The conversation around music and artist rights has never been more critical. Clear guidelines are urgently needed to foster progress in academia and industry alike, free from the risks of lawsuits and unethical practices.

Given the significance of this issue, I was honored to participate in a panel titled 'AI Output: To Protect or Not to Protect – That Is the IP Question' at the @WIPO Conversation, organized by the World Intellectual Property Organization – WIPO. Alongside leading legal experts like Linghan Zhang, Desmond Oriakhogba, PhD, and Ygor Valerio, we explored how generative AI is reshaping authors' rights and the future of intellectual property.

Some of the issues discussed included:
- Do AI-generated outputs infringe copyright in works contained in the input training data?
- Should AI-generated outputs benefit from copyright protection?
- How can we properly remunerate and respect artists?
- How should IP respond to AI-generated deepfakes and digital replicas?
- Can we detect which instances a model was trained on?

Being the only technologist in the panel, I was grateful for the opportunity to engage with legal experts and share insights on this quickly evolving landscape.

Our insightful moderator, Brigitte Lindner, highlighted a quote often attributed to Arthur C. Clarke (originally by Norbert Wiener): 'The answer to the machine is in the machine.' This thought-provoking idea sparked reflection: could the solution indeed lie within the machine itself?

At the Audio, Music, and AI Lab (AMAAI) at Singapore University of Technology and Design (SUTD), we’re exploring whether new AI models for music similarity detection could open new paths for distributing royalties based on training set similarity. This research could help transform how creators are compensated in the AI-driven music landscape.