Гендиректор Google Сундар Пічаї / Depositphotos
Google has updated its «principles for using artificial intelligence» page and removed the promise not to use the technology for weapons development.
The changes were first reported by Bloomberg, which notes that last week the promise was in place (now it is available in its original form in the archive).
«Removing this clause undoes the work that many ethical AI people and activists have been doing at Google, and even more problematic, it means that Google will likely now work to directly deploy technology that can kill people,» Margaret Mitchell, who led Google’s AI ethics team, told the publication.
In response to a request from journalists TechCrunch Google representative provided a link to a new page called «Responsible AI», which contains the following text:
«We believe that companies, governments, and organizations that share our values should work together to create artificial intelligence that protects people, fosters global growth, and supports national security».
Whereas in updated «Principles of AI usage» Google’s website now states that the company will work to «mitigate unintended or harmful consequences of using the technology» and will also align its activities with «widely recognized principles of international law and human rights».
It is worth noting that in recent years, Google’s activities with military organizations of countries (in particular, the company’s contracts for the provision of cloud services to the US and Israeli armies) caused internal sit-ins by employees, resulting in the dismissal of 28 peopleThere were also arrests at demonstrationsA recent Washington Post investigation also stated that Google provided Israel with access to AI technologies for military purposes immediately after the ground invasion of the Gaza Strip.