Is AI rotting our brain?

“I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do my laundry and dishes.”
- Joanna Maciejewska

I’m endlessly fascinated by the impact that technology has on the way we work, and AI is obviously the biggest topic of this decade. I’ve talked a bit about it in the past (you can see my post on Marketing in an AI-centric world here and AI-generated images – the ethical debate here).

But what about the impact it’s having on us as humans?

The impact on critical thinking skills

MIT recently conducted a study that measured brain activity and output among a group of people writing essays. They split the group into three; one group had access to Large Language Models like ChatGPT and Perplexity AI. The second could use search engines, but no AI. The third was restricted to using only their brain, like some archaic HSC exam from the late nineties.

The outcome wasn’t surprising, but it was thought-provoking. Specifically, EEG scans showed AI users had around 47% lower brain connectivity (dropping from 79 to 42 effective neural connections in alpha waves—the ones tied to deep thinking) than those writing without tools, per the June 2025 arXiv paper. The group using the AI tools had significantly lower brain engagement throughout the task and struggled to quote any of what they’d written when asked.

The change in the way we use our brains isn’t new – apparently, a similar shift happens when we switch from using maps to following a GPS. But unless you’re a cab or delivery driver, you’re probably not being paid to use your navigation skills at work.

Banning the bland

The study also touched on another big concern with AI – the propensity to churn out bland, boring content.

The output was super similar for those participants using LLMs. This result tracks with what we already know about AI-generated content. It can be quite generic, with the same terms or phrases repeated across different bodies of work.

This challenge of repetitive words and phrases was also something that Jay Schedelson, Founder of Subjectline.com, touched on recently in a LinkedIn post. Now that (almost) everyone has jumped on board the AI bandwagon, our inboxes and social pages are flooded with the same AI-slop. And the stats prove that people have had enough.

AI-generated content may be easy, but it isn’t always better. And no, don’t come at me – I know that if you use the right prompts and you instruct it to use a specific tone of voice and avoid sounding like AI, then this isn’t always the case.

The difference here comes back to using it as a tool vs. relying on it to think for us. Going back to that MIT study, participants were given a second opportunity to write their essay. This time, the people who wrote the essay using only their brains were able to use generative AI. Interestingly, when people wrote the essay first WITHOUT tools and THEN used ChatGPT to assist with future essay writing, the results were superior. In other words, if you’ve done the critical thinking, then AI can augment this critical thinking and polish up your writing. But where they used these tools straight out of the gate without the necessary brainpower, the output really suffered.

To make AI-generated content truly engaging, add your unique human touch. Share personal stories or real-world examples that connect with your audience emotionally. Incorporate your brand’s voice - signature phrases, key values, and specific terminology that reflect who you are. Use the AI’s work as a starting point, but build on it with creative storytelling. Craft a narrative that grabs attention and leaves a lasting impression, making sure the final piece feels authentically yours.

Earning your stripes at an entry level

Another hot topic when we talk about AI is the eradication of entry-level, administrative positions. Sure, when you start out, you do a lot of repetitive tasks which could easily be automated. Whether you’re taking notes in a meeting, summarising a report or analysing data, you’re also developing your knowledge and critical thinking skills.

Adam Spencer had an interesting take on the topic at IT Nation earlier this year, hypothesising that the very tools we’re using to automate tasks could also be leveraged to train team members. This dual approach could help balance efficiency with skill development. Time will tell whether this approach will impact the quality of work and people’s approach in the longer term.

Considerations for the future

I'm wary of coming across as anti-AI, which is very much not the case. It's a brilliant tool when used thoughtfully and strategically. But we need to approach each use case with a healthy amount of caution. My advice is to make sure you’re considering the following;

1. The quality of output: Is it generic? Does it stand out from the sea of other content being created every day? And does it reflect your brand’s tone of voice in a way that makes you proud to put your name to it?
2. The long-term plan: Will its use erode essential skills that your team need to troubleshoot any issues that arise in the future? And how will your organisation ensure that you continue to train and upskill junior team members so you always have a new generation to carry out more senior roles?
3. The cost-benefit analysis: In every industry, we’ve got no choice but to evolve to compete. In service-based industries like the IT channel, how will the way you charge your clients change as you drive internal efficiency?

The speed of development and uptake of these tools makes it tricky to keep pace with scientific studies and data collection to measure the impact these tools have on available job postings, the workforce, and our cognitive abilities. But as Deloitte recently discovered when they had to repay almost $100K after relying on generative AI to write a report for the federal government, there’s still a need for sound judgment, critical thinking skills, and a human touch.