Wouldn’t whoever is running the bot just be able to do a quick .replace() before feeding the raw data into it? What would be funnier is to get the bot to generate the string itself and then stop, without it being obiously present in the input. :)

Ryanair now makes you install their app instead of allowing you to just print and scan your ticket at the airport, claiming it’s “better for our environment (gets rid of 300 tonnes of paper annually).” Then you log in into the app and you see there’s an update about your flight, but you don’t see what it’s about. You need to open an update video, which, of course, is a generated video of an avatar reading it out for you. I bet that’s better for the environment than using some of these weird symbols that I was putting into a box and that have now magically appeared on your screen and are making you feel annoyed (in the future for me, but present for you).

This is fucking stupid. And now my day is ruined because I got annoyed thinking about this idea.

Just got here to post this link, I’m too slow!

… but the output is not deterministic as the letter count is sampled from a distribution of possible letter counts for a given word and letter pair; count ~ p(count | word = “blueberry”, letter = ‘b’)!

Thanks for the replies, I guess the “good” was vague on purpose, to see how people interpret it…

This popped up on one of my feeds today and I saved it, can’t remember from where, it’s relevant to the above so sharing here: https://oneproject.org/ai-commons/ (AI Commons: nourishing alternatives to Big Tech monoculture).

They talk about AI for good, at some point they mention how the term is sometimes used just for marketing.

I don’t know if this is the right place to ask, but a friend from the field is wondering if there are any examples of good AI companies out there? With AI not meaning LLM companies. Thanks!

Nuclear security? Should be fine, as long as it doesn’t involve counting letters in any words.

https://xkcd.com/221/

Knowledge distilation is training a smaller model to mimic the outputs of a larger model. You don’t need to use the same training set that was used to train the larger model (the whole internet or whatever they used for chatgpt), but can use a transfer set.

Here’s a reference: Hinton, Geoffrey. “Distilling the Knowledge in a Neural Network.” arXiv preprint arXiv:1503.02531 (2015)., https://arxiv.org/pdf/1503.02531

OpenAI whistleblower found dead in San Francisco apartment.

Thread on r/technology.

edited to add:

From his personal website: When does generative AI qualify for fair use?

Just received a newsletter from Mystery AI Hype Theater 3000 about their book The AI Con - How to Fight Big Tech’s Hype and Create the Future We Want being available for preorder. I’m looking forward to it!

Hi, I’m new here. I mean, I’ve been reading but I haven’t commented before.

I’m sure you all know about how cheap labour is used for labelling data for training “AI” systems, but I just came across this video and wanted to share. Apologies if it has already been posted: Training AI takes heavy toll on Kenyans working for $2 an hour.