Release delayed again, Meta continues to struggle in the "AI quagmire".
```
Meta has repeatedly delayed releasing its new AI model to developers.
On June 4, The Wall Street Journal reported that the API (application programming interface) for Meta’s latest AI model, Muse Spark, has been delayed multiple times, and as of Tuesday this week, there is still no confirmed launch date.
It has been nearly two months since Meta’s Chief AI Officer Alexandr Wang publicly promised that the API was “coming soon.”
After the news broke, a Meta spokesperson responded on Wednesday, stating that the company is testing the API with partners and plans to launch it this month.
Muse Spark is Meta’s first closed-source AI model, and the API is the only channel for developers to access it. The delay means the window for commercial monetization continues to shrink.
Broken Promises, Delayed Twice
According to reports, Meta originally planned to launch the API simultaneously with Muse Spark in April.
Two days after the model went online, Alexandr Wang posted on X:
The Muse Spark API is coming soon! Developers are extremely enthusiastic about this, and we’re thrilled — stay tuned!
However, the API was not released as scheduled. Citing informed sources, the report stated that the first delay (from April to May) was due to program vulnerabilities found in testing, as well as the need to build more infrastructure.
Later, the release was postponed again to June. So far, apart from a few authorized third-party evaluation agencies conducting tests, most developers still cannot access the model themselves.
Typically, AI companies launch accompanying APIs when releasing new models or within a few weeks to maximize influence among developers.
Continuous delays have not only weakened Muse Spark’s market hype but also reignited doubts about Meta’s ability to efficiently execute its AI commercialization roadmap.
Stalled AI Development, Not the First Time
The delay of the Muse Spark API is not an isolated case.
Wallstreetcn previously mentioned that last year, Meta delayed the release of another AI model, Behemoth, because engineers could not significantly improve its performance. The model ultimately was never officially launched.
Afterwards, Meta massively recruited AI talent and restructured its teams, appointing Alexandr Wang to lead the newly established Meta Superintelligence Labs (MSL).
A mysterious department under MSL called TBD Lab subsequently developed Muse Spark.
Muse Spark is also Meta’s first closed-source AI model. Previously, all models released by Meta were open-source and freely downloadable by developers.
According to internal Meta benchmark tests, Muse Spark’s performance is comparable to OpenAI and Anthropic’s flagship models, and it significantly leads xAI’s Grok in most tests.
But before the API is officially open, these figures are just numbers for external developers and are hard to translate into real market appeal.
Massive Investment, Surging Monetization Pressure
Meta plans to spend up to $145 billion on capital expenditures this year, mainly for building AI infrastructure. The company’s goal is to create personal and business AI agents for its 3.5 billion daily active users.
However, Wall Street is not buying this “money-burning plan.” In April this year, after Meta announced further raised spending forecasts, its stock price once dropped more than 5% in after-hours trading.
CEO Zuckerberg said in an investor call that converting surplus computing power into cloud computing business is “absolutely under consideration.”
Zuckerberg also revealed that every week, companies proactively seek Meta’s API services, but he did not give any specific timeline for the Muse Spark API launch.
Facing market pressure, Meta has begun announcing monetization strategies.
Last week, the company announced new subscription services for Instagram, WhatsApp, and Facebook, and said it would pilot paid subscriptions for its AI chatbot, Meta AI.
Risk Warning and DisclaimerThe market involves risks; investments should be made cautiously. This article does not constitute personal investment advice and does not take into account the specific investment goals, financial situations, or needs of individual users. Users should consider whether any opinions, perspectives, or conclusions in this article suit their circumstances. Any investment based on this article is at your own risk. ```