Goonatilake's current focus is cryptographic watermarking embedding hidden signals in AI-generated content to identify its source. "Right now with ChatGPT and Gemini you can generate anything and there's really no way to prove if it's human or if it's AI generated." Goonatilake said. "So what I'm working on is some scheme into the model, so if you want to determine if something is AI generated or not you can enter it in and see if it's generated by humans or AI model"