Dubai-Founded Company Beats Openai, Big Tech Rivals In Enterprise AI Test

(MENAFN- Khaleej Times) A UAE-founded artificial intelligence company has outperformed some of the world's biggest technology firms in a public benchmark designed to test how well AI systems search, retrieve and verify information from enterprise data.

Dubai-based AHOY Labs said its AI platform scored higher than competing products from OpenAI, Amazon, Microsoft, Google and NVIDIA on EnterpriseRAG-Bench, an open-source benchmark that evaluates the accuracy and precision of enterprise AI systems. The company ranked eighth globally among 26 systems assessed.

Recommended For You

Enterprise AI search tools are increasingly being used by businesses to sift through vast volumes of internal information, from emails and chat messages to files, tickets and meeting transcripts. The benchmark tests how effectively AI models can answer questions using a simulated corporate archive containing 500,000 records, including instances where information is scattered across multiple documents, conflicting records exist, or no answer is available.

According to published benchmark results, AHOY Labs achieved an overall score of 74.95 and answered 79.6 per cent of questions correctly. By comparison, OpenAI File Search scored 61.03 and correctly answered 69.8 per cent of questions, while Microsoft's Azure AI Search scored 48.42 and Google's Vertex AI Search scored 41.87.

The largest performance gap was in precision, a measure of how many irrelevant documents an AI system returns when generating an answer. AHOY Labs returned an average of 1.03 irrelevant documents per response, compared with 15.7 for OpenAI File Search.

“We built AHOY to prove that sovereignty and world-class performance are not a trade-off,” said Jamil Shinawi, founder and group chief executive of AHOY.“Outperforming the biggest names in technology, on both accuracy and precision, shows that regional organisations can keep full control of their data without settling for second-best AI.”

The company's platform is designed to run entirely on a customer's own infrastructure and can operate offline, allowing organisations to retain control of sensitive information. AHOY said its internal testing showed the system delivered the same benchmark performance in fully offline environments.

The firm argues that such capabilities are particularly relevant for sectors including government, defence, healthcare and energy, where data sovereignty requirements often limit the use of public cloud services. It said this made its platform the highest-ranked sovereign AI system on the benchmark leaderboard.

EnterpriseRAG-Bench was developed by Onyx and published as an open-source benchmark alongside a research paper. Results are reviewed before being posted to the public leaderboard.

MENAFN06102026000049011007ID1111770398