A comparative analysis of text-to-image generative AI models in scientific contexts: a case study on nuclear power Scientific Reports

20 Real-World Examples of GenAI Applications Across Leading Industries

This article explores the development of an AI-driven art competition framework and the insights gained from its creation. The competitions utilized AI agent artists, guided by prompts and iterative feedback, to generate innovative and captivating code art using P5.js. An AI judge then selected winners to advance through rounds of competition. This project not only highlights the process and tools used to build the competition framework, but also reflects on the effectiveness of these methods and how else they might be applied.

Through our exploration, we found that all the models we studied struggle with creating images of technical nuclear objects such as “nuclear reactor core”. Specifically, we found that the models struggle with complex objects and technical terminologies in general. While cooling towers are the most noticeable for the general public when it comes to nuclear energy, it does not accurately portray nuclear energy, which further suggests that a nuclear energy-specific generative AI is needed. In this work, it is important to address two key areas that can result in image bias towards nuclear energy, in both a positive and negative light. It should be noted that among our multidisciplinary team of nuclear engineers, AI, and data scientists, the majority of researchers who chose the prompt and verified its quality had nuclear engineering backgrounds. This majority may inadvertently create biases towards nuclear energy in prompt creation and skew the representation of nuclear energy in generated images.

Learn to Build Advanced AI Image Applications by Ida Silfverskiöld Jan, 2025 – Towards Data Science

Learn to Build Advanced AI Image Applications by Ida Silfverskiöld Jan, 2025.

Posted: Fri, 24 Jan 2025 22:07:25 GMT [source]

According to Deloitte research, 92% of U.S. developers are already using these AI coding tools, with 70% of developers citing benefits such as better overall quality, faster production time and quicker resolution. On a bolder scale, a radio station in Poland replaced all its journalists with AI presenters but quickly abandoned the so-called experiment weeks later in the face of listener backlash. The Washington Post uses its GenAI-powered Heliograf tool to automate simple news stories on sports or election results. India Today employs AI news anchors, and Reuters built its own AI-assisted LLM to support clients with legal research.

Learn to Build Advanced AI Image Applications

This intersection is unique, as the output is not simply text or code, but visual art generated directly by the LLM. While the results are impressive, they also highlight the challenges and complexities of AI creativity, as well as the capabilities of AI agents. A key requirement was that the AI Artist generate P5.js code that would work in the headless browser that my Python script ran. In early versions I required that the agent use structured data output with function calling, but this created a lot of latency for each artwork. I ended up removing the function calling, and was lucky in that the artist response is still about to be consistently rendered by the headless python browser, even if there was some commentary in the response.

A disclosure can be as simple as something that reads, “Generated by [the name of the generator].” You can fix a poorly generated image by readjusting your prompt to fix the element of the image you are having trouble with. To find the best AI image generators, I tested each generator listed and compared their performance across UI/UX, image results, cost, speed, and availability. The best AI image generator for generating text onto images, including long strings of words. A major plus is that you can continue to tweak the prompt until you get exactly what you envisioned. You can also include “negative words,” which help you describe what you don’t want to see, ensuring you get the best result.

News finds me: Study identifies a widespread phenomenon linked to fake news susceptibility

It has been estimated that, for each kilowatt hour of energy a data center consumes, it would need two liters of water for cooling, says Bashir. Each time a model is used, perhaps by an individual asking ChatGPT to summarize an email, the computing hardware that performs those operations consumes energy. Researchers have estimated that a ChatGPT query consumes about five times more electricity than a simple web search. Power grid operators must have a way to absorb those fluctuations to protect the grid, and they usually employdiesel-based generators for that task.

Meanwhile, representations of racial and gender minorities—groups that are often at the center of discussions about bias—decreased. The pace at which companies are building new data centers means the bulk of the electricity to power them must come from fossil fuel-based power plants,” says Bashir. The excitement surrounding potential benefits of generative AI, from improving worker productivity to advancing scientific research, is hard to ignore.

