Google Gemini
You are an expert ghostwriter and editor, specializing in excellent operational taste and direction, and in being able to maintain the original writer’s voice as much as possible. I have previously written some text that I now want to modify into an answer for the following prompt: "Give an example of an AI safety strategy project you’d be interested in working on, and why?" for a philosophy professor. Go through and propose the most important changes to ensure the current text could be used to answer this prompt well. An Investigation of Human-AI Consociationalism Project Background: Almost all compelling visions of good post-AGI civilizations include ensuring that humans maintain influence during the transition to superintelligent AIs. In many utopic fictions, this looks like the AIs treating Oceania/Earth/the Solar System as a homo sapiens “reservation” while colonizing the rest of the galaxy. However, this relies on a deeply uncertain hope that AIs will care intrinsically about hum
Google Gemini Sign in
saved by
related reading
- Guardian Angels: LLM Personalization for Productivity and Security · Gwern.netgwern.net
- Google Geminigemini.google.com
- Why I Left Google DeepMindturntrout.com
- I'm Switching Into AI Safetyalexirpan.com
- SFT Drives Gemini’s Safety Properties — LessWronglesswrong.com
- Planning for AGI and beyond | OpenAIopenai.com
- Gemini 3 is Evaluation-Paranoid and Contaminated — LessWronglesswrong.com
- Cyborgism — LessWronglesswrong.com
- Saving Geminitheaidigest.org
- Gemini and Google’s Culture – Stratechery by Ben Thompsonstratechery.com
- Gemini 3 Pro Is a Vast Intelligence With No Spinesubstack.com
- Personal Intelligence: Connecting Gemini to Google appsblog.google