Meta (NASDAQ:META) has announced a delay in the release of its new AI model, Avocado, shifting the anticipated launch from March to May. This setback brings Meta’s strategic objectives into question, as Avocado was anticipated to compete directly with prominent AI models from industry leaders such as Google (NASDAQ:GOOGL) and OpenAI. The postponement highlights underlying challenges within Meta’s AI development, marking a critical moment for the tech giant. Observers are closely watching to see how Meta navigates these hurdles without a cloud computing operation, in contrast to its competitors. Recent trends indicate rapid growth in AI capabilities, putting additional pressure on Meta to bridge the performance gap and deliver a competitive product.
Meta is reportedly considering licensing Google’s Gemini technology as a temporary measure while Avocado receives further development. If implemented, this tactic would reflect a significant shift from Meta’s established strategy of open-source AI models like Llama. Historically, Meta has advocated for open development, and adopting Gemini would alter the narrative and could affect its reputation as an industry innovator. This shift may suggest deeper questions about Meta’s ability to develop leading AI technology internally.
What’s Behind Avocado’s Performance?
Initial tests revealed Avocado’s shortcomings in areas like logical reasoning and programming, trailing behind competitors’ models such as Google’s Gemini. Despite outperforming Meta’s previous models and some older competing systems, Avocado did not reach its intended benchmark. This performance gap is particularly concerning as Avocado was meant to be a cornerstone of Meta’s upcoming AI strategy. Megan Fu, Meta’s Product Manager, described Avocado as a “major leap forward,” yet it still falls short of expectations as it awaits post-training enhancements.
Is Meta’s Spending Strategy Sustainable?
Meta has committed between $115 billion and $135 billion for AI-centric capital spending by 2026. This investment aims to bolster infrastructure capable of supporting robust AI models. Unlike Google, Amazon (NASDAQ:AMZN), and Microsoft (NASDAQ:MSFT), which offset costs through cloud services, Meta must justify this spending without similar revenue from AI infrastructure. Instead, Meta focuses on integrating AI advancements into Facebook, Instagram, and WhatsApp to improve services and drive revenue through enhanced user engagement.
Mark Zuckerberg has highlighted AI as a key growth driver for Meta, expressing optimism about its potential in recent earnings calls. Avocado was set to kickstart this growth, but its delay and performance issues cast doubts on the immediate strategy. As AI technology proliferates, player positions are recalibrating, with Meta looking to reinforce its foothold.
The delay also reflects broader challenges impacting Meta, as seen with its 2025 struggles to garner developer enthusiasm for releases like Llama 4. Despite these stumbles, Meta’s vision remains expansive, as underscored by continued work on projects like the Watermelon model and the Mango image system. However, questions linger on the timeline and feasibility of these plans.
The world of AI is rapidly evolving, creating a dynamic landscape where strategic pivots are continually required. As Meta recalibrates its efforts, the evolving industry may require them to explore partnerships and technology-sharing to remain relevant. Their next steps will be critical to ensuring their AI models can compete with the most advanced products on the market.
