The realm of creativity knows no bounds, particularly in the presence of Artificial Intelligence (AI). The recently viral trend regarding ChatGPT perfectly exemplifies this phenomenon. Internet users instruct the AI chat bot to generate a specific image and subsequently request further iterations through successive commands. This trend underscores the vast potential and limitless possibilities inherent in the integration of AI into creative processes.
The “make it more” trend is making folks laugh as they see other users prompting the AI to generate images and then repeatedly ask it to make certain aspects of the image more extreme or exaggerated. This trend has gained popularity on social media, with users sharing humorous and often surreal examples of the results.
Make it More Trend Goes Viral on Social Media
The community is having a lot of fun with this particular AI feature as seen from the following posts. Several users have been sharing images of the results from their peculiar prompts, while others are responding with their own. Here’s one prompt asking ChatGPT to generate image of a food item and then asking it again to make it more “spicy”. The result is of course exorbitant with heavy special effects. It surely looks like an alien food now.
There are plenty of other examples under this thread for having a good laugh. It is a unique, fun, and creative way to explore the capabilities of AI to generate surprisingly realistic and detailed images. Users love to see how far they can push the boundaries of AI-generated content. For example, who could have thought something as simple as making a coffee would be so multifaceted. The following image is a result of repeatedly asking ChatGPT to generate an image showing a person making a coffee. The AI delivers a bunch of pictures looking like they are taken in the future.
How is ChatGPT Able to do That?
The ability of ChatGPT to generate images that respond to these increasingly extreme requests stems from its underlying technology, which is based on a large language model (LLM) called GPT-3. LLMs are trained on massive amounts of text data, allowing them to learn the patterns and relationships between words and phrases. This enables them to generate text that is both grammatically correct and semantically meaningful.
The LLM analyzes the prompt and the context of the previous prompts to understand what the user is asking for. It then draws upon its knowledge of the world and its ability to generate text to produce an image that meets the user’s request. As the user continues to ask for more extreme modifications, the LLM adjusts its understanding of the task and generates correspondingly more extreme images.
This is the reason how ChatGPT responds with its best representation of the given prompt. Here’s another example where the AI chat bot generates an image of Europe’s most famous landmarks. Things get interesting when the user asks to make it even more “European”. Base on its learning of the world, the AI starts incorporating mythical elements and Greek lore into the image to try and provide what the user requested.
An Exciting Test for the AI-Image Generation
The “make it more” trend highlights the creativity and flexibility of LLMs like ChatGPT. It demonstrates their ability to not only generate text but also to adapt to changing instructions and produce unexpected and often funny results. This ability to engage in interactive and open-ended communication is one of the key advancements in AI research and opens up new possibilities for human-computer interaction.
The trend is also a way of poking fun at the often-nonsensical nature of AI-generated art. By asking ChatGPT to make images more and more extreme, users are highlighting the fact that AI is not always able to produce coherent or meaningful results.
On the other hand, this can also be a challenging and thought-provoking experience. This can help them to better understand the capabilities and limitations of AI. In addition to being fun and entertaining, the “make it more” trend also help users to develop their creativity. By seeing how ChatGPT can generate different variations of an image, users can learn new ways of thinking about and approaching creative tasks. By analyzing the results of the prompts, users can learn to identify patterns and trends in AI-generated art.
It is also important to note that image-generation is only available in the paid version of ChatGPT. OpenAI introduced DALL-E 3, its latest image generation model, into both ChatGPT Plus and ChatGPT Enterprise in October 2023. They are subscription services priced at 20 USD a month.