Claude scores highly in business AI test

Claude Opus 5.5 has delivered a strong performance in a recent comparison of business AI tasks conducted by Dust. This research, published on October 7, provides insights into how different AI models perform on key tasks such as analyzing financial data and creating dashboards. Dust emphasizes that the best choice for an AI model depends on the specific tasks one wants to perform.

In the comparison of Claude Opus 5.5 with other AI models such as GPT-6 Astra, GLM 5.3, and Grok 4.7, Claude emerged as the best for summarizing business information and analyzing financial data. However, when it came to creating dashboards and presentations, no clear winners emerged, indicating that the final choice has more to do with personal preference and design than with functional superiority. Dust advises companies to test various options, including less expensive ones.

A notable finding from the research is that a small number of shared AI assistants account for the majority of usage. Of all messages sent to AI agents on the Dust platform, 65 percent are directed to self-built agents, of which only 1.8 percent have at least fifty users. However, these small-scale agents perform well, responsible for 43.3 percent of all interactions. This indicates efficient knowledge sharing within teams that share the same instructions and information.

For companies, it is crucial to measure the costs of AI relative to the results delivered. A rising cost for AI might indicate higher employee productivity but could also point to inefficiency due to unsuccessful attempts. It is advisable to compare the performance of different models based on internal assignments, monitoring costs, quality, and need for correction. Dust's research offers insights that are specific to their platform and do not make general statements about all companies or AI models.

Read the full article from AI Insider.