AI News Roundup – Export controls on AI models, $3.3 billion AI datacenter, AI-created content labels, and more

Practices & Technologies

Artificial Intelligence

To help you stay on top of the latest news, our AI practice group has compiled a roundup of the developments we are following.

    • Semafor reports that AI conferences and academic publishers are warning researchers against using tools like OpenAI’s ChatGPT for peer-review, citing concerns regarding academic and intellectual integrity. A spokesperson for Springer Nature, publisher of the scientific journal Nature, said that while AI may be able to assist peer reviewers, “generative AI tools can lack up-to-date knowledge and may produce nonsensical, biased or false information.”
    • The U.S. Commerce Department is considering placing export controls on proprietary AI models, according to a new report from Reuters. The proposed rules would likely prohibit AI companies such as OpenAI, Microsoft, Google DeepMind, and others from selling their AI models or technologies to customers in Russia, China, North Korea, or Iran. Reuters also reports that the export control may only apply to models that exceed a certain computing power threshold (1026 integer or floating-point operations) to train, a statistic used in the AI executive order released by President Biden this past October.
    • Google’s AI subsidiary DeepMind has revealed a new version of AlphaFold, its protein-folding technology, according to The New York Times. In addition to protein folding, AlphaFold3 is capable of predicting protein shapes and the behavior of DNA and RNA within human cells. A bio start-up executive said to the NYT that the new technology could “save months of experimental work and enable research that was previously impossible” and “represents tremendous promise.”
    • President Biden visited Racine, Wisconsin last week to announce a $3.3 billion investment by Microsoft into an AI-focused datacenter on the site of an abandoned Foxconn facility in the city. According to a White House statement, the project will create 2,300 union construction jobs and 2,000 permanent jobs over time. Microsoft stated that the project will also involve “cloud computing and AI infrastructure, the creation of the country’s first manufacturing-focused AI co-innovation lab, and an AI skilling initiative to equip more than 100,000 of [Wisconsin’s] residents with essential AI skills.”
    • TikTok, the popular video-sharing social media app, will start labeling AI-generated content to combat deepfakes and other disinformation on the platform, according to the Associated Press. “Content Credentials,” as the labels are known, will consist of metadata to be attached to TikTok submissions. The company claims to be the first social media platform to support this technology.

AI News Roundup – Google unveils next-gen AI tools, controversy over U.K. scraping rules for AI training, Nvidia investing in AI-driven cars in China, and more

MBHB Partners Mike Borella and Joshua Rich discuss OpenAI’s stance on patents in Law360

AI News Roundup – U.S. announces further Chinese AI chip restrictions, Google’s AI weather prediction model, Trump names AI czar, and more