Tesla is moving to curb its AI costs. The company told staff that it plans to impose a $200 weekly limit on AI spending for employees. Over the past few months, some software engineers at Tesla were consuming thousands of dollars’ worth of tokens each week, according to two people familiar with the usage.

Introducing Devin Security Swarm A more cost effective and accurate way to find security vulnerabilities in complex codebases, based on a new architecture: Agentic MapReduce.

Nvidia is looking to expand access to its leading AI hardware by granting token credits to developers in exchange for a share of their future sales bloomberg.com/news/articles/…

Leopold Aschenbrenner’s Situational Awareness disclosed a 19.9% stake in $SHAZ totaling 5.4M shares.

I will take the under on Meta becoming a hyper scaler all day long. An AI compute provider, yes. Want to become a hyperscaler? 1. You need geographic coverage throughout the world. 2. In each region you need at least 3 availability zones. Each of these need to be connected and planned for nuclear fallout and other events. 3. Your security protocols and business continuity / disaster recovery must be top notch. 4. You must extend to offering a wide variety of services, from identity management to databases to serverless, etc. It is extreme. 5. You must get an ISV organization up and trained, and enable certifications on your infrastructure. 6. You must build out massive GTM organizations. The list goes on and on and on. This is not happening from the likes of Meta. Your best bet at a new form of this is @cloudflare.

$META IS BUILDING A CLOUD BUSINESS TO SELL EXCESS AI COMPUTE

This is basically what Amazon did with AWS, before the age of AI. They had all this server capacity and systems to run the largest retail website in the world, and then were like, “let’s scale it to run other businesses’ websites and applications, too.”

The market reaction of punishing $META because of the increase in CapEx will turn out to be very wrong IMO. In the $META Q325 earnings call, Zuck already explained well that if $META overbuilds their compute infrastructure for internal needs, they can sell it to external parties. We are in one of the biggest compute demand/supply imbalances that seems to be getting worse, not better. $GOOGL GCP CEO Kurian expects demand to be bigger than supply for 10 years. $META's enormous AI compute capabilities will either return ROI in the form of their products (which they are already showing) or/and give great ROI by selling compute to external companies. $META investing in AI data center compute is a big asset, not a cash burn, similar to what metaverse investments were in large. The market is viewing it as a negative, instead of an enormous strategic advantage in a compute-constrained world.