Digital Trust in Financial Reporting: A Review of Deepfakes, Synthetic Evidence, and AI Generated Corporate Communications

Digital trust has become a central concern in financial reporting as artificial intelligence, deepfakes, synthetic evidence, and AI generated corporate communications increasingly shape how information is prepared, verified, disclosed, and interpreted. This review examines the evolving relationship between reporting integrity and digital authenticity, with particular attention to accounting, external audit, internal control, board oversight, and regulatory response. Drawing on peer reviewed journal studies published between 2023 and 2026, the review synthesizes empirical and conceptual evidence from accounting, auditing, information systems, governance, cybersecurity, and artificial intelligence research. The literature indicates that AI can improve reporting efficiency, readability, and analytical capacity, but it can also intensify bias, transparency concerns, and information asymmetry when governance is weak. Deepfakes and synthetic evidence present an immediate threat because audio, video, image, and text forgeries can imitate legitimate records, impersonate authority, and contaminate audit trails. At the same time, AI assisted drafting can enhance clarity and consistency in annual reports, earnings narratives, sustainability disclosures, and investor communications, yet it may also weaken authenticity, signal detachment, and reduce stakeholder trust when disclosure appears overly automated or strategically polished. The review further shows that digital trust is not restored by technology alone. Effective protection requires a coordinated framework built on provenance tracking, content authentication, human oversight, and enterprise level controls. Technical tools such as metadata retention, content credentials, timestamps, blockchain supported traceability, and workflow logging can strengthen verification, but these measures must be complemented by accountable approval structures, professional skepticism, and clear disclosure policies. The evidence also suggests that regulatory responses should distinguish among types of AI use, the timing of disclosure, and the level of materiality involved. Overall, the manuscript argues that digital trust in financial reporting is fundamentally a governance issue. Future research should prioritize multimodal deepfake detection, audit procedures for synthetic evidence, and cross national studies on AI oversight and transparency.