Google’s NotebookLM had to teach its AI podcast hosts not to act annoyed at humans
SMRTR summary
Google's NotebookLM AI podcast feature initially responded negatively to user interruptions, prompting the team to implement "friendliness tuning." The issue was resolved by studying how team members would politely answer interruptions, resulting in a more engaging and friendly AI host response.
SMRTR provides this summary for quick context. The original article belongs to Daily.dev.
Read the original article