Good morning. It’s Monday, March 2nd.

On this day in tech history: In 2017, AMD released the Ryzen 7 series CPUs, featuring Zen architecture with up to 8 cores, simultaneous multithreading, and Infinity Fabric interconnects for 16 threads at 3.6 GHz base clocks. This affordable multi-core powerhouse democratized parallel computing, boosting local AI training on tasks like GANs and reinforcement learning without enterprise-grade hardware.

In today’s email:

You read. We listen. Let us know what you think by replying to this email.

This week, xAI revealed it is testing a voice cloning feature within the Grok iOS app, allowing users to create, deploy, and share their own synthetic voice profiles. This move positions audio as a customizable asset for creators and enterprise developers alike.
The new cloning feature leverages xAI’s existing high-speed infrastructure to turn a short recording into a persistent digital identity.

This allows for a consistent brand voice across all AI interfaces. No more generic assistants. Read more.

The AI industry reached a massive inflection point this week as OpenAI secured its position as the primary federal partner hours after the federal government banned Anthropic and labeled it a supply chain risk.

OpenAI signed a Department of War contract to deploy its models on classified networks. While Anthropic maintains that current frontier models are technically unfit for kinetic operations and is challenging its risk designation in court, OpenAI has agreed to all lawful applications.

Strategically, OpenAI negotiated a cloud-only deployment to prevent edge-based integration into autonomous weaponry and will embed engineers within the Pentagon to ensure model integrity. Despite taking the contract, OpenAI has publicly stated it does not believe Anthropic should be designated as a supply chain risk.

This move has sparked a mainstream ‘Cancel ChatGPT’ movement, as critics argue that agreeing to all legal means exposes citizens to data harvesting under the Patriot Act. While CEO Sam Altman claims models will not be used for surveillance, government officials noted the systems would support incidental data collection on Americans. This backlash has triggered a digital migration, propelling Anthropic’s Claude app to the top of the iOS and Android charts.

Simultaneously, OpenAI is navigating internal compliance. The company recently fired an employee for alleged insider trading on prediction markets like Polymarket, where blockchain analysis identified suspicious betting clusters totaling over $300,000 just before product launches. Furthermore, in its legal dispute with Elon Musk, OpenAI has moved to exclude testimony from researcher Stuart Russell, labeling him a doomer despite Altman previously co-signing Russell’s warnings on AI extinction.

OpenAI just raised $110 billion from Amazon, Nvidia, and SoftBank, matching a $111 billion burn forecast. The deal values OpenAI at $730 billion and secures 5GW of Nvidia capacity. Read more.

Connect your email and calendar and Attio instantly builds your CRM. Every contact, every company, every conversation — organized in one place. Then ask it anything. No more digging, no more data entry. Just answers.

Start your free trial →

📄 A Very Big Video Reasoning Suite

📄 The Trinity of Consistency as a Defining Principle for General World Models ​​

📄 From Statics to Dynamics: Physics-Aware Image Editing with Latent Transition Priors

📄 HyTRec: A Hybrid Temporal-Aware Attention Architecture for Long Behavior Sequential Recommendation

📄 From Blind Spots to Gains: Diagnostic-Driven Iterative Training for Large Multimodal Models

Your feedback is valuable. Respond to this email and tell us how you think we could add more value to this newsletter.

Interested in reaching smart readers like you? To become an AI Breakfast sponsor, reply to this email or DM us on 𝕏!