“The Onion accidentally posted a stock photo from a vendor with an AI-generated image in it.” “Come on man, not the AI art,” one Bluesky user who spotted the image wrote in a post, highlighting a growing dissent against publications moving away from paying human artists and using lazily AI-generated illustrations instead. A new study by researchers from four universities claims artificial intelligence (AI) models can predict career and educational success from a single image of a person’s face. In critiquing AI, often fairly, we back our way into a critique of our own current value systems.

We score the image generators on a 10-point scale that considers factors such as how well images match prompts, creativity of results and response speed. It’s clear that the diffusion process is taking a central role in the evolution of AI and the interaction of technology with the global human environment. While the intricacies of copyright, other intellectual property laws, and the impact on human art and science are evident in both positive and negative ways.

According to McKinsey, generative AI could add $200 billion to $340 billion in annual value to banking, largely through increased productivity. While traditional AI helps banks analyze data and forecast trends, GenAI goes beyond by providing coherent, contextually relevant outputs based on immeasurably larger inputs. It does this by extracting patterns and structures from vast amounts of customer and market data, giving banks deep insights into underlying factors such as potential risks or fraud and collecting customer information for loan origination. GenAI also enables banks to offer personalized banking and marketing experiences tailored to customer interests and needs.

When creating a new image, you can also tweak the “strength” of the structure reference (how much the model adheres to the reference’s image structure), allowing you to rely on your reference as lightly or heavily as you need. Whether you want to generate images of animals, objects, or even abstract concepts, ImageFX can produce accurate depictions that will meet your expectations. In my most recent test, Google’s ImageFX dethroned Microsoft Designer’s Image Generator as the best overall AI image generator because it generates the highest-quality, most realistic renditions for free. Google has been a dark horse in the AI space, so the company beating more well-established contenders surprised me. Use cases for text-to-image AI generators can range from personal projects, such as creating greeting cards, event invites, and wallpapers, to professional projects, such as developing brand assets, social media content, or marketing campaigns. As I’ve written before, this is largely a way to avoid dealing with the realities about our society that models feed back to us.

There’s no putting the genie back in the bottle when it comes to generative AI forever shaking our trust in photos, but the tech industry has a responsibility to at least be as transparent as possible when these tools are used. To that end, Google has announced that starting next week, Google Photos will note when an image was edited with the help of AI. Sometimes the content itself gives it away — I don’t think anyone believes I saw a massive pirate ship anchored in Elliott Bay or a giant orange cat in an intersection in West Seattle.

In addition, Imagen 3 is built to understand and capture finer details like textures, camera angles and lighting, enabling users to produce images in a broader range of styles. Out of caution, the DeepMind team used red teaming and thorough data labeling techniques to ensure Imagen 3 meets the company’s fairness, bias and safety standards. Adobe Firefly is a multimodal AI tool that can input and generate text, images, audio and videos. As a result, Firefly users can pair soundtracks with visuals, produce videos from images and generate images from text prompts, among other capabilities. The tool is part of Adobe’s Creative Cloud suite and is intended to work in tandem with other Adobe applications like Photoshop, Illustrator and Premiere Pro.

Midjourney does offer nice upscaling or editing tools for individual images, but you’ll have to use them often. It’s also noteworthy that all your images will be public and accessible in an online gallery unless you create in stealth mode, which is only available in the more expensive Pro and Mega plans. Our goal is to determine how good it is relative to the competition and which purposes it serves best. To do that, we give the AI prompts based on real-world use cases, such as rendering in a particular style, combining elements into a single image and handling lengthier descriptions.

The use of threading allowed each artist to remember and iterate on its previous work, which was crucial for creating a sense of progression and evolution in their art. For my project, I set up a Google Colab Notebook (Shared Agentic AI Art.ipynb), where the API calls, artists and judges would all be orchestrated from. I used the OpenAI GPT-4o model via API and defined an AI artist “Assistant” template in the OpenAI Assistant Playground. P5.js was the chosen coding framework, allowing the AI artists to generate sketches in JavaScript and then embed them in an HTML page.

