Overview
Below is this week’s tracker of the latest legal and regulatory developments in the United States and in the EU. Sign up here to ensure you do not miss an update.
AI Intellectual Property Update:
- Reddit struck a deal with Google to make its content available for training the search engine’s AI models. The contract with Google is worth about $60 million per year.
- Google released Gemma, a range of “lightweight” open AI models designed for text generation and other language tasks. Gemma is built from the same research and technology used to create the Gemini models and uses automated techniques to filter out certain personal information and other sensitive data from training sets.
AI Policy Update—U.S.:
- House Speaker Mike Johnson and Democratic leader Hakeem Jeffries announced a bipartisan 24-member Task Force on Artificial Intelligence. Led by Rep. Jay Obernolte (R-CA) and Rep. Ted Lieu (D-CA), the task force is charged with delivering a report on what Congress should do to set new regulatory standards for AI and spur investment in the technology.
- The Department of Commerce’s National Telecommunications and Information Administration is seeking comments related to “open-weight” AI models. These “open-weight” models are more accessible to a variety of stakeholders, such as smaller companies and researchers, but their customizable nature also makes it easier to remove safeguards and could have important national security outcomes.
- The Connecticut legislature introduced a bill against algorithmic discrimination. The legislation would require developers of high-risk AI systems to “use reasonable care to protect consumers from any known or reasonably foreseeable risks of algorithmic discrimination.” It also prohibits dissemination of certain synthetic images.
AI Policy Update—European Union:
- On February 21, 2024, the European Commission officially launched its AI Office which aims to play an important role in implementing the EU AI Act, especially for general purpose AI models, to foster the development and use of trustworthy AI and to facilitate international cooperation.
- The European Commission published a speech given by Margrethe Vestager, Executive Vice President and Commissioner in charge of competition policy, in which she discussed how to avoid potential harms from big tech having a monopoly on AI. She indicated that large tech companies should expect more scrutiny of their operations.
- The European Patent Office (EPO) published a preview of the amended EPO Guidelines for Examination of European applications and patents in accordance with the European Patent Convention and its Implementing Regulations. The amended Guidelines include, among others, several updates on the assessment of inventive step and sufficiency for AI inventions and clarifications on the topic of AI inventorship. The Guidelines will enter into force on March 1, 2024.
- On Thursday February 22, 2024, Members of the Legal Affairs Committee (JURI) and of the Committee on Internal Market and Consumer Protection (IMCO) of the European Parliament will vote on the provisional political agreement on the Proposal for an EU Directive on Liability for defective products (known also as “Proposed New Product Liability Directive”). The Proposed New Product Liability Directive will bring the EU’s product liability regime up to date with developments related to new technologies, including AI.
AI Policy Update—International:
- The United Kingdom and Australia signed a Memorandum of Understanding that is intended to foster collaboration on online safety proposals. The initial focus and scope of the MoU will be on harmful online behavior, user privacy and freedom of expression, lawfully access to data, and the impact of new, emerging and rapidly evolving technologies like generative AI, among other things.
