This looks like a way more sensible approach than the "everyone must stop building AI or we'll blow them up" thing.
ETHICS
-
Frontier Safety Framework Updates for Advanced AI Systems
By
–
Important updates to our Frontier Safety Framework, with expanded risk domains for advanced AI and refined assessment protocols. https://t.co/Y1VlTjkgtn
— Demis Hassabis (@demishassabis) 23 septembre 2025Important updates to our Frontier Safety Framework, with expanded risk domains for advanced AI and refined assessment protocols.
-
Major Labs Avoid Fine-tuning Competitors’ Models Due to Company Policy
By
–
Fair. I think a lot of that is because the major labs are commercial labs, and they can’t (due to company policy)/don’t like to finetune competitors’ models
-
Anthropic’s Closed Model Strategy: Profit Over Open Source
By
–
Dario Amodei contorts himself into a pretzel to justify why Anthropic’s models aren’t open source. (The real reason is that he needs to make money, but that doesn’t play well with the “save the world” pitch.) https://t.co/hbc4cMvGFL
— Pedro Domingos (@pmddomingos) 23 septembre 2025Dario Amodei contorts himself into a pretzel to justify why Anthropic’s models aren’t open source. (The real reason is that he needs to make money, but that doesn’t play well with the “save the world” pitch.)
-
Why Less Rigor in Spiritual Beliefs Than Science?
By
–
Many people feel that it's ok to be less rigorous about their religious/spiritual/ethical beliefs than about other matters (such as science, finance of history). Why would you give less importance and accuracy to the core of your agency than to peripheral details?
-
Teachers prioritize accuracy inclusivity utility in AI evaluation
By
–
1. Teachers valued many different criteria but placed highest importance on accuracy, inclusiveness, and utility.
-
AI EdTech Tools: Balancing Context and Student Privacy
By
–
2. Teachers’ evaluation criteria revealed important tensions in AI edtech tool design. For example, providing adequate context requires significant time investment and sharing sensitive student data that could compromise privacy.
-
Teachers develop nuanced AI evaluation through collaborative peer exploration
By
–
3. A collaborative approach helped teachers quickly arrive at nuanced criteria. When provided time and structure to explore, critique, and design with AI tools in community with peers, teachers develop nuanced ways of evaluating AI – even without having received training in AI.
-
Moral Invulnerability in AI: First Takers Most Immoral
By
–
If cloak of moral invulnerability is offered, the first to take it will be the most immoral
-
AI Expert Admits Willingness to Compromise Ethics for Camera Gear
By
–
Not many know that AI is just a sideline & my true passion is blowing all my money on overpriced camera gear If Hasselblad are listening and they want to gift me their new X2D II 100C camera and a couple of lenses, I will drop my ethics and start shilling for them