Meta AI with Llama 3 405B and Custom Characters Sharpens the Strategy
Summary
Meta is intensifying its focus on AI, driven by the need to maintain user attention for its ad-based business model and to counter rivals like OpenAI, Microsoft, and Google. The company's Meta AI assistant, initially powered by Llama 2 70B, now offers a limited-use option with the more powerful Llama 3 405B model, aiming for performance parity with competitors. Meta CEO Mark Zuckerberg stated the goal is for Meta AI to become the "most used AI assistant" by the end of 2024, leveraging its extensive distribution through Facebook, Instagram, and WhatsApp. The company also launched Meta AI Studio, enabling users and creators to build custom AI agents for engagement, with a particular focus on business messaging. However, Meta's previous celebrity AI personas, launched in September 2023, largely failed to gain user traction, indicating challenges in consumer AI adoption for entertainment.
Key takeaway
For AI Engineers and product managers focused on consumer engagement, Meta's aggressive push with Llama 3 and Meta AI Studio signals a shift towards embedding AI directly into social platforms. Your teams should explore integrating AI assistants into existing high-distribution applications, focusing on utility and connectivity use cases rather than pure entertainment, to capture and retain user attention against a competitive landscape.
Key insights
Meta's AI strategy prioritizes user attention and competitive positioning through free AI assistants and foundation models.
Principles
- User attention drives monetization.
- Free offerings disrupt rival revenue.
- Distribution is key for AI adoption.
Method
Meta is deploying frontier-quality LLMs (Llama 3 405B) and enabling custom AI creation via Meta AI Studio to integrate AI deeply into its social platforms.
In practice
- Utilize Meta AI for information search.
- Experiment with Meta AI Studio for custom agents.
Topics
- Meta AI
- Llama 3
- AI Assistants
- Generative AI
- User Attention Economy
Best for: AI Engineer, Machine Learning Engineer, NLP Engineer, AI Product Manager, Executive, Investor
Related on AIssential
Editorial summary, takeaway, and curation by AIssential. Original article published by Synthedia.