Meta’s Muse Spark AI Model Fails to Compete with OpenAI, Google, and Anthropic, Admits Executive

Meta’s Muse Spark AI Model Struggles to Keep Up with Industry Leaders

Meta’s highly anticipated Muse Spark AI model, developed under the company’s costly Superintelligence Labs initiative, has fallen short of expectations. An executive confirmed to Bloomberg that Muse Spark cannot compete with leading models such as OpenAI’s ChatGPT, Anthropic’s Claude, or Google’s Gemini.

Investor Enthusiasm vs. Reality

Following the announcement of Muse Spark, Meta’s shares surged by 6%, reflecting investor optimism. However, the company’s own assessment suggests the model is not yet on par with its competitors. In a blog post, Meta described Muse Spark as “an early data point on our trajectory,” while acknowledging the existence of larger models in development.

Why Release an Underperforming Model?

The timing and purpose of Muse Spark’s release remain unclear. Analysts speculate that Meta may be attempting to secure a position in the competitive AI landscape, as Wired suggested, by simply “getting its foot in the door.” This move comes as Meta faces challenges in maintaining relevance amid rapid industry changes.

Recent headlines have focused on Meta’s legal troubles, including a court ruling holding the company liable for harming underage users through social media addiction. These issues have overshadowed its AI initiatives, raising further questions about its strategic priorities.

Controversial Training Practices

Meta’s decision to train Muse Spark on third-party open-source models, including one developed by Alibaba, has drawn criticism. The practice, known as “distillation,” involves using a more capable “parent” model to train a “student” model. This approach has sparked controversy in the past, particularly given Meta’s history with its Llama open-source models.

Last year, Meta faced allegations of manipulating benchmark results to exaggerate the capabilities of its Llama 4 model. Former Meta AI head Yann LeCun, who left the company amid the controversy, admitted in January that the results were “fudged a little bit.” He stated, “Mark was really upset and basically lost confidence in everyone who was involved in this. And so basically sidelined the entire GenAI organization. A lot of people have left, a lot of people who haven’t yet left will leave.”

Meta’s Desperate Push for AI Relevance

Following the backlash over its Llama models, Meta embarked on an aggressive hiring spree, reportedly spending hundreds of millions of dollars to assemble a top-tier AI team for its Superintelligence Labs. The goal was to restore Meta’s standing in the AI race, which has been dominated by OpenAI, Anthropic, and Google.

These competitors have since expanded their offerings, targeting enterprise customers and developing advanced coding assistants. While some early benchmarks have portrayed Muse Spark in a more favorable light, the overall assessment suggests Meta still has a long way to go to achieve its ambition of competing at the highest level.

The new model won’t be able to keep up with OpenAI’s ChatGPT, Anthropic’s Claude or Google’s Gemini.