GE’s Bill Run nas a message for anyone who’s confused or doubtful about the company’s new focus on the" Industrial Internet’.The long story made short is that the amount of data being produced by industrial machine” is going to be more than anything you’re ever seen” , and analyzing this data is going to make everyone’s life easier.

Take gas turtbine (涡轮机）at power plants for example. “We’re almost putting a data center on a gas turbine,” Ruh said during a session at GigaOM’s Mobilize conference on Wednesday morning，referring to the hundreds of sensors(传感器）the company is placing on those machines to get data. If those sensors, combined with anyone’s software for managing and analyzing the data, are able to improve efficiency by just one percent, they could save nearly $6 billion a year.

Think about the air travel. Forty-one percent of unplanned downtime for airlines is caused by mechanical errors, Ruh explained, so GE wants to be able to predict when its engines or other airline systems will fail. With this knowledge, carriers can fix problems during scheduled downtime and save everyone's precious time.

Speaking of sensors, Ruh noted just how much potentially predictive data they’re getting. “Using a sensor,” he said,“we could get hundred terabytes(百万兆字节）a day.”

However, he acknowledged that the great effects of the Industrial Internet—cost savings, carbon-footprint reductions and efficiency gains—will come with changes in the employment sector that might not be good for everyone involved. ‘‘Some new kinds of jobs that don’t exist today will get created，”Ruh said. “It takes time and energy for people to get adapted to it.”

Global food demand will double by 2050, according to a new projection, and the farming techniques used to meet that unprecedented(空前的) demand will significantly determine how severe the impact is on the environment, researchers said.

The study researchers warned that meeting the demand for food would clear more land, increase nitrogen(氮) use and significantly add to carbon dioxide and other greenhouse gas emissions.

“Agriculture’s greenhouse gas emissions could double by 2050 if current trends in global food production continue,” study researcher David Tilman, of the University of Minnesota, said in a statement. “This would be a major problem, since global agriculture already accounts for a third of all greenhouse gas emissions.”

The researchers studied various ways in which the increasing food demand could be mentioned. They found that the most environmentally friendly and cost-effective approach would be for more food producers to adopt the nitrogen-efficient “intensive” farming method, which involves the heavy use of labor and the production of more crops per acre.

This approach was shown to be more effective than the “extensive” farming currently practiced by many poor nations, a method that includes clearing more new land to produce more food.

Different farming methods produce significantly different yields, the researchers found — in 2005, the crop yields for the wealthiest nations were more than 300 percent higher than what the poorest nations produced.

According to their analysis of the effects of extensive farming, if poorer nations continue using this method, by 2050 they will have cleared an area larger than the United States, about 2.5 billion acres. However, if wealthy nations help poorer nations to improve food yields by incorporating(吸收) intensive farming practices, that number could be reduced to half a billion acres.

The researchers stress that the environmental effects of meeting future food demand depend on how global agriculture expands and develops.

“Our analyses show that we can save most of the Earth’s remaining ecosystems by helping the poorer nations of the world feed themselves,” Tilman said.

Alexa is a form of artificial intelligence, or AI for short. Many people start their mornings by asking Alexa for the weather forecast or the latest news. A device(设备) that houses Alexa can also play music from your favorite playlists, keep a shopping list, order takeout food, answer questions, send messages and even run “smart” home controls.

Training AI systems to respond to problems with human-like intelligence and learn from their mistakes can take months, or even years. Consider Alexa and similar software, such as Apple’s Siri. To do the tasks its human owners ask, these systems must make sense of and then respond to sentences such as, “Alexa, play my Ed Sheeran playlist” or “Siri, what is the capital of India?”

Computers can’ t understand language as it is spoken by people. So AI researchers must find a way to help humans communicate with computers. The technology used to get computers to “understand” human speech or text is known as natural language processing. By natural language, computer scientists refer to the way people naturally talk or write. To teach an AI system a task like comprehending(理解) a sentence or responding to a person ’s last move in a board game, scientists need to feed it lots of examples.

AlphaGo is an AI system designed by Google that has beaten a human champion, Lee Sedol, at the strategy(策略) board game Go. To train AlphaGo, Google had to show it 30 million Go moves that people had made while playing the game. Then AlphaGo used what it learned to analyze those plays as it played against different versions (版本) of itself. During this practice, the program came up with new moves---ones never seen in games between people.