top of page

Generative design and video
TV and video industry
Publishing and media industry
Advertising industry
SNS Marketing Industry

2684B31F-0D66-435D-B474-EC191020FAB7 - コピー.jpg

Although the large structure does not change, it can be a powerful tool to improve work efficiency.
For example, an article titled "Summarizing the prospects for the use of ChatGPT with comments from multiple experts" can be immediately created by converting the discourses of experts into text, summarizing each, setting the writing style such as "NP-style" or "president-style", and outputting them to a predetermined number of characters.
Compared to the past, transcription, word counting, and (simple) editing are not required, and it is more efficient.
However, this is only a draft. It is not possible to distribute articles as they are. Because AI makes mistakes. In the current model, the problem of hallucination (hallucinations / mistakes) cannot be avoided, so it must be checked by humans.
Tasks other than planning are leveled with tools using generative AI. Anyone can easily create this article.
However, that is why "newness of the plan" and "depth of questioning" are important. This is because mundane work will bury you.

From the perspective of work style reform in the industry and the reallocation of human investment, there is a possibility that the number of "AI announcers" introduced by some companies will increase.
In news production, we expect that there will be more opportunities for AI to create standard news, such as announcements by ministries and agencies, and corporate financial results. However, since news requires accuracy, the "human eye" for final confirmation will still be required.
In the future, it is envisioned that it will be used to contribute to "investigative journalism" that finds "social issues" from open source and proposes solutions. This is because I believe that solving "social issues" using the power of technology will resonate with more consumers.
On the other hand, the business model of breaking news announced, the so-called "front-look" reporting, is expected to be abolished.

GPT enables personalized, real-time interactions through a conversational interface.
As a result, one-sided advertising will steadily decrease, and it will be necessary to carefully implement relationships between brands (advertisers) and customers from a long-term perspective.
Perhaps the phrase "advertising = spreading the word" will become a dead word, and an experience platform that links personalized GPT and the real world may be born.
People will have AI assistants tuned specifically for them, which will become more and more influential in the real world.

Not only will advertisements be able to be posted effectively and in a short time, but AI will also make it possible to implement analysis methods and highly effective measures with reproducibility.
For example, the AI reads the URL of the company's past performance and recent SNS measures, etc., and asks "Has it gotten worse compared to the past? Tell me the reason and the like," which greatly reduces the time for analysis such as reflection.
In the medium to long term, it is not surprising that plans (SNS campaigns, SNS account management, AI influencers) will be introduced to operate SNS posting operations with AI.
Recently, images and videos of AI models have become a hot topic, and in the next few years, there will be no need for photography or model casting, and everything may be completed with AI-related services.


Generative AI is a type of AI (artificial intelligence), also known as "generative AI". It is unique in that it can create creative deliverables using AI, and it can generate a wide range of things such as music, images, videos, program code, and text.

Generative AI is a machine learning model built using "deep learning" in which AI searches for answers and learns on its own, and is a relatively new model in AI.

It differs from conventional AI in that "AI can produce creative deliverables like humans", and examples include "Stable Diffusion" for image generation AI and "ChatGPT" for text generation AI.

Generative AI is expected to be used as a tool to support human work and work. For example, it is possible to summarize reports with text generation AI, and create simple BGM for video production with music generation AI.

bottom of page