AUSIT Webinar | 20261010 Conversation with Alan Melby on Translation Ethics in the Age of AI

Join us for a conversation with Dr Alan Melby about the labels and standards quietly deciding how AI is used in translation, and about the deeper question on the value of the human being and what a machine is actually doing when it produces text.

Alan Melby started working on machine translation projects in 1970 but, a decade later shifted his focus toward tools for human translators, provoked by an intellectual crisis regarding the nature of human language. He has spent the decades since in the rooms where these questions get settled in practice, chairing the Standards Committee of FIT and FIT North America, serving on the Interpreting SAFE-AI Task Force, working on MQM, and developing standardised labels so that a person receiving a translation can tell whether a qualified human was ever involved. Article 50 of the EU AI Act on transparency obligations commenced on 2 August. How translation gets labelled is being negotiated right now, largely by people most practitioners have never heard of.

In this fireside chat, Michael Broughton will discuss this advocacy with him, and then turn to the argument sitting underneath it. Is there something in human language a machine cannot reach and that the profession should protect?

Expect a wide-ranging discussion of agency and free will, of whether the difference between a person and a machine is one of kind or of degree, and of how our convictions, religious and philosophical alike, shape what we are prepared to believe about a machine that writes.

Meet our Presenter – Dr Alan Melby

Dr Alan Melby is Professor Emeritus of linguistics at Brigham Young University and president of LTAC Global, a nonprofit serving the language community. He is a member of the Standards Committee of FIT and is an ATA-certified French to English translator. He continues working on TBX (for terminology archiving and exchange) and MQM (for analytic translation quality evaluation), but his current work centres on standardised labels for translation output.