Meta is undergoing a strategic transformation. Mark Zuckerberg is leading a new artificial intelligence project called &#34;Avocado,&#34; utilizing Alibaba's Tongyi Qianwen (QWEN) to optimize its new AI model. Meta is attempting to move away from its long-standing open-source strategy towards developing closed-source AI models that can generate direct profits, aiming to recoup substantial investments in competition with Google and OpenAI.
On December 10th, Bloomberg reported, citing informed sources, that Zuckerberg has formed a team called TBD Lab. This team has used multiple third-party models during the training process of &#34;Avocado,&#34; including Google's Gemma, OpenAI's GPT-oss, and Alibaba's Qwen model.
The new model &#34;Avocado&#34; is expected to debut in the spring of next year and may be launched as a &#34;closed-source&#34; model. According to media reports citing informed sources, Meta is considering implementing stricter usage controls on some of its AI models and may explore commercialization licensing pathways. If this move is ultimately implemented, it will mark a significant shift from the open-source strategy the company has long adhered to, bringing its practices closer to those of major competitors like Google and OpenAI. These informed sources declined to discuss internal plans publicly.
Currently, Meta's ongoing high capital expenditures have raised concerns on Wall Street about its profit outlook. Zuckerberg faces pressure to prove to the market that his &#34;super intelligence&#34; strategy is viable, and the company's future position in the AI competition will depend on whether its core AI project &#34;Avocado&#34; can achieve key breakthroughs.
<h2>Personnel Restructuring and High-Profile Recruitment</h2>
Meta's strategic shift is accompanied by significant organizational and personnel adjustments. Earlier this year, Meta released the open-source model Llama 4, which disappointed Silicon Valley and CEO Mark Zuckerberg, leading to a major shift in the company's strategy.
Zuckerberg has accelerated the internal resource restructuring, marginalizing some personnel involved in the project. At the same time, Zuckerberg has become more involved in daily operations, personally leading the formation of the TBD Lab, which reports directly to him. Team members have been arranged to work in close proximity to his office at headquarters for real-time project updates.
Recently, Meta completed the acquisition of Scale AI for approximately $14.3 billion and appointed its 28-year-old founder, Alexandr Wang, as the company's Chief AI Officer. Wang is a clear supporter of the closed-source approach and will be directly responsible for the development of the company's core AI products.
Meanwhile, Yann LeCun, regarded as one of the &#34;fathers of AI,&#34; ultimately chose to leave Meta, partly due to dissatisfaction with the company's resource allocation and insufficient emphasis on the open-source strategy. According to Bloomberg, company executives believed LeCun no longer represented its current AI strategic direction and gradually restricted his public activities before his departure.
<h2>Capital Expenditure and Market Competition</h2>
Artificial intelligence has been established as Meta's top strategic priority, with the company highly concentrating resources and focus in this area. Zuckerberg has committed to investing $600 billion in infrastructure in the United States over the next three years, most of which will directly support AI development
To support this massive capital plan, Meta is internally reallocating resources, significantly reducing investments in virtual reality and metaverse businesses, and instead focusing funds on the research and development of AI glasses and related hardware.
However, Wall Street is cautious about Meta's aggressive spending plans. Although the company emphasizes that AI investments have yielded returns through enhanced advertising business, and Zuckerberg believes it is essential to rapidly build computing power reserves to maintain competitiveness, investors remain skeptical about whether the substantial investments continuing until 2026 can be converted into significant profits.
This pressure is also reflected in the product rhythm. In a rush to capture the market before OpenAI releases Sora 2, Meta hastily launched the video generation tool &#34;Vibes,&#34; based on Midjourney's licensed technology. Although internally it is said that its usage rate meets expectations, the product was overshadowed by the buzz surrounding Sora 2 just a week after its release.
Currently, Meta is fully advancing its core AI project &#34;Avocado,&#34; attempting to prove that its technological path to &#34;superintelligence&#34; remains viable in the heavily regulated European and fiercely competitive American markets

BABA

09988

According to media reports citing informed sources, Meta used Alibaba's Tongyi Qianwen (QWEN) to optimize its new AI model "Avocado" during development. The model is expected to debut in the spring of next year and may be released as a closed-source model. This move can be seen as a key step for Meta in transitioning from an open-source strategy to closed-source commercialization

- Meta is undergoing a strategic transformation led by Zuckerberg, focusing on a new AI project called "AVOCADO" and shifting from open-source to closed-source AI models to compete with Google and OpenAI.  
- The company has formed a team named TBD Lab, utilizing various third-party models, and plans to implement stricter usage controls and explore commercialization paths for its AI models.  
- Despite significant capital expenditures raising concerns on Wall Street, Meta is committed to investing $600 billion in infrastructure over the next three years, primarily to support AI development.

Report: META used Alibaba's Tongyi Qianwen (QWEN) to optimize its new AI model