologies. The universities include Carnegie Mellon, MIT, New York University and Rensselaer Polytechnic.
The initial research efforts will support the following investigations:
–MIT: How socio-technical tools can boost the collective performance of moderate-sized groups of humans engaged in collaborative tasks such as decision making.
–RPI: How advances in processing power, data availability, and algorithmic techniques can enable the practical application of a variety of artificial intelligence techniques.
–CMU: How systems should be architected to support intelligent, natural interaction with all kinds of information to help people execute complex tasks.
–NYU: How deep learning is impacting many areas of science where automated pattern recognition is essential.
But that’s just a start. Cognitive systems will require innovation breakthroughs at every layer of information technology, starting with nanotechnology and progressing through computing systems design, information management, programming and machine learning, and, finally, the interfaces between machines and humans. Advances on this scale will require remarkable efforts and collaboration, calling forth the best minds—and the combined resources–of academia, government and industry.
In the coming months, we plan on working with additional universities, researchers, IBM clients and government agencies to map out a common research agenda, set priorities and identify areas where substantial progress needs to be made. Businesses can play an important role by helping to identify situations where cognitive systems could provide substantial benefits—and by providing real-world test beds were researchers can evaluate emerging technologies.
Over time, armed with these powerful new tools, people and organizations will be able to take on challenges that have confounded the human race. Cognitive machines have the potential to help us live more sustainably so we can address the destructive effects of climate change. They can help doctors defeat cancer and other dread diseases. They can help businesses operate more effectively and dynamically, yet, at the same time, minimize the negative effects of commerce on people and the environment. And, most broadly, they can help individuals, organizations, and, indeed, all of society, function in ways that foster peace, economic plenty and individual fulfillment.
But these tools will also pose new challenges to society. How do we incorporate into our lives computing systems that increasingly think like we do, that discover and provide insight from vast amounts of information?. Issues of privacy, and security must be addressed. More than any information technology waves of the past, cognitive computing will require individuals and institutions to be aware of the changes that are coming and to guide the way the technologies are used.
Today, we can only make educated guesses about where cognitive technologies will take us. Like the great technology visionary and A.M. Turing Award winner Alan Kay has said, “The best way to predict the future is to invent it.” So, together, let’s invent the cognitive era.
Visit A Smarter Planet Blog on Oct. 2 starting at 8:30 a.m. for live blogging from IBM Research’s Cognitive Computing Colloquium, featuring speeches by Nobel Prize winner Daniel Kahneman, MIT professor Thomas Malone, A.I. visionary Danny Hillis and more.
Apply now. Be an IBMer.
The 2010 earthquake in Haiti increased two American doctors’ efforts to create a humanitarian, non-profit program named Colleagues In Care in an effort to provide education and training to those who were affected by the earthquake. With support from IBM, Colleagues In Care is utilizing SmartCloud to share medical information with this island, just three hours from the United States. From the gray avenues of New York City to the colorful streets of Port-au-Prince, and all around the world, doctors and technologists are collaborating to save lives.
Watch below for the story of two humanitarians and how their partnership with IBM is helping local Haitian doctors and patients.
Imagine a day in the not-too-distant future when your car will alert you to a dangerous condition a mile ahead so you can slow down pro-actively or take an alternative route.
Or sensors detect abnormal wear on your brakes and the car automatically arranges for an appointment at your repair shop and even checks the parts inventory at the shop to make sure there will be no delay in getting the brake job done. Or a rental car recognizes you when you slip into the driver’s seat and automatically adjusts to your preferences—queuing up your iTunes playlist, adjusting the mirrors and briefing you on the to-do list from your digital calendar.
These are just a few of the scenarios that automakers and their suppliers are dreaming up as the world enters the era of the connected car—all of them enabled by next-generation cloud computing services. But these sophisticated services won’t come quickly or flourish unless the major players in the industry borrow a page from the tech industry playbook. The Internet revolution that brought Yahoo! and Google and Facebook would not have been possible without agreement on a broad set of open standards aimed at easing the flow of communications and the sharing of data. Failure to do the same could put the brakes on the connected car.
Here’s a story of some connected-car technologies being tried out in Eindhoven, the Netherlands.
Automakers and tech companies have long talked about a host of new safety, convenience, maintenance and infotainment features coming to vehicles, but only recently, thanks to the arrival of a handful of technology improvements, are their promises starting to be fulfilled. We’re seeing self-parking and self-braking vehicles, cars you can start remotely with your smartphone, and apps that alert your Facebook friends when you’re running late for a party.
These new services are being made possible by advances in mobile communications, cloud computing, sensor networks and data analytics. But all of those technologies rely on the adoption of open technology standards to deliver the goods in seamless and easy to use ways.
Today, most automakers are going their own way when they choose the core technologies that enable their connected-car services. They picture their services as walled gardens where they can control every experience their customers have while traveling, from start to finish. The best analogy in tech-industry history is America Online, the first hugely successful online service, which lost momentum when the Internet came along and connected every person seamlessly to everyone else. AOL was slow to shift from its proprietary technologies to open Internet standards, and the company paid dearly for that mistake.
At IBM, we have long embraced open standards in computing, and we are every bit as committed to the concept in the connected car. In the past three months, we announced partnerships with mobile carrier Sprint and international auto supplier Continental aimed at enabling a wide array of new applications for vehicles. In each case, we and our partners are committed to making technology choices that assure easy connectivity and sharing of data.
We conducted a smarter traffic pilot program with the Dutch city of Eindhoven where we demonstrated how cars equipped with sensors and advanced networking technologies could be used to monitor road conditions in real time. Raw data from the vehicles identified 48,000 incidents over a six-month period from 1.8 billion sensor signals. A streaming data-analysis system made it possible for the city authorities to react to hazards, accidents and traffic jams in near real time. These results could not have been achieved without our compliance with open technology standards.
One of the most promising new standards is called MQ Telemetry Transport, or MQTT for short. MQTT is a key enabler of interconnectivity between sensors and the computing systems that harvest the data from them and make sense of all that data. It makes it possible for every device on a network to communicate and share information with every other device on the network—and to do it super-efficiently. IBM scientist Andy Stanford-Clark co-invented MQTT, but IBM has led the successful effort to turn it into an open standard.
At the beginning of this essay, we laid out some scenarios of new connected-car services that could delight car owners. But the car companies, dealers and repair shops stand to reap tremendous benefits from these new technologies as well. In addition to boosting sales by offering customers compelling apps, car companies will be able to team with business partners to jointly market additional products and services to their customers. For their part, dealers and repair shops will be able to harness a wealth of knowledge about their customers’ cars and driving habits to provide predictive maintenance. And when things go wrong, their mechanics will consult with cognitive systems—much like IBM’s Watson—to help them correctly diagnose complex problems.
Open technologies will not only hasten the pace of innovation in the world of connected vehicles, but they’ll make it possible for the entire industry to delight car owners with continuous improvements. In the future, your car will be like you iPhone or Android phone is today: Constantly updatable with the new apps of your choosing.
None of this will come easily unless the automakers, and, indeed, the entire ecosystem of auto-related companies, adopt the open technology standards that can pave the way for the connected cars of the future.