Wulfy<p><span class="h-card" translate="no"><a href="https://dair-community.social/@emilymbender" class="u-url mention" rel="nofollow noopener" target="_blank">@<span>emilymbender</span></a></span> </p><p>This is the paper the journalist references:<br><a href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5166938" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="ellipsis">papers.ssrn.com/sol3/papers.cf</span><span class="invisible">m?abstract_id=5166938</span></a></p><p>The AI antagonists, once again prove that humans do not need <a href="https://infosec.exchange/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> to generate <a href="https://infosec.exchange/tags/bullshit" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>bullshit</span></a>.</p><p>All that this "research" proves is that at the most generous these Professors are ignorant of how <a href="https://infosec.exchange/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>LLM</span></a> models work. Which is the most generous interpretation of the human <a href="https://infosec.exchange/tags/hallucinations" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>hallucinations</span></a> they created. Because the other explanation is less generous, they engage in academic fraud.</p><p>Briefly why:<br>Methodology.<br>They upload a specific full 185 page textbook into the LLM context window (presumably with full permission of the copyright owner), then they ask SUPER SPECIFIC questions, 👉without directing the AI to reference the specific text uploaded👈.</p><p>Once again, be sceptical when humans reference a "study" showing "Bad AI". So far, every time I have seen such a study, it's the human flailing at the controls.<br>Including the "famous" <a href="https://infosec.exchange/tags/BBC" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>BBC</span></a> study.</p><p>Newsflash: "Hammers cause thumb injuries in humans who are 👉untrained👈 in their use"</p><p><a href="https://infosec.exchange/tags/luddites" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>luddites</span></a> <a href="https://infosec.exchange/tags/aislop" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>aislop</span></a> <a href="https://infosec.exchange/tags/spicyautocomplete" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>spicyautocomplete</span></a> <a href="https://infosec.exchange/tags/ethicalai" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ethicalai</span></a></p>