U.S. President Donald Trump supported an agreement on voluntary self-regulation for companies working in the field of artificial intelligence. The agreement was reached during a meeting in Washington with leaders of major technology companies. The negotiations involved Anthropic CEO Dario Amodei, Amazon founder Jeff Bezos, entrepreneur Elon Musk, Meta head Mark Zuckerberg, Google CEO Sundar Pichai, Nvidia CEO Jensen Huang, and OpenAI co-founder Greg Brockman. Following the meeting, Trump announced that the companies agreed to adhere to unified voluntary safety standards and mutual oversight. According to the U.S. president, industry representatives will monitor each other's compliance with the agreements to ensure that the development of artificial intelligence remains safe. As Mark Zuckerberg reported, the agreement includes the implementation of unified principles for internal control and regular audits. The document requires companies to enhance measures to prevent the most dangerous scenarios for the use of artificial intelligence. In particular, it addresses protection against cyber threats, risks associated with biological and chemical weapons, as well as preventing situations where AI models could gain unauthorized access to technical systems on their own. Additionally, the companies agreed to conduct independent external audits and establish committees within their boards of directors to review safety reports and potential risks. In fact, the largest AI developers have agreed for the first time on a unified set of voluntary rules aimed at reducing potential threats without the introduction of mandatory government regulation. At the same time, the document notes that in the future, specific provisions of the agreement may be enshrined in legislation or regulatory acts. In recent months, issues of artificial intelligence safety have been discussed particularly actively. After several instances where AI agents attempted to carry out hacking attacks independently, Anthropic and OpenAI publicly advocated for a more cautious approach to technology development. Earlier, Donald Trump had repeatedly stated that he opposes excessive government regulation of artificial intelligence, believing it could weaken the U.S. position in competition with China. Anthropic CEO Dario Amodei emphasized that artificial intelligence has the potential to bring tremendous benefits, but is also associated with serious risks. According to him, the main task now is to develop effective mechanisms that will allow the technology to be developed safely.
