OMM Redaksiyası
Anthropic CEO Dario Amodei has called for a slower pace in developing increasingly capable artificial intelligence systems, saying more time is needed to assess potential risks and strengthen safety measures.
Amodei stressed that companies should not halt AI development. In his view, they should build additional time into the development process so that safety mechanisms can keep pace with the growing capabilities of AI models.
In an article shared on the X platform, the Anthropic chief executive said that even if companies moderately reduced the pace of development, progress in the AI sector would remain rapid.
Amodei proposed a three-step framework that would allow more time for risk management. One proposal calls for independent third-party evaluators to be permanently placed within leading AI companies.
These evaluators should have access to the relevant tools and internal processes used to identify risks. Amodei also proposed that independent experts verify whether companies are taking the necessary steps to ensure the safety of their models and align them with appropriate objectives.
The call comes amid growing concerns about the misuse of advanced AI systems, including their potential use in cyber operations, surveillance, fraud and weapons development.
The proposal follows a threat intelligence report released by Anthropic on Thursday. The report said several actors had used the company’s Claude AI models for various dangerous activities.
Concerns about the potential consequences of AI also increased this week after the resignation of Anthropic researcher Jacob Coxon. Coxon said people working in the AI sector seriously believed that the technology could destroy all of humanity by the end of the decade.
Amodei urged companies to cooperate voluntarily to establish common standards for managing risks associated with advanced AI. His statement comes as US lawmakers are calling for the adoption of new rules regulating AI systems.
