Multi-agent systems built on large language models (LLMs) promise enhanced problem-solving through distributed information integration, but may also replicate collective reasoning failures observed in human groups. Yet the absence of a theory-grounded benchmark makes it difficult to systematically evaluate and improve such reasoning. We introduce HiddenBench, the first benchmark for evaluating collective reasoning in multi-agent LLMs. It builds on the Hidden Profile paradigm from social psychology, where individuals each hold asymmetric pieces of information and must communicate to reach the correct decision. To ground the benchmark, we formalize the paradigm with custom tasks and show that GPT-4.1 groups fail to integrate distributed knowledge, exhibiting human-like collective reasoning failures that persist even with varied prompting strategies. We then construct the full benchmark, spanning 65 tasks drawn from custom designs, prior human studies, and automatic generation. Evaluating 15 LLMs across four model families, HiddenBench exposes persistent limitations while also providing comparative insights: some models (e.g., Gemini-2.5-Flash/Pro) achieve higher performance, yet scale and reasoning are not reliable indicators of stronger collective reasoning. Our work delivers the first reproducible benchmark for collective reasoning in multi-agent LLMs, offering diagnostic insight and a foundation for future research on artificial collective intelligence.

Collaboration in multi-agent autonomous systems is critical to increase performance while ensuring safety. However, due to heterogeneity of their features in, e.g., perception qualities, some autonomous systems have to be considered more trustworthy than others when contributing to collaboratively build a common environmental model, especially under uncertainty. In this paper, we introduce the idea of increasing the reliability of autonomous systems by relying on collective intelligence. We borrow concepts from social epistemology to exploit individual characteristics of autonomous systems, and define and formalize at design rules for collective reasoning to achieve collaboratively increased safety, trustworthiness and good decision making.