LECTORAAT CHANGE MANAGEMENT – De Haagse Hogeschool

I must admit: I’ve been suffering from a major existential crisis, ever since AI came into our lives. After having spent two months, last semester grading vague and vacuous student assignments visibly written by ChatGPT, I can’t help wonder: what’s the point of it all? Also, reading a Scientific American article claiming that more and more researchers are “misusing ChatGPT and other artificial intelligence chatbots to produce scientific literature,” really didn’t help.

I feel a sense of doom and gloom. Where are we heading to as a society? As depicted in the dystopian film ‘Afraid,’ are we setting ourselves up to becoming slaves to robots? Will our brains become irremediably atrophied? Will our students lose the ability to think?

And presently, do I really want to have ChatGPT as a co-author? Can I really call myself a writer or a researcher if I do? A teacher?

Until now, I’ve buried my head in the sand, carried on giving my lectures, doing my own research, happily ignoring AI. Some colleagues said I was being resistant to change, which is a pretty nasty thing to say to someone who is a member of the research group change management.

So, I decided to take the leap. I opened ChatGPT and asked: “What would be a good topic for a blog, that would be about gossip in organizations and another current issue?”

After a few seconds, an entire exposé flashed on my screen, with the title: “The Double-Edged Sword of Workplace Gossip: How AI Surveillance is Changing Office Culture.” The article explained that more and more organizations are implementing AI-driven surveillance systems “to monitor employee behavior, communications and productivity” and warned that this could affect office gossip in a very negative way. Not only is there a rise in tools that monitor employee productivity, but also in tools “that track employee communication, and how people interact.”

I can’t help but feel impressed, not only with the blog idea in general, but also with the chatbot’s ability to self-reflect. 

“You could explore how AI technologies affect informal communication like gossip,” it suggests.  

-Which tools, specifically, are we talking about? Do you have any concrete examples? I prompt.

In a few seconds, it lists a few examples: Microsoft 365 Productivity Score, it writes, “tracks how employees are using tools like Outlook, Teams, Word and Excel, and also monitors how often employees send emails or attend meetings.”

Organizations like Amazon make use of a tool that measures “time off task” (TOT), and monitors how long workers are away from their workstations and flags those with reduced productivity” leading to stress and injury.

Similarly, certain companies reportedly used AI “to analyze Slack messages to detect potential negative sentiment. The AI scans communications for keywords and patterns that could indicate dissatisfaction or toxicity in the workplace.”

 ChatGPT warns that due to these technologies, employees will feel watched and won’t feel safe at work. These could potentially discourage open communication, including casual or informal conversations, such as gossip. Not only will employees feel pressured to avoid anything that might be perceived as “wasting time,” but they will also be “less inclined to share their honest thoughts about workplace dynamics, leadership, or even engage in friendly gossip, fearing it could be misinterpreted or flagged by AI.”

As I discuss in my book, Have I Got Dirt for You, gossiping is part of human nature. Every time managers have tried to rid their departments of gossip, their attempts have backfired.

“AI surveillance can undermine trust between employees and management,” ChatGPT warns, “leading to more negative or critical gossip. Employees may view monitoring as an invasion of privacy, which could increase dissatisfaction and result in a rise of gossip about management or organizational policies.”

Employees may feel they have fewer channels to voice frustrations or concerns openly, leading to “increased back-channel communication and whisper networks, where people can express themselves anonymously.” Such channels are definitely breeding grounds for “toxic gossip” and disinformation. As they are underground, management cannot get wind of employee concerns and address these before they fester into more serious issues.

According to ChatGPT, some AI surveillance systems are now sophisticated enough to detect negative sentiment or specific keywords in communication channels. “For example, if employees are using internal tools like Slack or Teams to gossip about leadership or co-workers, AI could flag these conversations for further review. This creates the potential for gossip to directly result in disciplinary action, reinforcing the idea that no conversation is truly private in a monitored workplace.”

So, as I walk over to the coffee machine at work, I wonder: should I first look whether a colleague is holding onto the latest i-phone, or wearing a FitBit? Should I check the settings on the coffee machine?

Or is it me, who’s just hallucinating?

Disclaimer

While ChatGPT was extremely instrumental in writing this blog, I did need to summarize more than six pages of information, and fact check to make sure it was not “hallucinating.” While most of the information turned out to be accurate, ChatGPT accused a certain logistic company to use Slack to monitor their employees’ communication. Maybe the company uses the algorithms well to bury such controversies, as I found no article to back those claims. So, I took out the company’s name, reminding how important it is not to trust AI blindly.