Meta executive in charge of the internal AI revamp leaves after just two months.

      The memo that was released addressed a transition rather than a departure. Emily Dalton Smith, the Meta executive who had only been with the company for two months overseeing its AI reorganization efforts, is stepping down. She joined Meta in 2015 and is departing just as her role was expected to gain momentum.

      The timing of her exit is significant. In April, Meta announced that Dalton Smith would spearhead product efforts to consolidate and enhance the company’s internal AI tools, part of a broader initiative designed to make AI agents central to Meta's operations.

      Her division managed Metamate, the company's primary internal enterprise assistant. Now, just two months later, she is reportedly leaving, according to sources familiar with the situation.

      Dalton Smith stated that she would continue working with Andrew Bosworth, Meta’s chief technology officer, until her successor is appointed. Meta has not disclosed who that successor will be or where Dalton Smith plans to go next.

      The company's transformation dubbed ‘AI for work,’ which is the official term for the overhaul, proceeds without the executive who was recently placed in charge of a critical component of it.

      This situation presents a challenging image for a company that has spent the past year asserting that AI is the core principle guiding its future. CEO Mark Zuckerberg has invested considerable resources that clearly indicate intent: Meta has been heavily funding its infrastructure and a Superintelligence Labs unit, which has been partly established through acquisitions.

      In this context, losing the lead for the internal tooling initiative after only two months feels less like a standard reshuffle and more like a hiccup in a plan touted as inevitable.

      Her departure also takes place during a year of upheaval at Meta. In May, the company made the decision to cut 8,000 jobs while reporting record quarterly revenues, a trend that has become common in Big Tech as companies shift their workforce costs into AI investments. High turnover among senior management, whether voluntary or not, complicates that narrative.

      Meta's aspirations for AI agents extend well beyond just internal tools. The company has been expanding Superintelligence Labs through acquisitions, most recently acquiring Moltbook, an AI-agent 'social network' whose founders have joined the lab.

      Additionally, Meta has been moving away from the open-source model that characterized its Llama era, now focusing on a proprietary next-generation model. All these efforts rely on the same requirement that the ‘AI for work’ initiative does: the talent capable of delivering results.

      This strategic shift marks a significant departure from Meta’s recent history. The company spent years establishing Llama as an open alternative to the closed models from OpenAI and Anthropic, making weights available that thousands of developers leveraged.

      The new proprietary model, codenamed ‘Avocado,’ will not be available for outside developers to download and operate freely. Rearranging the company around agents while restricting access to the model is a bold gamble, and it was the responsibility of Dalton Smith’s unit to help implement this internally.

      The internal tooling initiatives she oversaw may lack the allure of the model race but are arguably more impactful for the daily operations of Meta. Tools like Metamate and the consolidated assistant layer are what tens of thousands of Meta employees will utilize in the agent-focused company that Zuckerberg envisions.

      Placing one executive in charge of this and then losing her after two months raises practical concerns about continuity: who will now oversee the roadmap, and will there be delays during the transition?

      It is still unclear what Dalton Smith’s departure means for the timeline of Meta’s internal reorganization. The company has not revealed whether this transition will slow the development of consolidated AI tools, nor has it commented on the reasons for her exit.

      Currently, the most concrete information available is from the memo: the individual chosen by Meta to lead a key aspect of its AI reorganization is leaving just two months after commencing the role.