Tech giants urge Trump to avoid restricting open-source artificial intelligence
A growing debate over the future of artificial intelligence regulation in the United States has brought major technology companies into the spotlight, with several industry leaders urging the Trump administration to avoid imposing strict restrictions on open-source AI models.
The companies argue that excessive regulation could slow innovation, weaken the competitiveness of the US technology sector and give Chinese AI developers an opportunity to narrow the technological gap with American firms.
An open letter reportedly supported by major companies including Nvidia, Meta, Microsoft and Palantir called on policymakers to maintain a regulatory environment that encourages innovation and allows developers to continue improving open-source AI systems.
The absence of companies such as OpenAI, Anthropic and Google from the list of signatories has also attracted attention. These companies primarily develop closed AI models, which are controlled directly by their creators and are generally less accessible for modification by outside developers.
The debate comes as concerns grow in Washington over the rapid progress of China's artificial intelligence industry. The emergence of new Chinese AI models has intensified competition between the two countries, with the race increasingly viewed as a key issue for national security, economic leadership and technological influence.
US officials have also raised concerns about the possibility that Chinese companies could use advanced American AI technology to accelerate their own development. Michael Kratsios, director of the White House Office of Science and Technology Policy, has accused Chinese developers of using a technique known as "distillation," in which the outputs of an advanced AI model are used to help train another system.
US officials have stressed the importance of protecting American intellectual property and preventing the unauthorized use of advanced technologies. Treasury Secretary Scott Bessent has also indicated that sanctions could be considered if Chinese AI companies are found to have violated US intellectual property rights.
At the same time, the US administration has maintained that it does not oppose the development of open-source AI models domestically. The focus, officials say, should be on protecting American innovation while allowing the technology sector to remain competitive.
Nvidia CEO Jensen Huang has been among the prominent voices defending open-source AI. He argues that open models can broaden access to artificial intelligence, reduce costs for developers and encourage faster innovation. Huang has also described the rapid development of Chinese AI technology as evidence of the intensity of global competition.
The Nvidia chief has repeatedly urged Washington to allow US companies to continue competing in the Chinese market, arguing that maintaining access to international markets can strengthen American technological leadership and encourage continued innovation.
Supporters of open-source AI say these models offer developers and businesses greater flexibility while reducing dependence on expensive proprietary systems. They also argue that greater transparency allows researchers to identify vulnerabilities and improve the security of AI technologies more quickly.
Critics, however, warn that open models can potentially be modified and misused by third parties to develop harmful applications. Once such systems are publicly released, controlling their subsequent use can be difficult.
The debate reflects a broader challenge facing governments around the world: how to regulate rapidly advancing artificial intelligence without undermining innovation. As the United States and China compete for leadership in the sector, policymakers will have to balance national security concerns, intellectual property protection, economic competitiveness and the potential benefits of open technology.
The outcome of this debate could have lasting consequences for the global AI industry, shaping how artificial intelligence is developed, distributed and regulated in the years ahead.
-
13:19
-
12:15
-
08:15
-
17:02
-
16:00
-
11:45
-
11:03
-
10:45
-
18:30