Human computation - sometimes called crowd computing - is the emerging designation for projects that employ thousands of volunteers to work with computers on small elements of a bigger problem. For instance, the approach often used in studies where huge batches of images need to be individually analyzed by, you know, actual people.
Researchers have long realized that, for certain tasks, the human brain is vastly superior than even the fastest and most sophisticated computer or artificial intelligence. When there's a huge amount of data to be parsed, however, practical problems emerge: Research teams only have so many human brain-hours to work with.
But what if we could get tireless A.I. and efficient crowdsourcing to work together?
A.I. Takes A Stroll Through Amsterdam
That's the concept behind an article published this week in the journal Science, in which researchers propose a new vision for the future of human computation. By more effectively combining computer intelligence and human intelligence, we could set these hybrid computational systems to work on our most intractable planetary problems - like disease, climate change and geopolitical strife.
These kinds of super dilemmas have a name in the world of systems analysis - they're called "wicked problems." They have so many complex moving parts that they simply defy traditional problem-solving techniques.
Researchers hope that, with properly deployed A.I. assistance, human cognitive abilities can be effectively combined into multidimensional collaborative networks. For instance, newly developed techniques can provide A.I. management to crowd-based inputs, so that individual contributions can be processed by a computer, then sent on to the next person for a different task.
Story Time! A.I. Generates Interactive Fiction
In the Science journal article, researchers from the Human Computation Institute (HCI) and Cornell University cite several instances in which this new kind of human computation is already being deployed on a limited basis. The YardMap.org project uses crowd-based input to map local and global conservation efforts.
The new approach could radically speed up medical studies, as well. The authors discuss an Alzheimer's disease research project that recently incorporated human computation and crowd-based microtasking, potentially reducing treatment discovery from decades to just a few years.