Hi Reader,

Here are your top AI stories for the week:

• ChatGPT can weaken your brain
• Claude shares nerve gas recipe
• Amsterdam ends AI experiment due to bias

Read more below! 👇

Sponsored by: Brain.fm​ Transform Your Focus With Brain.fm​ I know you're always on the hunt for tools that genuinely improve your life—which is why I'm excited to introduce you to Brain.fm's groundbreaking focus music. ​Brain.fm's patented audio technology was recently validated in a top neuroscience journal, showing how it actually changes your brainwaves to enhance concentration. They even offer a specialized ADHD mode, specifically engineered for brains that process attention differently. ​Brain.fm is a powerful companion for any serious writing or deep thinking work. For a limited time, my readers can get a full month of Brain.fm for free. ​Try Brain.fm Free for 30 Days →​

​

🧠 ChatGPT can weaken your brain

A new MIT study showed that using ChatGPT to write for you has a negative impact on your cognitive development. The researchers monitored the brain activity of 54 highly-educated adults while they wrote essays, and found that those who only used ChatGPT showed much weaker brain connectivity than those writing on their own or using Google.

However, brain connectivity increased when participants started by writing on their own and then used ChatGPT for editing. This highlights the importance of integrating AI tools into our educational system in a thoughtful and deliberate way that maximizes the benefits to young minds.

🧪 Claude shares nerve gas recipe

Since the introduction of modern AI chatbots, there has been a theoretical concern that they could be tricked into giving out truly dangerous information, but that concern is becoming more real every day.

In this case, a security researcher convinced the new Claude 4 Opus model to provide detailed instructions for creating sarin gas. This represents a greater risk than that of a search engine because the chatbot can provide “practical guidance on specific steps in response to follow-up questions, like how to disperse the nerve gas.”

With many AI companies sprinting to release new models ahead of their competitors and dominate the market, it’s hard to imagine that they are all taking these safety concerns seriously enough.

❌ Amsterdam ends AI experiment due to bias

Despite its best efforts to follow every principle of responsible AI, Amsterdam’s attempt to detect welfare fraud using AI ended in failure due to persistent algorithmic bias.

This was initially caused by bias in the training data, but even after the city corrected for that, new bias in the model emerged. This highlights the ongoing danger of using AI to make decisions in any context in which human rights are at stake.

Thanks for reading, and feel free to share it with a friend! 🤝

- Kevin

P.S. Looking for a beginner-friendly AI course? Check out Build an AI chatbot with Python ($9).