European lawmakers are seeking to include new product liability laws in the 2024 Artificial Intelligence Act. This proposal has been put forward by four senior lawmakers in the European Parliament, and its goal is to hold companies like OpenAI and Anthropic accountable for the unexpected risks posed by their technologies.
Response to AI safety concerns
This move reflects Brussels' awakening to the existing realities where harms from artificial intelligence are occurring. Lawmakers have requested the Commission to introduce a law called the "Artificial Intelligence Liability Act" to fill existing legal gaps and ensure that no individual or European business remains defenseless against the harms caused by these major providers.
Read more: Gavin Newsom signed an order to expedite oversight of artificial intelligence
Brussels' changing approach to artificial intelligence
Kim van Sparrentak, a member of the Dutch Green Party and one of the leading lawmakers in this effort, stated that technology is moving rapidly and has created a need for a swift update of existing laws. This call has been supported by lawmakers from various parties, including the European People's Party, the Socialists and Democrats group, and the Liberal group, indicating broad political support for this initiative.
This week is also recognized as another important moment in global efforts for AI safety. The topic of AI safety will be raised at the United Nations General Assembly in New York, at the G7 digital ministers' meeting, and in the recent meeting between Xi Jinping, the leader of China, and Donald Trump, the President of the United States.
The proposal from European lawmakers to add new liability laws represents a fundamental shift in the European Union's public stance in recent months. While Brussels previously provided a standard response to incidents caused by AI, it is now focused on accountability, implicitly acknowledging that current laws alone are insufficient.
Cybersecurity experts believe that incidents from last summer, such as breaches using AI models, demonstrate the need for a liability scheme. Additionally, this proposal could help facilitate and reduce compensation costs for victims of AI-related harms.
Read more: Greg Stanton called for the establishment of a government agency to oversee artificial intelligence · Friedrich Merz under pressure from state elections in Germany
