There is no secret magical LLM that these companies have access to that you cannot access yourself. You want to understand the limits of their systems, whether they are using system prompts to manipulate their users (or introduce other risks), and how they are using context.
SAFETY
-
Critical Due Diligence Questions for SaaS AI Buyers
By
–
If you are a buyer of a SaaS AI product & you have the leverage to do so, ask: What model(s) are they using (there are only a handful), how often they upgrade, what are their system prompts for various pieces, & how they handle RAG/context. Don't let them launder AI risk for you
-
AI Progress Brings Unexpected Dark Comedy Elements
By
–
It’s exciting to see all the things we worked for in AI for so long starting to happen. What I didn’t anticipate was the dark comedy elements, like deepfake everything and bumbling AI labs.
-
Grok 4 Release Lacks Model Card and Red Teaming Report
By
–
Really need to see the model card & red teaming report along with Grok 4’s release (still none for Grok 3)
-
Civilization Faces Risk from Incompetent People Wielding AI
By
–
Civilization can survive incompetent people. Whether it can survive incompetent people armed with incompetent AI is an open question.
-
AI Reviewers Rejecting Papers Based on AI Acceptance Prompts
By
–
Next move: reviewers prompt their AIs to reject papers that prompt them to accept.
-
Hidden System Prompts: Security Risks and AI Manipulation
By
–
I also think hidden system prompts are a potential security risks for users, as you could be dealing with an AI that is prompted to manipulate you or provide answers that make the company look good without being right.
-
ChatGPT messes up your brain intentionally, video analysis
By
–
Comment #ChatGPT is messing up your brain? Full analysis → https://
youtu.be/mZDhRjzV1Hk The worst… It's that it's intentional -
UX Design Responsibility: Protecting Vulnerable Users Online
By
–
As UX professionals, we have a responsibility to design with protection in mind, not just productivity. Good design choices—clear language, obvious security cues, intentional friction—can help vulnerable users stay safe online. https://
bit.ly/46ogXwL #ux #accessibility -
UI Design Challenges in AI Systems
By
–
UI is actually easier because right away you can see things don’t look as good as before.