By AI welfare I specifically mean concern about AI's subjective experience, moral patienthood, etc.
Examples of things that would count for a yes resolution:
A major AI lab employee quits, citing model welfare concerns, and draws mainstream attention to a degree analogous to Jacob Coxon's quitting/tweets
Legislation focused on model welfare is passed or seriously debated/considered
Lab employees/other stakeholders publish a coordinated letter focused on AI welfare, drawing a comparable number of signatures or level of public attention to the Pacing the Frontier letter
Examples of things that would not count on their own:
A lab CEO tweets that they're taking model welfare seriously or building a team dedicated to researching it
Researchers make breakthroughs in model welfare research, but their findings don't get significant mainstream attention
Please ask for further clarification! Since the resolution is partially subjective I won't bet on it.