An AI safety specialist confronts fears about the future of large language models...<br/><a href="https://www.science.org/doi/10.1126/science.adu1567" class="button purchase" rel="nofollow noopener" target="_blank">Read More</a>