AI Guides and Research
Practical AI guides, research explainers and product evidence from SentX SI, a Superintelligence research company registered in Dubai, UAE.
How to Compare Video Generators Before You Pay
How should I compare video generators before paying for a project?
How Machine Intelligence, Machine Learning, Product Memory, and Learning After Deployment Differ
How do machine intelligence, machine learning, product memory and learning after deployment differ?
Keeping Your Voice When You Write With a Model
How can I use a model for writing, creative prompts and haiku while keeping my own voice?
Using a Research Assistant for a Literature Review Without Losing Your Sources
How can I use a research assistant for a literature review without losing track of sources?
What the Event-Frame Asymmetric Stereo Paper Actually Establishes, and Where Its Evidence Stops
What does Bidirectional Cross-Modal Prompting for Event-Frame Asymmetric Stereo actually establish, and what are its limitations?
Turning a Design Brief Into a Generated Image: A Prompt Method and a Review Process You Can Reuse
How can I turn a design brief into a generated image with a clear prompt and a useful review process?
A Full-Paper Reading Test: Correct Tables, Two Interpretation Errors
What did a complete-paper reading test in SentX get right, and which claims still needed checking?
Turning a Visual Brief into a Still and a Short Video Concept in SentX
How can I turn a visual brief into an image and a short video concept in SentX?
Why an Assistant's Memory Can Reinforce a Mistake
Why can an assistant’s memory reinforce a mistaken belief, and what do the research benchmarks actually show?
Google Aligns Guidance on Model-Generated Content With Rater Guidelines; What Website Owners Should Check
What did Google change in its model-generated-content guidance on 1 October 2026, and what should website owners check?
How to Correct or Remove an Outdated Memory in ChatGPT and Claude
How do you correct or remove an outdated memory in ChatGPT and Claude, and check whether it still affects answers?
Moving Project Context Between Assistants Without Mistaking a Summary for a Memory Transfer
How can you move useful project context between assistants without mistaking a summary for a complete memory transfer?
What a pasted checkpoint actually preserved, and what still needed checking
What can a supplied project checkpoint preserve in a fresh assistant conversation, and what still needs checking?
Choosing a Memory-Enabled Assistant for a Long Project: A Test You Can Run Yourself
How should I choose a memory-enabled assistant for a long project?
Continuing a Research Project in a Later Conversation Without Losing Its Sources or Decisions
How can I continue a research project in a later assistant conversation without losing its sources and decisions?
How to Check a Model-Generated Research Summary Against the Original Paper
How do I check a model-generated research summary against the original paper?
What Survives When You Start a New Conversation: Context Windows, Saved State, and Product Memory
What can an assistant retain when I start a new conversation, and how is that different from its context window?