Elon Musk has declared that Grok 5, xAI's next-generation artificial intelligence model, will be the system that reaches artificial general intelligence (AGI), a milestone that would match or exceed human performance across a broad range of intellectual tasks. This announcement positions Grok 5 beyond the company's planned Grok 4.8 and 4.9 releases, signaling an ambitious roadmap for xAI's AI development.

What Is Grok 4.8 and When Will It Launch?

Grok 4.8 is a 2.5-trillion-parameter model built on xAI's new C++ software stack that is scheduled to finish training this week before entering reinforcement learning, a process that fine-tunes the model's responses based on human feedback. To put this in perspective, a trillion-parameter model contains roughly one trillion individual data points that the AI uses to make predictions and generate responses. This represents a significant scaling up of xAI's computational capabilities.

Musk explained that Grok 4.8 should improve noticeably over current versions, while Grok 4.9 is expected to reach performance levels comparable to Anthropic's Fable and Astra models, which are among the most capable AI systems currently available. However, he acknowledged that xAI still needs to strengthen multimodal performance, which refers to the model's ability to understand and process multiple types of information simultaneously, such as text and images together.

How Does Grok 4.7 Compare to Competitors?

Musk stated that Grok 4.7 is roughly comparable with Anthropic's Opus 5.0, performing stronger in some areas while weaker in others. This positioning places xAI's current model in direct competition with leading AI labs, though the company still has work to do to match the most advanced systems on the market. The comparison suggests that xAI is making steady progress but has not yet achieved clear superiority over established competitors.

When asked directly how close Grok 4.8 would come to AGI, Musk responded simply: "That will be Grok 5." He added that Grok 5 might be "maybe better than anything," before cautioning, "We shall see." Notably, his AGI prediction included no benchmark, timeline, or measurable criteria for what would constitute reaching AGI.

Musk

Steps to Understanding xAI's AI Development Strategy

• Model Scaling: xAI is rapidly increasing the size and capability of its models, moving from Grok 4.7 through 4.8 and 4.9 before attempting AGI with Grok 5, demonstrating a clear progression in computational ambition.
• Infrastructure Investment: The company has scaled quickly through large compute clusters and data tied to X and SpaceX, giving it unique resources and data advantages compared to other AI labs.
• Performance Benchmarking: xAI is positioning its models against competitors like Anthropic's Opus and Fable, using comparative performance as a measure of progress toward more capable systems.

Why Are Safety Concerns Intensifying Alongside AGI Claims?

Musk's AGI announcement arrives on the same day he publicly backed Anthropic CEO Dario Amodei's call for the AI industry to slow capability development due to safety concerns. This apparent contradiction highlights a central tension in frontier AI development: the race to build more powerful systems while managing risks related to control, misuse, job displacement, and concentrated corporate power.

The stakes are particularly high for xAI because the company has access to enormous computational resources and proprietary data from X (formerly Twitter) and SpaceX. This combination of scale and data gives xAI significant advantages in training advanced models, but also raises questions about oversight and safety protocols as the company pursues AGI.

Musk has also argued that advanced AI could sharply expand global economic output and accelerate progress in fields including science, robotics, and software development. This optimistic view contrasts with industry warnings about AI risks, placing Grok 5 at the center of a broader debate about how quickly AI should advance and under what safeguards.

What Happened to xAI's Legal Disputes?

In a separate development, Musk's X Corp and SpaceXAI resolved their antitrust lawsuit against Apple on September 14, 2026. The companies moved to dismiss the lawsuit they had filed in federal court in Texas, though the court filing did not explain the reasons for dismissal or confirm whether a settlement had been reached. The original lawsuit had accused Apple of conspiring to unlawfully monopolize markets for smartphones and generative AI chatbots by exclusively integrating ChatGPT into Apple Intelligence features.

However, X and SpaceXAI stated they will continue to pursue claims against OpenAI, ChatGPT's maker, which remains a defendant in the case. This suggests that while the dispute with Apple has been resolved, the companies maintain their legal challenge to OpenAI's market position and partnerships.

The resolution of the Apple lawsuit removes one legal obstacle as xAI continues its aggressive push toward AGI with Grok 5. Whether Musk's bold claim about reaching artificial general intelligence will materialize remains uncertain, but the company's rapid scaling and substantial resources suggest that xAI intends to be a major player in the race to build the next generation of AI systems.

" }