Sherlock Analyzer - AI That Checks Explanations

Sherlock Analyzer demonstrates how coordinated AI agents can perform structured reasoning instead of simple text generation. Traditional chatbots provide persuasive explanations but do not show how conclusions were reached. Our system breaks reasoning into a controlled pipeline of specialized agents. A question-framing agent defines the investigation scope. A timeline agent constructs a sequence of verified events from primary sources. A claims-analysis agent compares public statements with documented actions. Finally, a synthesis agent produces a concise explanation and assigns a confidence level. The output clearly shows: • what was publicly claimed • what actually happened • where contradictions occur • how confident the conclusion is The demo analyzes the Volkswagen Dieselgate case and automatically detects conflicts between official corporate statements and regulatory evidence. Potential users include investigative journalists, compliance teams, auditors, regulators, and policy researchers who require transparent, auditable AI reasoning.