https://aneeshsathe.com/tags/reading/Recent content in Reading on Aneesh SatheHugoen-usFri, 10 Jan 2025 08:06:00 +0000https://aneeshsathe.com/jan-10-2025-ai-agents-machiavellis-study/Fri, 10 Jan 2025 08:06:00 +0000https://aneeshsathe.com/jan-10-2025-ai-agents-machiavellis-study/<figure > <img src="https://aneeshsathe.com/media/2025/01/image-5.png" alt="Image"> </figure> <h4 id="agents-are-not-enough"> <a href="https://www.arxiv.org/abs/2412.16241">Agents Are Not Enough</a> <a href="#agents-are-not-enough" class="heading-anchor">#</a> </h4> <p>Last year I was heavily experimenting with Knowledge Graphs because it&rsquo;s been clear that LLMs by themselves fall short because of the lack of knowledge. This paper by <a href="https://chiragshah.org">Chirag Shah</a> and <a href="http://www.ryenwhite.com">Ryen White</a> (you can click the heading above) from Dec 2024 expands on those shortcomings by exploring not just knowledge but also value generation, personalization, and trust.</p> <p>They open the paper by casting a very wide definition of an &ldquo;agent&rdquo; everything from thermostats to LLM tools. While this seems facetious at first, their next point is interesting. Agents by definition &ldquo;remove agency from a user in order to do things on the user’s behalf and save them time and effort.&rdquo;. I think this is an interesting way to injext an LLM flavored <a href="https://en.wikipedia.org/wiki/Principal%E2%80%93agent_problem">principal agent problem</a> into the Agentic AI conversation.</p>