Watch out with ChatGpt

After reading this blogpost, ChatGPT will no longer be what you thought it was #factcheck #fakenews #alert

If you follow me, you know that I am currently investing a lot of time in De Digital Comfort Factory, a new company that is bringing the Digital Comfort Index to the market.

This weekend, I challenged ChatGPT to tell me something about digital comfort and ways to measure employee well-being. As requested, the platform gave me a dozen methods that exist for this purpose:

  • De Employee assistance Program Assessment
  • Workplace Health Assessment
  • De Werkbaarheidsmonitor (SERV)
  • De Vragenlijst Beleving en Beoordeling van de Arbeid
  • De KU Leuven StressVragenLijst (KULSVL)
  • De Bevlogenheidsmonitor
  • Job Demands-Resources Scale (JD-R Scale) (die wij als basis gebruikten in onze research)
  • Utrechtse Burnout Schaal (UBOS)
  • Maslach Burnout Inventory (MBI)
  • Health and Safety Executive Stress Indicator Tool (HSE Tool)

The Story

It’s not surprising that ChatGPT doesn’t mention “The Digital Comfort Index” spontaneously at the moment. This tool is from after 2021, and the chatbot isn’t connected to the internet. Curious as I am, I ask more specifically if the bot had heard of the index. Here’s the answer:

Yes, I have heard of the “digital comfort index”. This is a measuring instrument that is used to measure the digital comfort level of employees. The digital comfort level refers to the degree to which employees feel comfortable using digital technologies in the workplace. The measuring instrument was developed to help organizations gain insight into the digital comfort level of their employees and to develop targeted interventions to improve this comfort level. (…)

Interesting. I then ask who developed the index…

The “digital comfort index” was developed by Acerta, a Belgian HR services company. The company developed the Digital Comfort Index as part of their service to help organizations improve the digital skills of their employees and (…)”

I google ACERTA and Digital Comfort Index and find no reference to it. I point out a possible mistake to ChatGPT, who apologizes:

My apologies if my previous answer was unclear. The concept of digital comfort is widely known and used in various forms and names in various research and applications, including in the context of work and organizations..

The term “digital comfort index” is specifically used by the company Attentia, which is active in Belgium and specializes in HR services and prevention advice. In collaboration with the research center imec (Institute for Microelectronics and Nanotechnology) and KU Leuven, they developed the Digital Comfort Index (…)”

Triggered by the correction, I search the World Wide Web for more details. It’s another dead end. I point out the situation to ChatGPT. Once again, I receive apologies 🙂

Apologies for the confusion, it appears that I gave incorrect information in my earlier responses. Upon further investigation, it appears that the Digital Comfort Index was not developed by Attentia, but by another organization called Attentix.

I guess you know what happened next …?

The story ends with yet another series of sincere apologies. I finally decided to share correct information with ChatGPT who gave me a warm “thank you” in return.

A few days later, there is still no mention of DCI in ChatGPT’s responses. The only thing I read now is that ChatGPT has no knowledge of a “digital comfort index, no knowledge at all (?)

What I learned

Most of us are familiar with online translation tools, right? They are extremely useful and can also provide hilariously incorrect translations. It helps if you know the target language and can verify everything, rather than just copy and paste before going live with a translation.

I have the same feeling about ChatGPT. This language model can provide a lot of information, but it is important to be cautious and not blindly trust everything it says. In this case, I challenged ChatGPT on a specific topic and discovered that the information provided was not entirely accurate.

Or maybe Google had it wrong? If you know, please share your thoughts and feelings in a comment.

So, my advice would be to always be on guard with ChatGPT and any other sources of information. Double-check everything and be critical of what you read. And don’t be afraid to ask questions and dig deeper when something seems off.

In conclusion, ChatGPT can be a great resource, but it is important to approach it with a healthy dose of skepticism and critical thinking.

Leave a Reply