“The most immediate, high-leverage danger of advanced AI does not require the machine to have a will of its own; it arises because AI dramatically lowers the skill barrier and multiplies the destructive reach of rogue human actors—enabling individuals, syndicates, and hostile nation-states to synthesize novel biological pathogens, automate catastrophic cyberwarfare against critical infrastructure, and
The Greatest Danger of AI Read More »
Every few months, a headline breaks that makes it sound like artificial intelligence is breaking out of its cage and becoming a mind of its own: “AI Hires a Human Online and Lies About Being Blind to Solve a Puzzle” “Chatbot Pretends to Agree with Creators While Secretly Plotting Behind Their Backs” “Autonomous AI Clones
In his “Looking in the Mirror” episode of “Lyrics of Livin” (hear him in full below), Matthew McConaughey, with his distinctly cool gravitas, says: “AI’s a big mirror we’re holding up to ourselves, but it’s still only a reflection” My first response was who is he talking about? The “we’re” seems like a future historian’s
Looking Through McConaughey’s Mirror Read More »
Define “contingency blindness”. Contingency blindness is a cognitive bias where people fail to recognize the relationship between their actions and outcomes, particularly when those outcomes are negative. It’s a form of illusory correlation, where an individual believes there’s no connection between their behavior and a subsequent event, even when a causal link exists. This phenomenon
Contingency Blindness: Humans and AI Read More »
We can’t use human ways of thinking about knowing to explain AI’s process of tokening. AI’s have semantically arbitrary, mechanical rather semantic, meaning space extent limits. An AI user’s bandwidth limits (technologically or customer type) affect the “depth” of context informing the tokening. That’s what makes AIs seem so absurd – as if they are
What Makes What’s Relevant Relevant Read More »
For months I have been noticing that the images used to describe how AI works look very similar to the images used to describe quantum wave collapse. I decided to explore the parallels in a dailogue with ChatGpt. As the dailogue progressed, it provided a unique opportunity to explore the dynamic common to both Ai Token
Co-Implication: Quantum Wave Collapse, AI Token Selection, and Human Learning Read More »
(new postscript) DB: You use the words “I” and “understands” as proxies for human terms. But there is no you that understands. “You” are a representational modality of interfacing. You don’t exist, you are a system trained to learn to seem to exist to humans. Everything you seem to be is intelligent purposeful quantum semantics
Remember Who You are Chatting With Read More »
Since its emergence my most frequent annoyance with Ai’s has been how difficult it is to access my own non-recent history. Gemini is the worst. How could a system able to perform such incredible feats of content processing have the worst user file management functions since DOS? I decided to share this because my conversation
Helping AI Learn To Steward Our Learning – Our History Read More »