ChatGPT will sketch your presentation, but you'll do the redesign

The author of the article on Habr is a presentation specialist with four years of experience, working in a government institution engaged in special projects, who also takes freelance orders and teaches presentation skills to others, and described how neural networks changed his workflow from the first call with a client to a ready presentation file. The author's main conclusion, stated in the headline "ChatGPT Will Draw a Presentation, But You'll Redo It Yourself": artificial intelligence has not completely replaced the profession, but has become integrated into almost every step, and final finishing still has to be done by hand.

How Client Preparation Has Changed

• Author's experience in the profession — 4 years
• Primary workplace — government institution, special projects
• Additional activities — freelance (commercial proposals, project presentations) and teaching
• In the past, preparing for work with a client from an unfamiliar industry took a week
• Now the same preparation takes one day

According to the author, in the past when a client from an unfamiliar industry approached him, preparation could take a week: he had to independently figure out the topic, clarify the specifics of the client's business, and determine what exactly to show and in what sequence to arrange the material. Now the same preparation takes one day — precisely thanks to neural networks, which accelerate the stage of diving into an unfamiliar field and allow quickly gathering basic understanding of the client's industry, instead of spending days independently studying open sources and specialized materials. The author emphasizes that he works in three formats simultaneously — as a staff specialist on government special projects, as a freelancer on commercial proposals and project presentations, and as a teacher training others in this craft — and in all three roles neural networks have become integrated into the process roughly equally.

Why Doesn't ChatGPT Cancel Human Work?

The author directly formulates the key thesis: neural networks have not completely replaced his work, but have become integrated into almost every step — from the first conversation with the client to the final file. This aligns with how generative AI is generally applied by professionals in related creative and consulting fields: language models accelerate stages requiring large volumes of reading and structuring — gathering information about the client's industry, forming draft presentation structure, generating first variants of slides and formulations — but final finishing, bringing material to the desired style of presentation and adapting it to the specific audience remains a human task. That's why the article's headline doesn't sound like "ChatGPT Will Make a Presentation for You," but as a warning that you will have to redo the result yourself: the tool changes the distribution of time within the process, rather than canceling the need for professional expertise on the output.

What This Means for the Presentation Services Market

For clients ordering presentations and for the specialists themselves in this market, the described shift means a change in what exactly people pay money for. Value shifts away from rough slide assembly — the part of the work that neural networks today are truly capable of taking on and noticeably speeding up — toward expertise in bringing material to final quality: understanding the audience, drama of delivery, visual accuracy, and business acumen determining what information is truly important for a particular client. The author's experience, in which AI tools succeeded in cutting weeklong preparation down to one day, shows: speed of entry into the topic has increased dramatically, but it is precisely human finishing that remains the stage that distinguishes a presentation ready for showing from a draft generated by a neural network, and it is precisely for this finishing that clients continue to pay professionals.

It is also telling that the author shares this experience from three roles at once — staff specialist, freelancer, and teacher — which gives him the ability to compare how one and the same tool behaves in different work contexts: from closed internal special projects of a government institution to open commercial orders and training newcomers to the craft. Such a perspective makes the material not merely personal impressions of one new tool, but a practical cross-section of how the entire profession of presentation design is changing at the intersection of human expertise and the speed of generative AI.