“Daniel Cocotelo advocates for greater transparency in AI development to solve alignment, arguing that the problem is more likely to be solved by a larger community of 100-500 experts across academia and industry rather than a small, siloed team of 10 experts at the leading lab.”