AI @ Noon | August 1

Bloomberg: US weighs restrictions on China’s access to AI memory chips

+ SK Hynix and Samsung could be affected by the new US rule

+ China’s AI memory chip capability lags behind global rivals


Nvidia and ASML surge after reports of US export control exemptions: Nikkei reports US tech stocks rally on renewed AI enthusiasm.

China wants to start a national internet ID system: NYT reports the government said the proposal would protect online privacy. Critics said it could further concentrate government control over the internet.

EU calls for help with AI: The European Union has kicked off a consultation on rules that will apply to providers of general-purpose AI models under the bloc’s AI Act, its risk-based framework for regulating applications of AI.

More than 100 Silicon Valley investors pledge to support Kamala Harris: The group, including Democratic donors such as Reid Hoffman and Vinod Khosla, has been organized under an effort called VCsForKamala. NYT

TC: Copyright Office tells Congress: ‘Urgent need’ to outlaw AI-powered impersonation

How the Kids Online Safety Act was dragged into a political war: 
The Senate overwhelmingly passed the Kids Online Safety Act on Tuesday, but the legislation faces an uphill battle in the House because of censorship concerns. NYT

Lawmakers want to carve out intimate AI deepfakes from Section 230 immunity: The Intimate Privacy Protection Act would require platforms to have a ‘reasonable process’ to address cyberstalking and digital forgeries. The Verge

Senators introduce long-awaited bill to protect artists from AI deepfakes: Rolling Stone reports the NO FAKES Act would establish protections for when an individual's voice or likeness are used without permission with artificial intelligence.

AI startups ramp up federal lobbying efforts TC

Meta sees AI costs surge as digital ad revenue grows: WSJ reports Meta Platforms said quarterly digital advertising sales increased rapidly even as its investments in artificial intelligence and the metaverse weighed on profits.

Meta sees off fears of AI overspending: The Times reports the social media giant beat Wall Street’s expectations with $39.07 billion revenue.

Gartner released a report suggesting that around a third of generative AI projects in the enterprise will be abandoned after the proof-of-concept phase by year-end 2025. The reasons are many — poor data quality, inadequate risk controls, escalating infrastructure costs and so on.

Microsoft last week previewed its own AI-powered search experience, called Bing generative search.

A CIO canceled a Microsoft AI deal. The reason should worry the entire tech industry. BI

Will AI upend white-collar work? Consider the Hollywood editor. Actors and writers won strict limits on artificial intelligence in last year’s contract negotiations, but editors and artists face a growing challenge. NYT

Power-hungry data centers are gobbling up Texas amid AI boom: A state’s infamously strained electric grid faces a new threat. Bloomberg

Enjoy the ride + plan accordingly.

-Marc