Governments and regulators around the world are becoming increasingly concerned about the risks posed by advanced AI systems.
This week, the Bank of Italy revealed that it has initiated discussions with major AI companies over cybersecurity threats associated with next-generation AI models. Concerns include automated hacking, financial system vulnerabilities and the potential misuse of increasingly capable AI systems.
At the same time, reports suggest U.S. officials are considering frameworks that would require frontier AI developers to provide advance notice before releasing highly capable models.
The debate marks a significant shift from discussions about AI opportunity toward concerns about governance and security.
