Studying AI safety

Besides all the things I talk about AI capabilities, AI safety study must be done too.

I think Drupal AI might contribute to the solution.

My intuition is: people have to use AI capabilities in order to be at the same strength as AI, instead of ignoring it, and thus being weaker, and so, being completely controlled and taken over by AI.

This follows on the similar logic to Elon Musk's Neurolink project to solve AI problem.

Also I am inspired by Herrari's arguments (recent video with 1.5 million views: AI and the future of humanity | Yuval Noah Harari at the Frontiers Forum)

AI safety community following a comprehensive approach (recently found) https://www.gatoframework.org/

1 page summary of the approach: https://drive.google.com/file/d/167TbSLTTD19kG9vM2q-EzzAASjLmCHXu/view

Interesting youtube channel from the lead of the community: https://www.youtube.com/@DavidShapiroAutomator/videos

Here's some handy links on AI security:

• OWASP Top 10 for Large Language Model Applications
• OWASP LLM Top 10 (PDF)
• Cloudflare: Firewall for AI

Very nice Drupal AI security module was proposed: ✨ [META] Create an AI Security module for custom moderation calls Active
A practical work within Drupal in this field.

@solideogloria thanks for the links.