Logo
Subscribe
Log In

Evidence-Based Prompts

💻 Powerful prompts to turn AI into your personal PhD-Level Behavioural Designer—delivered weekly(ish).

Sign up for the newsletter and enjoy access to exclusive updates.


Dive into our featured stories ↓

🧠 Dangerous AI Biases

Communications

+1

🧠 Dangerous AI Biases

When tested against 180+ cognitive biases, AI revealed 9 human biases it mirrors the most — and they’re the riskiest AI biases you need to watch out for.

🧠 The Real Effect of Duolingo

Communications

+1

🧠 The Real Effect of Duolingo

Over 8 million learners have achieved a 365-day streak or more. Yet, its design choices spark debate about effectiveness versus engagement.

🧠 Unintended Consequences for Consultants Using AI

Communications

+1

🧠 Unintended Consequences for Consultants Using AI

AI can enhance the speed of your client recommendations, but it can also distort what you recommend. The IN CASE framework helps you anticipate what might go wrong when using AI — before it affects the people you serve.

Catch up on the newest stories from us…

Explore our top tags:

Goal-Setting

Communications

Artificial Intelligence

🧠+🤖 Will AI Ever Be Unbiased? (SCIENTIST REVEALS)

🧠+🤖 Will AI Ever Be Unbiased? (SCIENTIST REVEALS)

A behavioural scientist's review of 2025 research on AI biases and what actually works to mitigate them

🧠 Two Bias Fixes Tested on LLMs

🧠 Two Bias Fixes Tested on LLMs

Human crowdsourcing is prone to cognitive biases. Researchers have two proven strategies to fix this. But when these same strategies were tested on AI, one worked and the other completely failed.

🧠 5 Signs Your Prompts Suffer from Context Rot

🧠 5 Signs Your Prompts Suffer from Context Rot

Learn what Chroma Research discovered when testing 18 AI models—including the latest GPT, Claude, and Gemini versions—on how context length impacts performance. Each sign comes with evidence and a practical fix.

🧠 Prompt That Teaches AI When to Think Again

Artificial Intelligence

🧠 Prompt That Teaches AI When to Think Again

In ode to late Nobel laureate Daniel Kahneman, author of Thinking, Fast and Slow, researchers from IBM Research, University of Oxford, and Tulane University have built an AI system that doesn’t just think fast or slow—it knows when to rethink.

🧠 Meet Your Systems Thinking GPT Coach

Artificial Intelligence

🧠 Meet Your Systems Thinking GPT Coach

Your systems thinking coach offers you three frameworks to see: (1) Root causes, (2) Leverage points, and (3) Stakeholders to influence.

🧠 Dangerous AI Biases

Communications

+1

🧠 Dangerous AI Biases

When tested against 180+ cognitive biases, AI revealed 9 human biases it mirrors the most — and they’re the riskiest AI biases you need to watch out for.

Load more

Recommended reads for you

Vishal George
🧠 Unintended Consequences for Consultants Using AI

Communications

+1

🧠 Unintended Consequences for Consultants Using AI

Vishal George
🧠 The Subtle Art of Asking AI

🧠 The Subtle Art of Asking AI

Vishal George
🧠 Level Up Your ChatGPT Prompts with Design Thinking

Artificial Intelligence

🧠 Level Up Your ChatGPT Prompts with Design Thinking

Subscribe for new reads…


💻 Powerful prompts to turn AI into your personal PhD-Level Behavioural Designer—delivered weekly(ish).

Quick Links

Subscribe

Archive

Tags

Subscription

Signup

Login

Reset Password

Update Password

Search

© 2025 Behavioural by Design.

Privacy policy

Terms of use

Powered by beehiiv