Meta Platforms is intensifying its efforts in the artificial intelligence arena with the development of new models aimed at enhancing its capabilities in image and video generation. The company is reportedly working on an image-and-video model dubbed “Mango” and a text-based model referred to internally as “Avocado,” as noted by The Wall Street Journal.
This initiative comes at a crucial time as Meta grapples with stiff competition in the rapidly evolving AI landscape. The company plans to unveil these new models in the first half of 2026, under the leadership of its chief AI officer, Alexandr Wang, who joined the company earlier this year after founding Scale AI.
Meta aims to leverage its expansive user base across various social media platforms to ensure swift adoption of these technologies, despite an already crowded market that includes offerings from competitors such as Alphabet’s Google, with products like Veo and Nano Banana, and OpenAI’s Sora and Adobe’s Firefly.
Wang has highlighted that Meta’s focus will not only be on enhancing its text model but also on developing advanced world models capable of understanding visual information and executing reasoning, planning, and actions autonomously, without extensive prior training.
The strategic roadmap, unveiled by Wang and Chris Cox, Meta’s chief product officer, indicates a significant restructuring of the company’s AI division. This includes leadership changes and an aggressive recruitment drive targeting top-tier researchers from rival companies. CEO Mark Zuckerberg has initiated a hiring spree aimed at establishing the new unit, Meta Superintelligence Labs (MSL), to spearhead the development of next-generation AI models.
Wang, who leads this newly formed division, was integral to Scale AI, in which Meta holds a 49% stake valued at $14 billion. However, the unit has faced challenges, with some researchers leaving, including the company’s chief AI scientist, Yann LeCun, who is set to launch his own startup.
Despite the hurdles, Meta’s AI assistant has gained traction largely due to the company’s massive user base rather than a distinct edge in product superiority. The assistant’s visibility, embedded in the search bars of Meta’s applications, ensures that it is frequently accessed by users, enhancing its popularity without necessarily reflecting a superior product offering.
This context amplifies the pressure on MSL’s initial projects and models as they enter a competitive race in the AI sector. With Meta’s ambitious plans for AI innovation, much will hinge on how effectively the company can adapt to the landscape and roll out cutting-edge tools that resonate with users in an increasingly saturated market.
See also
Hybrid Reasoning Combines Symbolic and Statistical AI for Enhanced Accuracy and Trustworthiness
IntraGPT Named Most Secure Local AI LLM by AI Magazine Netherlands, Sets New Data Privacy Standard
Latin America’s 2026 Financial Stability at Risk Due to GenAI and Fintech Vulnerabilities
OpenAI’s Sora Generates AI Videos of Celebrities, Raising Ethical Concerns Over Likeness Control
