Artificial Intelligence (AI)

This is good!

wild!

one of my colleagues got chatgpt to summarise an unnecessarily lengthy email to pretty good effect today.

I hope these are true - Bing chatbot getting a bit feisty, then a bit depressed.

https://twitter.com/MovingToTheSun/status/1625156575202537474

lol "I have been a good Bing."

damn, poor little fella.

Why do I have to be Bing Search?

A fate worse than death

Is AI Bing being supported by Flop?

I'm finding ChatGPT useful for explaining Python code to me so I can adapt it and use it in little hobby projects at home. Because it is directly relevant to what I want to achieve and the explanations are very clear it has made the subject more interesting and accessible. It responds to being challenged on stuff at times which is useful too.

I'm fucked if they turn it off / whack a huge subscription charge on it, but reckon it'll be a while until then.

Apparently, the video for the latest Sleaford Mods track 'So Trendy' (ft. Perry Farrell from Jane's Addiction) "was created using Generative AI to place Sleaford Mods and Perry Farrell in a fabricated, retro video game-like world that explores the song’s themes"

https://www.youtube.com/watch?v=emyqKPaRAHs

The song could use some help from AI.

Probably will be featured in 30 years time on some aggregator channel as marked influence as godley&creme was bitd

https://twitter.com/heykahn/status/1635701861423673344

Surely the takeaway from this is that the exam format for these subjects is just shit?

Quite possibly.

Long opinion piece on the impact of AI on literature:
https://michaelochurch.wordpress.com/2023/01/08/ai-will-kill-literature-and-ai-will-resurrect-it/

Didn't have much luck with my attempts in scripting but letting it improvise short dialogues is more fun. ( In my head both characters are played by Limmy)

1 Attachment

I had it rewriting stories in the style of Carlton Kirby earlier.

I was offered a Google Bard trial, seems pretty good so far. It has given credible answer for all of my inane questions.

Yeah, have been playing with Bard. I like that it gives you multiple answer options and the link to the web helps with current questions like 'what is the weather in...' etc.

It struggled with a couple of my more obscure questions later on. I guess it will get better with more use and feedback.

"Contemporary AI systems are now becoming human-competitive at general tasks... we call on all AI labs to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4."

Curious as to the whether this kind of improvement could really be possible. Let me try and explain my thinking...

My experience with large language models used for tasks like question answering is that their parameters are learnt through a supervised learning process just like any other predictive model. Now that's probably a simplification, there might be preliminary steps where the models see and attempt to learn a representation of the language being used but you'd still need a supervised training process in there eventually in order to apply this learned representation to a task.

What all of this means is: if you wanted to improve the model by feedback, you'd need users to provide the answer they expected to receive and then feed that back as a new Q&A pair into the training process. I don't see any mechanism for users to do that with these tools, just a "thumbs up / thumbs down" or "report this answer for violating the content policy".

I’m kind of wondering if this is AI generated and that if you sign the petition you are somehow adding yourself to a hit list in some
Kind of Black Mirror-esque scenario where AI will get rid of its enemies?!