Google has presented a new policy proposal responding to the U.S. government’s call for an “AI Action Plan.”
 The company is advocating for relaxed copyright rules in AI training and fewer restrictions on AI-related exports.
Google argued that excessive regulations could slow down AI innovation and weaken the country’s global leadership in the field.
One of the key points in Google’s proposal is the use of copyrighted content for AI training.
The company insisted that fair use and text-and-data mining exceptions should allow AI developers to train their models using publicly available data, including copyrighted materials.
Google claimed this approach would support scientific research and innovation without negatively affecting content owners.
The company is currently facing lawsuits from data owners who argue that their copyrighted content was used without permission.
 U.S. courts have yet to decide if fair use laws protect AI developers from such legal challenges.
Google also criticized certain AI-related export controls imposed by the U.S. government.
 It believes these restrictions could hurt American companies by making it harder for them to compete internationally.
 While some firms, such as Microsoft, have expressed confidence in complying with these regulations, Google argues that the rules place an unnecessary burden on U.S. cloud service providers.
Another area of concern for Google is AI regulation at the state level.
The company is urging the federal government to establish a unified AI policy to prevent confusion caused by different laws across various states.
It also wants the government to invest more in AI research and provide access to public datasets that could help advance AI technology.
Additionally, Google is pushing back against AI transparency requirements similar to those being discussed in the European Union.
The company argued that forcing AI developers to disclose too much about their models could expose trade secrets and weaken national security.
