A “landmark deal,” as the host called it Prime Minister of the United Kingdom Rishi Sunak, the two-day first international conference on the security of Artificial Intelligence, which concluded today at Bletchley Park outside London, was a success.

In particular, the 28 represented governments and major companies active in the field agreed to safety tests of new AI models before they become widely available.

The organization coordinating this project will be based in Britain and will be called the “Artificial Intelligence Security Institute”.

The conference, Mr. Sunak said, “tipped the scales in favor of humanity,” amid concerns about misuse of advanced technologies to the extent that they could pose an “existential threat” for the human.

It was also agreed that the so-called “Godfather of Artificial Intelligence”, Canadian Joshua Benzio, will lead the drafting of a “State of the Science” report, which will lead to a common understanding from all parties involved about the possibilities and risks of the most advanced forms of Artificial Intelligence.