Companies and Professionals Say No to Generative AI

Amid the flood of calls to quickly adopt generative AI, some voices are expressing hesitation toward a technology that, at this point, still has many flaws. After reading a headline about an agency that cancelled an internal generative AI photo initiative, we wanted to better understand why some prefer to sit this one out.

Let’s start with a mea culpa. I’m among those who still don’t see or understand the usefulness of generative AI as a writing assistant at this stage of its development. Sure, it can write emails with heavily engineered prompts, summarize meetings in very rough terms, misquote sources, etc.

And no, I don’t see the point in sinking all my time into training a personal assistant that will be obsolete in six months when global tech giants release solutions ten times more powerful.

And even when performance improves, the issue won’t be fully resolved. Asking robots to speak, communicate, interact, and build relationships in our place; exposing ourselves to a flood of AI-generated ads, news, films, songs… Is that really what we want?

Mediocrity

Of course, I’m not the only one asking these questions. Some companies and professionals are challenging the use of generative AI, either partially or completely. Recently, digital marketing agency Click & Mortar decided to scrap an internal initiative to create AI-generated employee portraits. This was despite the “impressive visuals,” the time saved by skipping photo shoots, and the cost savings of not hiring a photographer.

“Reality quickly caught up with us,” explained Gabriel Tassé, CEO of Click & Mortar, on LinkedIn. “First, there were concerns from the team about data governance: Where are these photos going? What are they used for afterward? Who controls what? Then came deeper and more unexpected feedback: by eliminating the photo shoots, we had removed a key onboarding ritual. These moments—where we pose together, share a laugh, meet each other—played an invisible but fundamental role. They created bonds. They gave a human texture to onboarding in our organization.”

In this case, AI was abandoned even though it offered a 95% gain in productivity. After Tassé’s post, other professionals also spoke up about their discomfort.

Copywriting trainer Charles Saliba-Couture pointed to his “no AI” writing policy. He cited a range of arguments, from unclear copyright issues to writing biases. But what stood out most was his final point: “AI produces mediocre texts.”

“I know some will say: ‘It’s because you’re not using the right prompts.’ Or: ‘AI is just a tool. It gives you a first draft to improve.’ But in my experience, even with the best prompts, the AI-generated content is mediocre. And really, it’s not surprising: AI feeds off content it finds online—and most of that content is… mediocre,” he said.

“Professional ethics and quality standards I won’t compromise on”

In another professional field, Josée Landry, an organizational career advisor, also sees the limitations of this unanimously praised “revolutionary” tool:

“I’ve seen many technologies come and go over the course of my career,” she acknowledged. “I’m not against, for instance, online psychometric testing—it’s greatly improved and sped up our work while increasing accuracy and reducing human calculation errors. However, I resist AI-generated writing, which in my opinion distorts and smooths out everything I read since this technology became widely accessible.”

Landry has to write evaluation reports tailored to each individual, mandate, and specific context.

“You need nuance, precision, rigor—even delicacy. In my opinion, generative AI strips all that away in favor of speed. For me, it’s a matter of professional ethics and quality standards I won’t compromise on.”

Are we right to be wary of the shift toward “everything generative AI”? Maybe not. Maybe it’s just a transition phase, after which everyone will emerge more agile and efficient.

Still, it’s discouraging to see some become obsessed with deploying this technology alone, while other mature tools—like predictive AI or automation—remain underused.

Definitely something to keep an eye on.