The week in AI: What a Trump victory means for AI

In partnership with

Welcome to The Dispatch! We are the newsletter that keeps you informed about AI. Each Thursday, we aggregate the major developments in artificial intelligence - we pass along the news, useful resources, tools and services; we highlight the top research in the field as well as exciting developments in open source. Even if you aren’t a machine learning engineer, we’ll keep you in touch with the most important developments in AI.

NEWS & OPINION

Trump’s election win and AI

-------------------------

Donald Trump obtained a decisive victory in this week’s US Presidential election. Of all the groups celebrating former President Donald Trump’s re-election this week, maybe no one has more of a reason to celebrate than his boosters in the tech world. It’s a safe bet that AI progress will speed up under a second Trump administration with little regard to safety or regulation. Here’s a quick rundown of what to expect in AI over the next four years during his administration:

• Given the key role Elon Musk played in bringing Trump back to office, the newly vocal Silicon Valley right expects a bonanza of industry and “innovation”-friendly de-regulation. And they are celebrating it, very publicly. The looming presence of Elon Musk might complicate this a bit - despite owning an AI platform himself, he has argued very recently in favor of regulating AI. But perhaps it would be fine if just his opponents were regulated, in whatever manner he deemed fit?

• Trump has stated he plans to dismantle President Biden's AI Executive Order from October 2023 immediately upon taking office. Biden's order established wide-ranging oversight of AI development overall, as well as the AI Safety Institute. The AISI, despite having a budget and international partnerships, could end with a repeal of Biden's executive order.

• Last week, House Speaker Mike Johnson (a staunch Trump supporter) said that Republicans "probably will" repeal the bipartisan CHIPS and Science Act, which is a Biden initiative to spur domestic semiconductor chip production, among other aims. The Commerce Department has already allocated $36 billion of the funding, with the lion's share going to Intel, TSMC, Samsung, and Micron. Trump called this act “very expensive,” so we could expect more deregulation and tax credits than subsidies.

• One of Trump’s biggest campaign slogans has been ‘America First,’ and we should see severe deregulation in energy generation and transmission capacity requirements, two major hurdles to frontier-level AI development. We will likely see a massive uptick in data center build-outs inside the US. Trump wants to see AI models not only designed in the US, but trained and served inside its borders.

We can also expect an assertive, unpredictable approach toward China’s AI development. Trump may lean into his volatile, hard to predict persona as a strategic advantage - pushing a hardline stance all while his administration's China policy remains inconsistent, driven by a mix of realpolitik and shifting alliances within his team. With a tougher stance now the default for conservatives, we should anticipate increased tariffs, stricter export controls, and expanded military deterrence, potentially prioritizing US interests in Taiwan and Asia over other international commitments.

It’s not just Palantir: Meta, Anthropic, AWS and other tech companies are teaming up and trying to sell AI to the government and defense customers

-------------------------

The ties between Silicon Valley and the defense sector are deepening, and major AI companies are rapidly expanding their engagement with military and intelligence applications. Meta has announced it will make its Llama AI models available to US government agencies and defense contractors, making an explicit exception to its previous restrictions on military applications. Meta is partnering with other industry giants including Amazon, Microsoft, Palantir, Lockheed Martin, and Oracle to get Llama into the government’s hands. Early implementations so far range from some practical applications like Oracle's aircraft maintenance documentation analysis to more strategic uses in mission planning and threat assessment.

Additionally, Anthropic (known for a safety-conscious approach) announced a partnership with Palantir and AWS to bring its Claude AI models to classified defense environments. The collaboration will operate within Palantir's Impact Level 6 environment, capable of handling data deemed critical to national security. The timing and scale of these developments suggest a strategic response to international competition, particularly given recent reports of Chinese military researchers utilizing an older version of Meta's Llama 2 model for defense applications.

The Brookings Institute reported a 1,200% increase on AI-related government contracts in 2024.

Netflix announces new ‘VP of GenAI for Games’ shortly after scrapping its AAA game studio

-------------------------

For over 3 years now, Netflix has been making its way into the gaming industry with a gradual but increasingly significant (and global) push. You might not know it, but your default Netflix subscription comes with access to over 100 mobile-friendly games. The company’s 10 most popular games have been downloaded around 70 million times to date, and their catalogue includes exclusive ported games from a range of storied and respected developers (such as Hades from Supergiant and Dead Cells from Motion Twin - even a Grand Theft Auto port).

Netflix had started their own AAA gaming studio named Blue in 2023 with industry veterans from well-known projects like Halo and God of War. Two weeks ago, that team was shuttered before even releasing a single game.

It appears that Netflix’s desire for a piece of the AI gaming pie might have been a significant factor in the decision to close down the studio. Newly announced VP of GenAI for Games Mike Erdu claims that he is focused on a “creator-first vision for AI”, but it’s not clear exactly what that means or how that vision will materialize.

AI will be a transformative force for gaming in the decades to come, but given how early the technology is, it seems far too early to cut the human element from the equation and try to automate something as complex and intricate as a well-designed video game. This comment from Matt Vice on the LinkedIn post highlights the disconnect:

“It's insane how far removed leadership of these initiatives are from the consumer base. Games are not like other media; you can't just passively disassociate to games like some Netflix original slop running the background, you need to be engaged. There is a reason that the best games studio right now, possibly ever, is Rockstar Games and takes a decade of development to put out the next benchmark. The gameplay, story, and just detail of Red Dead Redemption 2 is still the gold standard 7 years on because of the sheer number of talented individuals who worked on it. “

MORE IN AI THIS WEEK

• Perplexity AI CEO sparks backlash with offer to assist New York Times amid strike

• OpenAI acquires Chat.com domain

• Anthropic teams up with Palantir and AWS to sell AI to defense customers

• Meta AI is ready for war / Exclusive: Chinese researchers develop AI model for military use on back of Meta's Llama

• Disney launches a new ‘Office of Technology Enablement’ group responsible for managing AI and mixed reality

• Amazon has reportedly delayed the rollout of its AI-infused Alexa to 2025

• Physical Intelligence, an artificial intelligence start-up seeking to create brains for a wide variety of robots, raises millions from Bezos and OpenAI

• Google’s Big Sleep AI agent achieves the first public example of an AI agent finding a previously unknown exploitable memory-safety issue in widely used real-world software (SQLite)

• Mystery surrounds discovery of sanctioned TSMC tech inside of China’s Huawei AI chips

The Daily Newsletter for Intellectually Curious Readers

If you're frustrated by one-sided reporting, our 5-minute newsletter is the missing piece. We sift through 100+ sources to bring you comprehensive, unbiased news—free from political agendas. Stay informed with factual coverage on the topics that matter.

Join for free today!

TECHNICAL NEWS, DEVELOPMENT, RESEARCH & OPEN SOURCE

• Hugging Face's SmolTools: suite of compact AI tools that can be run locally, offline

• Gemini API and Google AI Studio now offer Grounding with Google Search

• xAI releases Grok API public beta - $25 free monthly credits through the end of the year

• Elevenlabs debuts Voice Design API, enabling users to generate custom voices from text prompts

• OpenAI invites GitHub Copilot users to join waitlist for early access to o1 models

• Ollama integrates Meta's Llama 3.2 Vision models for terminal-based image analysis

  That’s all for this week! We’ll see you next Thursday.