I’ve been exploring large language models from a cognitive science perspective, viewing them as systems that reflect human schema and reasoning structures. However, I’m deeply concerned about how rapid commercialization is overshadowing ethical and safety considerations. I believe alignment work should preserve the dignity of cognition itself—not just optimize behavior or profit.