Google's quest for the intelligent cloud
This month, the search giant is publishing a series by its top scientists on how they see the Internet evolving in the next 10 years.
Google is publishing a series of brief articles during September by 10 of its top scientists on how the Internet will evolve in the next 10 years. In the first article, Alfred Spector, a vice president of engineering, and research scientist Franz Och, outline how Google's search engine will evolve over the next decade.
Traditionally, systems that solve complicated problems and queries have been called "intelligent", but compared to earlier approaches in the field of 'artificial intelligence', the path that we foresee has important new elements. First of all, this system will operate on an enormous scale with an unprecedented computational power of millions of computers. It will be used by billions of people and learn from an aggregate of potentially trillions of meaningful interactions per day. It will be engineered iteratively, based on a feedback loop of quick changes, evaluation, and adjustments. And it will be built based on the needs of solving and improving concrete and useful tasks such as finding information, answering questions, performing spoken dialogue, translating text and speech, understanding images and videos, and other tasks as yet undefined. When combined with the creativity, knowledge, and drive inherent in people, this "intelligent cloud" will generate many surprising and significant benefits to mankind.
It appears that Google will stay on the same path that it is on today, taking advantage of Moore's Law in terms of faster and cheaper systems, as well as faster and cheaper storage and networks, and moving from hundreds of thousands of servers to millions working in parallel to deliver more relevant and media rich answers to queries.
Google isn't betting on pure artificial intelligence, replicating all the functions of the human brain, as the way to put more intelligence in the network. The army of Google software engineers will continue to focus on machine-learning and human-engineered relevancy algorithms to unpack trillions of data bits collected from Web crawling and user inputs.
At this point in time Google performs somewhere around 71 percent of Internet searches in the U.S. Google appears destined to increase that share over the next decade unless Microsoft or some company just hatching can come up with a substantially superior search experience.