GENERATIVE AI - AI Dynamics

Yeah, most end-consumers still aren’t great at prompting and might never want to be. End-state of things here is someone’s mom or dad can use a GPT like they download Weather from the App Store.

→ View original post on X — @localghost

12 November 2023

Loving This Midjourney Style for Creative AI

By

@fofrai

–

12 November 2023 18h58

I’m loving this Midjourney style

→ View original post on X — @fofrai

12 November 2023

Pre-trained base LLMs offer more diversity than instruct-tuned models

By

@goodside

–

12 November 2023 18h57

Re: 1, yes — see the parenthesis in the second screenshot. Re: 2, you get more diversity from pre-trained base LLMs than the RLHF/instruct-tuned LLMs most people know. You can also compensate by prompting with random dictionary words for inspiration, à la TinyStories.

→ View original post on X — @goodside

12 November 2023

NVIDIA AI Platform Brings New Capabilities to Microsoft Azure

By

@nvidia

–

12 November 2023 18h52

On November 14-17, join us at @Microsoft Ignite to see how the NVIDIA AI platform is bringing new AI capabilities to Microsoft Azure. Read the blog: https://
nvda.ws/3FUIJTq #GenerativeAI #MSIgnite #NVIDIADGX

→ View original post on X — @nvidia

12 November 2023

Don’t play 20 Questions with LLMs due to superposition of simulacra

By

@goodside

–

12 November 2023 17h59

You shouldn’t play 20 Questions with an LLM. Why? Because LLM behavior less resembles a simulation of a person than a superposition of many such simulacra. In @Nature
, Shanahan, McDonell, and Reynolds eloquently illustrate what this means:

→ View original post on X — @goodside

12 November 2023

Limitations of LLMs in Real-Time Web Data Sourcing

By

@testingcatalog

–

12 November 2023 17h52

Exploring GPT for news sourcing, I found that ChatGPT can’t access Twitter due to robots.txt restrictions. Interesting to note, Elon Musk said Grok from X will have full Twitter data access, offering an edge. However, while Googlebot accesses Twitter, it seems to lag with about a

→ View original post on X — @testingcatalog

12 November 2023

LLM Anthropomorphism Design: Separating Reality From Companionship Claims

By

@johnchavens

–

12 November 2023 17h48

Interesting. But the design biases or practices of LLMs use anthropomorphism by design. So separating myself from those responses means, for instance with terms about loneliness, that I’m supposed to believe ChatGPT is real and can provide “companionship.” Thoughts?

→ View original post on X — @johnchavens

12 November 2023