Elon Musk's xAI claims its Grok Imagine video generation tool is profitable, contrasting sharply with OpenAI's decision to shut down Sora after losing approximately $1 million per day. The shutdown of OpenAI's video platform, announced just six months after its December 2024 launch, reveals a critical divide in how AI companies approach expensive generative video technology. While Sora struggled with high compute costs and declining user engagement, Grok Imagine appears to be charting a different financial path .

What Went Wrong With OpenAI's Sora?

OpenAI's Sora launched as a standalone text-to-video generation model that quickly rose in popularity, topping app charts and amassing millions of downloads. However, the economics didn't hold up. The platform will shut down by April 26, 2026, with its developer API discontinued by September of that year .

The reasons behind this strategic withdrawal are primarily financial. Investigations by multiple outlets revealed that Sora's operating costs were extremely high, with the platform losing around $780,000 to $1 million per day. User engagement also sharply decreased from its initial launch surge, with active users dropping by roughly half. That combination of high compute demands and declining adoption made it difficult for OpenAI to justify continuing to operate the platform at scale, especially as the company prepares for a potential initial public offering and seeks more stable revenue streams .

A failed partnership with The Walt Disney Company, which reportedly had plans to invest up to $1 billion in licensing characters for Sora, also collapsed amid the shutdown. Some industry observers believe Sora didn't shut down because people stopped caring, but because it was simply too expensive to keep alive .

How Is Grok Imagine Actually Making Money?

Musk responded to Sora's shutdown by emphasizing that xAI's Grok Imagine operates differently. In his statement, he argued that video data is critical for achieving artificial general intelligence, or AGI, a hypothetical AI system with human-level intelligence across all domains. He also claimed, succinctly, that xAI's video tool is "positive gross margin" and "not a money loser," a direct contrast with the narrative around Sora's financial challenges .

Musk

The phrase "positive gross margin" matters significantly. In simple terms, Musk appears to mean the tool is bringing in more money than it costs to operate directly. That does not necessarily mean it is fully profitable once wider expenses such as research, engineering, and long-term AI development are included. However, it suggests xAI has found a more efficient operational model than OpenAI's approach .

Pricing estimates show users may be paying around $2.30 to $3.10 per minute for certain video features, which gives one of the first real hints at how xAI could be making money from the tool. To put this in perspective, the cost to produce one minute of AI video varies dramatically across platforms. Grok Imagine reportedly costs approximately $4.20 per minute to produce, compared to Google Veo 3.1 at $12 per minute and OpenAI's Sora at $30 per minute. This efficiency gap suggests xAI has optimized its infrastructure more effectively than competitors .

How to Evaluate AI Video Tools for Your Needs

• Cost Per Output: Compare the actual cost to generate one minute of video across platforms. Grok Imagine's $4.20 per minute is significantly lower than competitors, making it more accessible for budget-conscious projects.
• Quality Benchmarks: Check independent rankings and leaderboards where video tools compete. Grok Imagine has reportedly performed better than some rival video tools, including products linked to Google and OpenAI's now-closed Sora.
• Financial Sustainability: Assess whether the platform operator has disclosed positive margins or profitability. Tools with sustainable economics are less likely to shut down unexpectedly, as happened with Sora.
• User Engagement Trends: Monitor whether active user bases are growing or declining. Sora's user base dropped by roughly half before shutdown, a warning sign of platform viability.

Grok Imagine is also scheduled to receive a major upgrade in the first half of April, with reports suggesting it will continue to improve its competitive position. The tool's strong performance on leaderboards, combined with its lower operational costs, positions it as a viable alternative to more expensive competitors .

However, strong performance does not automatically translate to huge profits. AI video is inherently expensive to run, and the costs of computing power, training, and infrastructure can accumulate quickly. xAI has not publicly shared full revenue or profit figures, so the bigger financial picture remains unclear. What is clear is that Musk's assertion about positive gross margin suggests xAI has found a more efficient path than OpenAI's Sora, at least in the near term .

The contrast between Grok Imagine's claimed profitability and Sora's shutdown illustrates a fundamental challenge in the AI industry: building generative video tools that are both impressive and economically sustainable. As AI companies race to develop cutting-edge capabilities, the ability to operate these tools at reasonable cost may become as important as the quality of the output itself.