Seven AI head companies, including OpenAI and Google, have pledged to add watermarks to artificial intelligence content and open third-party censorship

CTOnews.com July 21, Reuters reported that the U.S. government announced today that it has obtained voluntary commitments from several artificial intelligence head companies to watermark AI-generated content and other measures to help improve the security of the technology.

Amazon, Anthropic, Google, Reflection, Meta, Microsoft and OpenAI all agreed to improve the security and transparency of their systems, including allowing third-party experts to review their models.

In a statement sent to TechRadar, the U.S. government said: "Companies that are developing these emerging technologies have a responsibility to ensure the safety of their products. To realize AI's full potential, the U.S. government encourages the industry to uphold the highest standards to ensure innovation does not come at the expense of Americans 'rights and security. "

First, the seven companies committed to conducting internal and external security testing of AI systems before releasing them to the public, and sharing information with relevant industry players, governments, academia, and the public to help manage AI risks.

The companies will also commit to cybersecurity investments and internal threat controls to "protect proprietary and unpublished model weights" that are critical to driving the operation of data models that generate AI. They also agreed to facilitate third-party investigations and report any security vulnerabilities in the system.

The seven companies also agreed to take steps to boost public trust in their systems, including developing a way to ensure people know when they see AI-generated content, such as watermarks or other measures. The companies will also prioritize the social risks posed by AI models, including racial discrimination and other forms of bias that may result, as well as "protecting privacy."

However, CTOnews.com found from the announcement that these commitments are voluntary measures taken by the relevant enterprises, and there are no penalties for violation, so there is no 100% guarantee of strict implementation.