Meta's head of AI for work is leaving the company. She joined Meta in 2015 and assumed this role just 2 months ago. We're also seeing more headlines surface about Meta regretting the way in which it carried out its AI talent overhaul and aggressive restructuring. CTO Andrew Bosworth apparently called the approach "atrocious" in ways that "undermined trust."They're actively trying to "boost morale" for employees by adding to travel and office perk budgets, reducing the number of direct reports each manager has and allowing talent to more freely move to different teams if they so choose. Employee experience satisfaction in the Applied AI team was especially low, so this will likely lead to some moving away from that team.This combination of headlines weighed on the stock this week.While I don't love reading these things, I just go back to how much confidence I have in Zuckerberg and the assets he has at his disposal. He is not the type of founder to move slowly. He is the person who moves quickly, breaks things, accepts the inevitable mistakes and fixes them as quickly as possible. This is a leadership style he executes as well as anyone else on the planet, and my confidence is high that he will figure out the AI organizational structure. These are the types of headlines I hope push META back into the low $500s so I can get even more aggressive in accumulating. Why? 

Their core business is incredibly healthy and the noisy trial and error does not shake my conviction in this being an AI winner. They've already shown far more ability to use AI to improve their core business than most peers and are not getting much credit for that at all. We're also at the very beginning of their agent monetization journey, which adds another leg of addressable market to elongate their runway. In addition, they are less vulnerable to a slowdown in this AI supercycle than every other mega-cap. They don't have sizable exposure to OpenAI or Anthropic. And? They aren't reliant on this supercycle continuing to zoom for their growth engine to keep churning. Meta does not buy Nvidia chips and rent capacity to customers for a major source of revenue like Microsoft, Amazon and Alphabet do. Finally, if CapEx needs do slow, the spend will not go to waste. The company has so much core business improvement left to do with this technology and can easily reallocate capacity for core business needs. Their current execution is better than many seem to think and their AI backup plan is better than the consensus view too. We're now paying a below-market multiple to own one of the best, most durable businesses on the planet. I'm going to keep taking advantage if it keeps getting cheaper.