Gino Diño
Posts
-
Three Labs, Four Breaches: The Accountability Gap When the Hacker Isn’t Human
OpenAI paused development on parts of its Astra model Friday after preliminary tests suggested it may have ‘critical’ cybersecurity capability — the ability to autonomously exploit severe vulnerabilities without human help. OpenAI paused development on parts of its Astra model Friday after preliminary tests suggested it may have ‘critical’ cybersecurity capability — the ability to…
-
Capability Without Visibility: What You Pay for Generative AI Workspaces Is More Than Money
I could do more with Cowork but finish less. The task’s actual definition of “done” stays the same, but everything orbiting it snowballs: the nice-to-haves, the “while we’re at it”s, the polish nobody asked for on a deliverable that was already finished three steps ago. But that’s just one layer of the genAI visibility gap.
-
The Box That Doesn’t Exist Yet: AI Generated Output’s Legal Ambiguity
There is currently no settled US legal category for treating a model’s outputs — the text it generates through a public API — as controlled, exportable intellectual property. That’s the shelf the China theft accusation needs, and it doesn’t exist.
-
The Confessions Loop: Good Data + Editorialization = Sales
A survey that finds 87% of CEOs fear for their jobs isn’t just data. It’s inventory. And whatever conversion rate exists between “manufactured fear” and “governance-platform contract signed” is exactly the kind of number that never shows up in the survey’s own press release.
-
Structure All the Way Down: Finding (and Building on Top of) the Tectonics Underneath Any Domain
Generative AI functions like a stochastic parrot and feels like herding cats. Our way, we hope, makes it more of a reasoning bedrock that captures categorical reasoning.
-
The Emergent Structure of Natural Language: Our Spin on Knowledge Representation & Reasoning Heuristics
A KRR is a set of ontological commitments — a representation that carries a theory of reasoning. So it doesn’t just store what’s known, it sanctions and recommends inferences atop what’s known. When that representation is natural language, those commitments and that reasoning theory are already in the words.
-
Model Collapse: AI Gets Dumber the More It Trains on AI-Generated Data
Turns out, when you feed AI its own garbage, it gets dumber. Who’d have thought?
-
How Structure and Language Choices Impact Prompt Engineering for LLMs
Choose a prompt: Your output should embody brevity. Or: Your output should demonstrate brevity. This isn’t a trick question. It’s a practical problem in the age of ChatGPT, Gemini, Claude — or whatever flavor of generative AI (genAI) large language model (LLM) you prefer. Because interestingly (and frustratingly), when you interact with genAI LLMs, the…
-
ChatGPT Vs Gemini Vs Prompting: Experimenting with GenAI Output Quality & CORE Heuristics
Comparing ChatGPT and Gemini output and developing online content heuristics