This blog post details an experiment testing the ability of LLMs (Gemini, ChatGPT, Perplexity) to accurately retrieve and summarize recent blog posts from a specific URL (searchresearch1.blogspot.com). The author found significant issues with hallucinations and inaccuracies, even in models claiming live web access, highlighting the unreliability of LLMs for even simple research tasks.
An analysis of how well different AI systems perform in describing images and answering questions about them. The article compares ChatGPT, Gemini, Llama, and Claude using four images: a hand, a bottle of wine, a piece of pastry, and a flower.
This article explores the use of Google's NotebookLM (NLM) as a tool for research, particularly in analyzing the impact of the Aswan High Dam on schistosomiasis in Egypt. The author details how NLM can be used to create a research assistant-like experience, allowing users to 'have a conversation' with uploaded content to gain insights and answers from the material.
The article discusses the ability of AI systems to interpret images, particularly focusing on the limits and reliability of these systems in answering questions about visual content. The author, Dan Russell, challenges readers to evaluate how well AI can identify objects in provided images and what kinds of questions can be reliably answered by AI.
- "Deep Research" is a new trend in AI-driven research using large language models for multi-step investigations.
- The article compares Deep Research systems, highlighting capabilities and limitations like generating tangential content and handling nonsensical queries.
- Includes systems such as Gemini Advanced 1.5 Pro, OpenAI’s Deep Research, Perplexity’s Deep Research Mode, and You.com’s Research Feature.
This blog post discusses strategies for staying up-to-date on the rapidly evolving field of AI, covering resources, tools, and techniques for tracking news, research, and developments.
Key concept: Setting mental models can help users understand how to interact with products that adapt over time. This chapter covers: Identifying existing mental models Onboarding in stages Planning for co-learning Accounting for user expectations of human-like interaction Key concept: To build effective mental models of AI-powered products, consider what you want people to know about your product before their first use, how to explain its features, and when it will need feedback from them to improve.
- This blog discusses using Large Language Models (LLMs) such as Bard and ChatGPT4 to summarize lengthy texts.
- The author compares the performance of these LLMs on summarizing texts, particularly focusing on the classic gothic tale, Frankenstein by Mary Shelley, and Chapter 10 of their book, The Joy of Search.
- While both Bard and ChatGPT4 show promise in creating decent summaries, there are notable differences between the two, with ChatGPT4 being more adept at handling larger amounts of information.
Syllabus Outline
Outline of the course - HCIAI - Zürich Fall 2023
Course Introduction: Course mechanics, intro to the topic Designing AI with the Mind in Mind Fairness, Accountability, Transparency, and Ethics in AI Building AI with Humans in the Loop Natural Language Interfaces Data Visualization for understanding fairness and bias Rethinking the AI-UX Boundary for Designing Human-AI Experiences AI, Art, Music & Sound Synthesis Image Generative AI Self-Driving Vehicles Humans and Robots Sentience? Consciousness? Coming to grips with these questions Building AI systems for sensemaking Final Presentations part 1 - FINAL PROJECT due today. Video Final Presentations part 2 - FINAL PROJECT (second half) Video Show controls (Ctrl+S) ☰
- Search engines sometimes provide incorrect results due to misinterpretations of the query by the user, highlighting the need for careful evaluation of search results.
- Modifying queries and pursuing multiple avenues of research concurrently improves the effectiveness of searching.
- Advanced search techniques such as site:, filetype:, and double-quoting phrases enhance accuracy.
- Critical thinking skills are essential when analyzing search results to avoid accepting the first answer seen without questioning its validity.
