Grok 4.7 was built to work for hours. It still fails most of the time

Billy Odell Tucker-Robinson Founder & Host — Banking With Billy Network • Intelligence Network • Data Science • AI Research • World News
Published: 2026-09-21T19:32:58.867Z • Permanent link

Grok, an artificial intelligence (AI) software company, has been making headlines for its ambitious project to develop a coding agent capable of working for hours without errors. Grok 4.7, the latest iteration of this agent, was touted as a significant improvement over its predecessors. However, recent reports suggest that the agent still fails to meet expectations, raising questions about the company's claims and the broader implications for the Global Infrastructure domain.

The story begins with Grok's CEO, Benoît Carré, who has been vocal about his vision for a self-sustaining AI agent that can learn and improve over time. In 2020, Carré announced that Grok 4.7 had reached a milestone in its development, with the agent able to run for 24 hours without errors. However, subsequent reports have raised concerns about the agent's performance, with many experts questioning the accuracy of the company's claims.

In a recent interview, a Grok spokesperson acknowledged that the agent still experiences frequent failures, despite its ability to work for extended periods. The spokesperson attributed the failures to "isolated incidents" and emphasized that the company is committed to improving the agent's performance. However, industry insiders have expressed skepticism about Grok's claims, citing concerns about the lack of transparency and the absence of independent verification.

The failure of Grok 4.7 to meet its promised performance has significant implications for the Global Infrastructure domain. Companies that rely on AI-powered systems, such as financial institutions and technology giants, are increasingly dependent on these systems to manage their operations and make critical decisions. If AI agents like Grok 4.7 are unable to deliver on their promises, it could have far-reaching consequences for the industry.

For example, financial institutions that use AI-powered trading systems could be at risk of significant losses if these systems fail to perform as expected. Similarly, companies that rely on AI-powered systems for customer service and support could experience a decline in customer satisfaction and loyalty. In the worst-case scenario, the failure of AI agents like Grok 4.7 could lead to a loss of confidence in the technology, resulting in a decline in investment and innovation.

Grok 4.7 is not an isolated incident, but rather part of a larger trend in the development of AI-powered systems. In recent years, there has been a surge of investment in AI research and development, with many companies and institutions competing to develop the next generation of AI agents. However, this trend has also led to a proliferation of false claims and exaggerated promises, as companies and researchers seek to attract funding and attention.