Recent outbreaks of avian influenza (AI) have highlighted the necessity to improve existing tests and to develop new methods, in order to detect spread or new outbreaks more quickly, which is vital for the early and successful implementation of control strategies. Conventionally, the time between clinical suspicion and laboratory confirmation of AI can be relatively long because of the logistics of sending samples to laboratories and their capacity for providing high throughput of sensitive and specific assays. Increasingly, new-generation assays based on molecular diagnostics have become available and applied successfully to disease investigation or active surveillance programmes. There has been widespread application of techniques based on the amplification of specific nucleic acid sequences by polymerase chain reaction (PCR), ligase chain reaction and nucleic acid sequence-based amplification (NASBA). The approaches generally offer high specificity and sensitivity. One of the most promising technologies is real-time PCR, which enables amplification of nucleic acids and detection of the amplified products through specific probes at the same time. A rapid diagnosis can be achieved, together with potential for high throughput resulting from process automation. Currently, microarray technology is developing rapidly and has been applied to diagnosis of influenza A virus but generally lacks the necessary sensitivity for direct application to clinical specimens. In addition, these new technologies have been increasingly applied to rapid and reliable subtyping of AI viruses. The application of molecular technologies to the "field" is now potentially an option, through the availability of portable machines for conducting such tests, with prospects for radically changing diagnostic approaches for AI in the future.