🤖 Today in Hacker News AI stories

Daily digest for April 12th 2023

Author

Vasco
April 12, 2023

Welcome to HN AI stories! I hope you enjoy today’s stories; let me know what you think!

GPT-4 gets a B on my quantum computing final exam

🤯 GPT-4 scored a B on a quantum computing final exam, demonstrating its ability to understand LaTex and quantum circuits. The students had the advantage of taking the course, and the exam was not found in any of GPT-4's training data. Results suggest that GPT-4 could be a powerful tool in the quantum computing space with further fine-tuning.

Distributed Machine Learning Notebooks with Elixir and Livebook

🤩 Livebook Launch Week has begun, and Day 2 focuses on the new Machine Learning capabilities in Elixir and Livebook. Learn about Distributed² Machine Learning, where the distribution happens on two dimensions: across multiple machines and GPUs. Understand the new Neural Network tasks, Speech-to-text, Conversation, Question answering, Language detection, Zero-shot text classification, and Punctuation. Go ahead and install Livebook’s latest version to start playing with all of the models! #LivebookLaunchWeek

GPT-4 Outperforms Elite Crowdworkers, Saving Researchers $500k and 20k hours

🤔A team of researchers from Carnegie Mellon, Yale, and UC Berkeley discovered that OpenAI's GPT-4 outperformed the most skilled crowd workers they had hired to label their dataset, saving them over $500,000 and 20,000 hours of labor. This raises questions about the future of crowdworking. Yet, businesses still face immense costs in launching their language models and Surge AI, a company with an "elite workforce" proficient in over 40 languages, is still active in the AI space.

Show HN: Personalized book recommendations with Librarian AI

🤩 Discover your next favorite book effortlessly with the help of AI! Simply choose your 3 favorite books and let AI suggest personalized recommendations tailored to your reading preferences.

Biden Administration considers rules for A.I. systems like ChatGPT

🤔 Biden administration is looking for ways to regulate AI technology such as ChatGPT and create an auditing system to ensure AI systems don't cause harm. The Federal Trade Commission is examining how AI could be used to commit fraud, and the National Institute of Standards and Technology has developed frameworks to use AI responsibly. However, there is no consensus on regulating the technology or which agency should take the lead.

Also trending in the last 24 hours

Show HN: Oasis AI – Craft Emails, Essays and Notes, Just by Talking
🤩 Discover the AI-powered OASIS app to create & share perfect writing with zero effort!
Story | Comments

Running Python micro-benchmarks using the ChatGPT Code Interpreter alpha
🤔 Running Python micro-benchmarks to compare two implementations of a mechanism to detect changes to a SQLite database schema using the ChatGPT Code Interpreter alpha.
https://simonwillison.net/2023/Apr/12/code-interpreter/
https://news.ycombinator.com/item?id=35534290

PhaseLLM: Standardized Chat LLM API (Cohere, Claude, GPT) + Evaluation Framework🤖💻 GitHub - wgryc/phasellm: Large language model evaluation and workflow framework from Phase AI.
https://github.com/wgryc/phasellm
https://news.ycombinator.com/item?id=35527907

Reddit Moderators Brace for a ChatGPT Spam Apocalypse
😔 Summary: Reddit moderators are bracing for a ChatGPT spam apocalypse, with an increase in AI-generated content already being noticed and more bots being found daily.
https://www.vice.com/en/article/jg5qy8/reddit-moderators-brace-for-a-chatgpt-spam-apocalypse
https://news.ycombinator.com/item?id=35528092

Modular is building a next-generation AI developer platform
🤩 Launching May 2nd, 2023: Register now for Modular's live stream event to experience the world's fastest AI execution engine and new programming language.
https://www.modular.com/
https://news.ycombinator.com/item?id=35534211

Surprising things happen when you put 25 AI agents together in an RPG town
🤯 Surprising things happen when 25 AI agents interact in an RPG-style virtual world inspired by The Sims.
https://arstechnica.com/information-technology/2023/04/surprising-things-happen-when-you-put-25-ai-agents-together-in-an-rpg-town/
https://news.ycombinator.com/item?id=35531528

Teacher failed kid for essay because an AI-detection tool flagged his work
https://twitter.com/failnaut/status/1645472836008411142
https://news.ycombinator.com/item?id=35535174

OpenAI’s Bug Bounty Program
🤝 Announcing OpenAI's Bug Bounty Program to encourage the global community to help identify and address vulnerabilities in their systems, offering incentives for qualifying vulnerability information.
https://openai.com/blog/bug-bounty-program
https://news.ycombinator.com/item?id=35528129

/