Doctoral researchers are turning to generative AI to assist in their research. How are they using it, and how can supervisors and candidates have frank discussions about using it responsibly?