ai-petition.neocities

Petition Embed

Sharing the petition with your own audience is a great way to give it more exposure! Please feel free to talk about this on your own sites or simply embed the petition.

<iframe src="https://docs.google.com/forms/d/e/1FAIpQLSfZ6LSErE6Z3syNb-d_XgRLEiPMgnJkdG9Y70_qGjVeBqjeEw/viewform?embedded=true" width="640" height="860" frameborder="0" marginheight="0" marginwidth="0">Loading…</iframe>

Timeline

On June 2nd 2023, Kyle Drake posted a blog post about his thoughts on AI. You can view this blog here(archive).

On April 1st 2024, a coding assistant (AI) called "Daria" was introduced to the editor.

On April 21st, I noticed that in the editor there was another coding assistant called Penelope. I asked it "Is this AI?" and it responded with whats in the image below.

Penelope was no longer visible sometime between me sending a message about it and when Jodie finally responded. I have heard nobody else talk about this initial sighting of Penelope so I felt it was important to dig up this chat log.

On July 7th 2024, it was discovered that the code is still in place for Penelope.
Link to Reddit Post(archive)

In the post, Kyle responded to a comment. This account is a moderator for the neocities subreddit so I am pretty sure it's not someone pretending to be him.

Penelope is for sure still in the source code of the editor, these are screenshots that I took for my personal site's blog recently.

If you change "none" to "block" on the line below, penelope's chatbox window appears
<div class="col right-col chat-container" style="display: none">

Attempting to talk with penelope results in a error

If you have firefox and want to check it out yourself, open the source code within the editor, press ctrl + f, then type "penelope". It should be the second one.

Personal thoughts from my blog

I have a hard time justifying the majority of what Kyle said in his blog. Getting caught up on the idea that people compare AI's danger to that of nuclear threat is... interesting albeit missing the point. AI poses lots of danger but getting caught up on just one thing feels wrong. He does mention that AI contributes to climate change but said it in a way that feels like he's trying to discredit safe.ai.

In the reddit comment that Kyle made, he states that he wants to impliment AI in a way that does not alienate people that don't want to use it so it's on hold for now. I don't think this is possible with any form of social media including neocities without alienating the group that does not want it and/or an outright worse experience for everyone. This also ignores the fact that Kyle stated in his blog post that there are climate change concerns with AI. The fact that it's "on hold" implies he still plans on implementing it at some point.

If you want some help to avoid ai webcrawlers you can setup a file called robots.txt. Copy and paste the contents of https://owly.fans/robots.txt into your robots.txt file and you should be good to go. It's not going to stop everything but it will stop well known AI web crawlers that respect robots.txt. Not all of them do, I am currently looking for a better option in regards to neocities sites.

The environmental impact of AI

AI requires a ton of power to function. Most AI is hosted within datacenters which adds a very noticable amount to the amount of water required to get rid of the heat generated. The data centers themselves require a ton of materials and rare metals to make... only for you to need more when something like this happens.

E-waste is big concern with all of this. Some people will try to act like AI is great for the environment but any percieved "good" is massively outweighed by the amount of resources required to run AI.

"Childish" and "Divisive"

The AI sucks image is here to stay. If you read everything on this page, have not been living under a rock for the past few years, and still think it's "childish" or "divisive" then you very likely suck. Also the environmental impact section of this page is here mostly to point out the hypocrisies of Kyle's actions and words, not to steal and rehost the research that has already been done on the topic! Below are links to some external articles explaining the risks of AI.