Meta bets billions on ‘muse,’ risks repeating past tech fumbles

Mark Zuckerberg has thrown another hefty sum at artificial intelligence, unveiling ‘Muse,’ its latest closed-source chatbot model, a move designed to aggressively challenge industry leaders like OpenAI and Google. But the strategy raises questions about Meta’s ability to truly translate investment into competitive advantage – a pattern the tech giant has struggled with repeatedly.

A closed door approach – and a 6% stock surge

The announcement followed a massive, multimillion-dollar influx of talent and infrastructure, including sprawling data centers, reflecting Zuckerberg’s desperate attempt to catch up after a period of significant lag. As a result, Muse represents a critical litmus test for the newly formed AI lab, a proving ground for whether this latest gamble will finally pay off.

Following the unveiling, Meta’s stock jumped a notable 6% in New York trading – a momentary reprieve amidst broader concerns about the company’s overall trajectory. The recent downturn highlights a growing anxiety within the tech sector: is Meta becoming the next tabula rasa, a cautionary tale of investment gone awry?

To achieve this, Meta has adopted a surprisingly agile approach to its AI division, granting researchers considerable autonomy and deliberately minimizing its traditional, hierarchical structure. This shift, however, hasn’t immediately translated into superior results. Muse Spark, the initial model, currently lags behind competitors like ChatGPT, Claude, and Gemini in several key areas, according to internal assessments.

‘It’s a first milestone on our journey,’ one team member admitted. ‘We’re still in the early stages of development.’ Despite this acknowledgement, Meta insists that Muse Spark – internally dubbed ‘Avocado’ during its development – represents a fundamental recalibration of the company’s AI strategy, pivoting away from its previous reliance on open-source models like Llama.

Meta intends to maintain the chatbot's accessibility, offering it freely to users while simultaneously exploring potential subscription tiers for future monetization. The model's primary focus will be enhancing Meta’s existing ecosystem – spanning Instagram, Facebook, and WhatsApp. The company is quietly laying off hundreds of roles as it prioritizes investment in its AI initiatives.

But the core question remains: can Meta, with its history of ambitious overreach and failed ventures, finally deliver on its AI ambitions? The Muse project – and the billions invested – will provide a compelling, and likely unsettling, answer.