Matt Goldenberg comments on Are we certain that gpt-2 and similar algorithms are not self-aware?