G7 will approve a code for AI development companies – Kommersant

The G7 countries on Monday, October 30, will approve a special code for companies developing advanced artificial intelligence (AI) systems. This is stated in the G7 document, which the agency cites Reuters.

According to the agency, the governments of the G7 countries are thus seeking to reduce the risks of using artificial intelligence for other purposes. The special code will become a guideline for developments in the field of AI.

The 11-point code “aims to promote safe, trustworthy AI around the world and will provide voluntary guidance to companies developing the most advanced AI systems, including core models and generative AI systems,” the document says.

The Code encourages AI developers to take steps to identify, assess and mitigate risks throughout the development cycle of AI technologies, and to prevent misuse of AI products once they are placed on the market.

In addition, according to the document, development companies must publish reports on the capabilities, limitations and misuse of artificial intelligence systems, as well as invest in robust security controls.

G7 countries (Canada, France, Germany, Italy, Japan, UK and US), as well as EU states, began work on AI regulation in May at a ministerial forum in Hiroshima.

In August it became known that the American company OpenAI, the developer of ChatGPT, testing systems for identifying and removing unwanted information based on artificial intelligence.

Anastasia Larina