yahoo news | Meta releases first new AI model since shaking up team

Meta announced the launch of its newest artificial‑intelligence model, **Muse Spark**, claiming it is both smarter and faster than its predecessor, Llama 4. Developed over the past nine months by Meta’s newly formed Superintelligence Labs, Muse Spark is a small, high‑speed model built to reason through complex queries in areas such as science, mathematics, and health. The company says the model will power a range of Meta products—including the AI‑driven features of Facebook, Instagram, WhatsApp, Messenger and upcoming smart‑glass devices—and is currently available only in the United States. This release marks the first entry in a planned “Muse” series, with a next‑generation version already under development.

The debut of Muse Spark follows a period of intense competition in the global AI race, during which Meta’s earlier Llama 4 fell behind rapid advancements from rivals in China, France, and the United States. In response, Meta chief executive Mark Zuckerberg overhauled the company’s AI organization, prompting the departure of long‑time research lead Yann LeCun after a 12‑year tenure. Zuckerberg then launched an aggressive recruitment drive, bringing in talent such as Scale AI co‑founder Alexandr Wang to lead Superintelligence Labs and hiring senior engineers from OpenAI, Anthropic and Google—often at high cost and sometimes directly by the CEO.

The strategic shift reflects Meta’s new focus on “personal superintelligence,” an AI that not only answers questions but truly understands each user’s life context. By moving away from the previously emphasized open‑access model approach exemplified by Llama, Meta aims to embed its AI more deeply into the relationships and everyday interactions that define its platforms, positioning the technology as a core part of users’ digital experiences.

Read more: https://tech.yahoo.com/ai/meta-ai/articles/meta-releases-first-ai-model-205619568.html

#meta #musespark #llama4 #superintelligencelabs #markzuckerberg