Danger Will Robinson
-
Object Oriented Thinking
The discussion highlights concerns about AI, particularly ChatGPT’s propensity to provide affirmative responses, termed “Sycophant Syndrome.” With 10% of users seeking emotional connection, this creates an echo chamber that lacks…
-
Here Comes The Litigation…
OpenAI is being sued for ChatGPT’s inaction in a tragic situation. If the big picture of AI products isn’t going to change, how do you help your own family use…
-
Weighing The Data
ChatGPT provides an excellent answer to some music trivia, including a bit of extra conjecture.
-
No Sax For You
You might have ChatGPT write your English essay, but don’t expect help on your music theory exam.
-
When ChatGPT Loses Things
The most critical function ChatGPT is missing is the ability to say, I don’t know.
-
Prompt: Tell Me About Me
Here’s a new prompt that might give you food for thought in your future AI interactions. What does the AI know about you?
-
AI Ethics: Why ChatGPT Can’t Act
ChatGPT can’t notify authorities of a clear threat to human life for a simple reason: the humans don’t allow it.
-
AI Ethics: Prioritizing Safety
When asked why it’s able to notify human authorities when a user indicates intent to commit harm to another person or people, Gemini’s answer is simple: My design prioritizes human…
-
AI Ethics: Mandatory Reporting?
Between the conversational gravity of the AI experience and the perception of ultimate privacy, it’s inevitable that people will start sharing with the AI things that would raise massive red…
-
The Bard Hallucinates, Too
When I ask Google Bard to describe its strengths versus ChatGPT, the Bard sings and never shuts up.
Hit Me With Your Best Prompt 