Mark Zuckerberg says Meta reached the AI frontier as it launches Muse Wonder

In short: Meta has released a new AI model called Muse Wonder, and Mark Zuckerberg says the company has caught up with top AI rivals and should not slow down.

What happened

Meta released Muse Wonder on Monday, according to an interview Mark Zuckerberg gave to the Financial Times. Zuckerberg said Meta is now at the “frontier” of AI, meaning it can build systems that are among the most advanced in the field.

Meta says Muse Wonder beats comparable models from OpenAI and Anthropic on some benchmark tests. Benchmarks are like standardized exams that try to measure how well an AI does certain tasks, although they do not always match real life use.

Muse Wonder is meant to power Meta’s Muse app, a personal AI assistant that can use “agents,” or automated helpers that do tasks for you (like a digital assistant that can click buttons and fill forms). The Muse app has been the most downloaded app in the US on Apple and Android app stores since its launch in early September, the FT reported.

Unlike some earlier Meta AI models, Muse Wonder will be closed, with the best versions sold to businesses. Meta said pricing could be as low as 20 cents per million “tokens” (tokens are small chunks of text that AI reads and writes), or up to $200 a month for a premium plan called Muse Wonder Full.

Zuckerberg also rejected calls for an industry-wide slowdown in AI development. He argued that each AI lab should be responsible for building safety “guardrails,” meaning rules and checks designed to prevent harmful behavior.

Why it matters

For everyday users, this is another sign that big tech companies are racing to build AI assistants that can handle more personal and sensitive tasks, like understanding a lease or health information. It also raises practical concerns about privacy and control, especially as AI tools move from answering questions to taking actions on people’s behalf.

Source: Financial Times