On Wednesday (May 8th) local time, Democratic Senate Majority Leader Chuck Schumer stated that he will propose a framework for legislation on artificial intelligence (AI) in the coming weeks, and Congress will gradually enact relevant legislation instead of introducing a comprehensive regulation at once.
Since the rise of generative AI, the United States has been working on legislation on AI, but progress has been slow. US lawmakers hope to encourage innovation and development while also managing the potential risks brought by AI as much as possible.
Schumer gave a media interview at the AI Summit held in Washington on Wednesday, stating that the AI legislative framework will be based on expert opinions and focus on the thorny issues brought about by AI, from its impact on intellectual property and labor rights to healthcare and the "end of the world scenario".
Schumer said, "This roadmap will be released in the coming weeks, with many good guidelines and ideas on how Congress should act." He added, "You will see different committees working and translating these contents into specific legislation."
US officials have warned that AI may exacerbate the risk of bias and interfere with elections through deep fraud. They are also concerned that powerful AI models may be used by some to manufacture biological weapons or launch cyber attacks.
Schumer emphasized that lawmakers will address these issues one by one. He said, "Some committees have gone further than others, and we won't wait for a massive comprehensive plan that covers all aspects."
Unlike the United States, the European Union is developing a comprehensive regulation on AI, known as the EU Artificial Intelligence Act, and the progress is faster than that of the United States.
In March of this year, the European Parliament approved the Artificial Intelligence Act, which will become the world's first comprehensive legislation on AI. Although the Artificial Intelligence Act still needs formal approval from the European Council in order to become law, this process is expected to be just a formality.
The EU Artificial Intelligence Act will gradually come into effect in the coming years, with specific rules including prohibiting AI from being used for certain purposes, introducing new transparency rules, and requiring risk assessment of AI systems defined as high-risk.