🥟Chao-Down: #7 ChatGPT as MD, JD, and MBA? The exam-taking capabilities of LLMs

In the last couple weeks, we’ve seen academics pit ChatGPT and its competitors against several tests of specialized knowledge and professional competency. Large language models have shown that they can generate responses to the quality of a fairly competent law, medical, and business student. With such broad capabilities, are we actually in the age of true artificial intelligence? Or do we need new ways for how we test and assess knowledge?

• ChatGPT passes MBA exam given by a Wharton professor (Paper)

• Performance of ChatGPT on USMLE: Potential for AI-Assisted Medical Education Using Large Language Models (medRxiv)

• Anthropic’s Large Language Model Claude Passes Law and Economics Exam (Link)

-Alex, your resident Chaos Coordinator.

What happened in AI? 📰

Azure releases HIPPA-Compliant Large Language Models for the Healthcare Industry (Twitter)

How ChatGPT Will Destabilize White-Collar Work (The Atlantic)

Microsoft and OpenAI officially extend their partnership (The Official Microsoft Blog)

Always be Learnin’ 📕 📖

• Public reading list for learning topics in robotics (link)

• Dataset Distillation: A Comprehensive Overview (arxiv)

Projects to Keep an Eye On 🛠

InfiniCity: Infinite-Scale City Synthesis (Link)

SymbolicAI - A Neuro-Symbolic Perspective on Large Language Models (Link)

The Latest In AI Research 💡

Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor (arxiv)

Discovering Language Model Behaviors with Model-Written Evaluations (arxiv)

Is ChatGPT a good translator? A preliminary study (link)

The World Outside of AI 🌎

Netflix’s New Chapter (Stratechery)

A condensed version of the Netflix founding story. From its battle with Blockbuster, to its entrance into streaming and its future beyond, Ben Thompson dives into how the stepping down of founder Reed Hastings marks the beginning of a new era for the content and entertainment tech giant. 

One Last Bite 😋