Google agrees not to use artificial intelligence for weapons

The company had been criticized for a contract with the Pentagon called Project Maven, designed to distinguish people from objects in videos of drones

        Google announced on Thursday that it will not develop artificial intelligence for weapons or to "cause or directly cause injuries to people," by presenting a set of principles for the use of this technology. "We will continue our work with the government and the armed forces in many other areas," wrote the company's CEO, Sundar Pichai, amid pressure from a contract with the military sector that the technology giant said last week that It would not be renewed.

       Pichai detailed the seven principles on which Google bases the use of artificial intelligence (AI) or any advanced computing that can simulate intelligent human behavior. He said Google uses AI to "help people address urgent problems" such as predicting forest fires, helping farmers, diagnosing diseases or preventing blindness.

"We recognize that such a powerful technology raises equally powerful questions about its use," Pichai said in the blog. "As AI is developed and used it will have a significant impact on society for many years. As leaders in AI, we feel a deep responsibility to do this correctly. "

The CEO ensured that AI programs will be designed for applications that are "socially beneficial" and that "will avoid creating or reinforcing unfair prejudices." The principles include that the projects are "built and tested for safety", that they are "responsible" and "incorporate principles of privacy design".

       Google will avoid the use of any technology "that causes or may cause widespread harm," Pichai wrote. This means avoiding "weapons or other technologies whose main objective or implementation is causing or facilitating injuries to people" and systems "that collect or use information for surveillance that violate international standards".

      The announcement comes amid a growing concern that control of automated or robotic systems may be lost, creating chaos. Several technology companies have agreed on general principles to use artificial intelligence for good, but Google was more precise in its standards.

The company, which is a member of the Artificial Intelligence Alliance, which includes dozens of technology companies committed to the principles of AI, was criticized for a contract with the Pentagon called Project Maven, designed to distinguish people from objects in videos of drones .

      After a petition signed by thousands of employees and facing criticism outside the company, Google announced that the $ 10 million contract would not be renewed, according to press reports.