Companies hope that EU regulations will ensure the continued development of AI in the region.
Several tech and artificial intelligence companies, including GitHub and HuggingFace, have written to European policymakers asking them to provide more support for the open source development of artificial intelligence (AI) models, The Verge reported July 26.
In the document, which contains a list of recommendations to the European Parliament, the companies asked for clearer definitions, among other things.
Companies co-signing the document include Creative Commons, EleutherAI, LAION, and Open Future.
The companies are taking action to ensure that the AI legislation currently being developed by EU policymakers meets the best standards for ensuring the development of AI.
The European Union is one of the regions that is taking AI regulation seriously, and its draft AI bill has been released, but the draft has already attracted criticism from many quarters.
Most of the criticism has focused on the excessive focus on the application layer and the lack of a concise definition of AI technology. Not surprisingly, these companies have proposed clearer definitions.
Other suggestions include distinguishing between enthusiasts and researchers working on open source AI models and those working on commercial, profit-making ones, setting requirements for different models, and regulating actual testing of AI projects.
The bill bans all forms of real-world AI testing. But the companies argue that such an outright ban would only affect research and development. They also want to share AI tools on open source libraries, free from regulation.
If passed, the bill could set a precedent for laws regulating AI elsewhere, which is why AI companies have been lobbying aggressively to ensure the rules aren’t too restrictive and don’t hobble the industry’s growth.
While EU policymakers are already working on laws to regulate AI, regulation in the U.S. is currently in the early stages. Lawmakers hope to avoid the current regulatory ambiguity that continues to plague cryptocurrencies in their handling of AI.
Top AI companies including Google, Meta, and OpenAI recently agreed to implement watermarks on AI-created content as part of a voluntary effort to improve safety.