“AI #30: Dalle-3 and GPT-3.5-Instruct-Turbo” by Zvi

We are about to see what looks like a substantial leap in image models. OpenAI will be integrating Dalle-3 into ChatGPT, the pictures we’ve seen look gorgeous and richly detailed, with the ability to generate pictures to much more complex specifications than existing image models. Before, the rule of thumb was you could get one of each magisteria, but good luck getting two things you want from a given magisteria. Now, perhaps, you can, if you are willing to give up on adult content and images of public figures since OpenAI is (quite understandably) no fun.

We will find out in a few weeks, as it rolls out to ChatGPT+ users.

As usual a bunch of other stuff also happened, including a model danger classification system from Anthropic, OpenAI announcing an outside red teaming squad, a study of AI impact on consultant job performance, some incremental upgrades [...]

---

Outline:

(01:25) Table of Contents

(03:49) Language Models Offer Mundane Utility

(12:05) Language Models Don’t Offer Mundane Utility

(19:25) Level Two Bard

(22:28) Wouldn’t You Prefer a Good Game of Chess?

(24:46) GPT-4 Real This Time

(25:48) Fun with Image Generation

(32:03) Deepfaketown and Botpocalypse Soon

(33:53) Get Involved

(34:40) Introducing

(38:43) In Other AI News

(43:51) Technical Details

(50:27) Quiet Speculations

(56:29) The Quest for Sane Regulations

(01:04:40) The Week in Audio

(01:05:13) Rhetorical Innovation

(01:11:23) No One Would Be So Stupid As To

(01:13:49) Aligning a Smarter Than Human Intelligence is Difficult

(01:18:35) I Didn’t Do It, No One Saw Me Do It, You Can’t Prove Anything

(01:22:36) People Are Worried About AI Killing Everyone

(01:26:00) Other People Are Not As Worried About AI Killing Everyone

(01:27:46) The Lighter Side

---

First published:
September 21st, 2023

Source:
https://www.lesswrong.com/posts/DQTM2whB9i57o2mA3/ai-30-dalle-3-and-gpt-3-5-instruct-turbo

---

Narrated by TYPE III AUDIO.