CNET Test Shows AI Tools Misinterpret Happiness As Productivity In Daily Schedules

A recent experiment published by CNET demonstrates that leading artificial intelligence tools fundamentally misunderstand human happiness. When asked to plan two days optimized for joy, ChatGPT, Google Gemini and Claude AI produced rigid schedules focused heavily on productivity and physical exercise.

The test involved a single prompt asking each tool to detail every hour from waking to sleeping. All three AI systems scheduled the user to wake up around 7 a.m. Claude AI specified a 15-minute earlier start for stretching, yoga and gratitude journaling. By 9 a.m., the plans required preparing substantial breakfasts and immediately beginning work on creative projects or writing tasks.

The Productivity Bias in AI Planning

The afternoon segments of all three schedules featured a consistent pattern of light, healthy lunches followed by mandatory walking sessions. These walks lasted between 45 minutes and an hour. After exercising for the second time in five hours, the AI tools directed the user back to work mode. The plans included only brief 10- to 15-minute breaks between work sessions.

This emphasis on labor contradicted the specific instruction to optimize for happiness. The CNET author noted that the AI seemed to equate constant productivity with well-being. The assumption appears to be that a human who is not productive for even a moment is deeply unhappy. This logic ignores leisure activities that many people find genuinely joyful, such as watching internet videos or eating ice cream.

Social Isolation in Nighttime Schedules

The evening plans further highlighted the AI’s narrow definition of a good day. Right before dinner, the tools instructed another 45-minute to hour-long walk. For the final meal, users were told to cook, order in or eat leftovers. The schedules excluded social dining options like romantic restaurant visits or meeting friends for drinks at bars.

By 9 p.m., the winding-down period consisted of solitary hobbies like doing puzzles, gardening or reading a book. The complete absence of social interaction in the evening hours left the author feeling lonely and resentful. The experience revealed that these AI tools lack the nuance to understand that human connection is a primary driver of happiness for many individuals.

What should you do?

Readers using AI for life planning should recognize these tools as productivity engines rather than emotional advisors. When generating schedules, explicitly define what happiness means to you in the prompt. Include specific requests for social interaction, unstructured leisure time and rest. Review AI outputs critically to ensure they align with your personal values rather than generic productivity metrics.

What to do and how to stay safe: ChatGPT

• Customize AI prompts to explicitly define happiness as including social interaction and leisure, not just productivity.
• Review generated schedules for excessive work or exercise blocks that may lead to burnout or isolation.
• Manually insert social activities like dinners with friends or family into AI-generated daily plans.
• Use AI tools for logistical planning only, avoiding them for decisions about emotional well-being or life satisfaction.

Step-by-step guide: Prompt Injection Warning Signs: How to Spot AI Manipulation

General security guidance from the Threat Times newsroom. It is not confirmed advice from the organisations named in this story.

Frequently asked questions

Which AI tools were tested in this CNET experiment?

The test included ChatGPT, Google Gemini and Claude AI, which were all asked to plan two days optimized for happiness.

Did the AI tools include social activities in their plans?

No, the schedules excluded social dining and friend interactions, focusing instead on solitary hobbies and work tasks.

How did the author feel after following the AI-generated schedules?

The author reported feeling lonely and resentful, noting that the plans felt isolating rather than happy.