Gradually, then Suddenly: Upon the Threshold

As Ethan said a year ago, it takes 10 hours of monkeying around to get into it. But I think a good AI executive training program can dramatically shorten the window, assuming a growth mindset on the exec’s part. I ran into a retired architect the other day who was bemoaning wanting to work in retirement but he had never learned CAD-CAM! Now he’s tw... See more

First, Claude 3 Opus was excellent at handling large amount of text. I gave it a 300 pages PDF of my sales pages and sales emails for the past few years, with just a one page explaining intro, and asked him to basically make my copywriting for a new product. It was at least 90% as good as what I would have done.

- Then, GTP4o was released, and the i... See more

GPT-4 and 4-Turbo have always been available for free in Microsoft Copilot. 4o is now free in ChatGPT. Claude 3.5 Sonnet is now free in Claude. So there are many that have never subscribed to a premium plan that have been using the best models all along still

Conn McQuinn

Jul 4

First, if we don't have sufficient in-brain skill and knowledge, we won't know what to ask the AI to do or how to interpret the products it creates. Second, if we turn over our thinking to AI, we are doing ourselves harm. I don't mean that in the metaphorical sense. Our brains are like muscles, and the harder they are used, the st... See more

Mickey Schafer

Jul 4

This past semester, a student of mine despaired over analyzing a 20+ question survey with 92 responses. She uploaded the spreadsheet to NotebookLM, a tool we'd used in class, which not only cheerfully assured her it would do the task, but also returned basic R values with short statements about strong relationships. She was deli... See more

But there's just something about the immediacy and interactivity of Claude's "Artifacts" window, combined with the model's less stilted tone that brings it home.

For the most avarege users, these seemingly minor difference matter so much more than raw performance on abstract LLM benchmarks.

Arbituram

Jul 4

Same here; I'm not rubbishing the benchmarks, they have their uses, but Sonnet 3.5 just *feels* so far ahead of anything else, I had to go for a long walk on the beach after using it to contemplate how long humanity has left.

Sonnet 3.5 is able to have thoughtful and nuanced conversations on complex topics and to be proactive in codin... See more