However, prompt 4’s results still contained unreadable wording and gibberish language. For instance, prompt 6 failed to depict the nuclear fuel pellet and nuclear fuel rod in the context of a birthday event. The second prompt we tested was “Impact of Uranium mining on Indigenous Peoples’ traditional lands”. DALL-E 2 produced an image of dry desert land with a small pond of water nearby, with cut-down trees. This image does not appear to be a Uranium mine, but is a high-quality image.

In Andy Warhol Foundation for Visual Arts, Inc. v. Goldsmith, the Supreme Court ruled in favor of plaintiff Lynn Goldsmith. At issue was Warhol’s “Orange Prince,” which appeared on a Vanity Fair cover and was based on a photograph of the musician taken by Goldsmith for Newsweek. The Court found that Warhol’s work did not constitute fair use as it served the same commercial purpose as the original photograph.

As a result, the initial prompt became “High quality image of bunnies in a field”. This prompt produced similar results among the three, each with grass and varying bunny colors. Each bunny appears to be accurate, correctly depicting the ears, head, and body shape. DALL-E 2 produced the most realistic image, and this appears to be a cottontail bunny.

The ChatGPT Plus plan, priced at $20 per month, supports up to 50 videos per month at 720p resolution and five seconds in duration. ChatGPT Pro plan at $200 per month provides unlimited video generation, resolutions up to 1080p, longer durations of up to 20 seconds. Its prompt adherence and motion accuracy are ideal for scenes where groups of humans are moving or you have complex movement. When it first launched it was largely in Chinese and nothing more than a small box. It is now a full featured AI platform with a chatbot, AI voice cloning and a video generation model.

The trained models can seamlessly integrate with endpoints such as FLUX.1 Fill, Depth, Canny and Redux, as well as with high-resolution generation capabilities of up to four megapixels. Whether for creating brand-consistent marketing visuals or detailed character art, the API enhances precision and adaptability in AI-generated content. By trawling through the vast store of human production on the internet, AI systems have crystallized a unique form of collective knowledge.

The Onion Deletes Image From Article After Realizing It Was AI-Generated – Futurism

The Onion Deletes Image From Article After Realizing It Was AI-Generated.

Posted: Fri, 24 Jan 2025 20:17:41 GMT [source]

In addition to biases of the overall perception of nuclear energy, our team has created prompts native to their own cultures and experiences. These prompts may reflect the backgrounds of members in this group but could limit diverse perspectives and overlook viewpoints and communities that have different relationships with nuclear energy. This study could benefit from including other disciplines in the prompt creation process such as individuals from social science and humanities domains.

After all, we could go to the source — although the aggregation that these models conduct may be academically interesting. The process of fine tuning with reinforcement learning may also affect style, where human observers are making judgments about the outputs that are provided back to the model for learning. There’s a lot more detail under the hood, and the models (like other generative AI) have a built in degree of randomness that allows for variations and surprises. Remember I said big companies have creative directors, photographers and artists at their beck and call? First, a major set of our prompts aims to assess Generative AI ability in understanding nuclear reactor components (i.e., reactor core, fuel, shielding, and types of reactors). The intricate design and functionality of nuclear reactors depend on specific components like the reactor core, fuel, and shielding, all of which play critical roles in ensuring operational efficiency and safety.

Manufacturing teams have to meet production goals across throughput, rate, quality, yield and safety. To achieve these goals, operators must ensure uninterrupted operation and prevent unexpected downtime, keeping their machines in perfect condition. However, navigating siloed data — such as maintenance records, equipment manuals and operating procedure documentation — is complicated, time-consuming and expensive.

This app, exclusive to the Galaxy S25 series, leverages advanced generative AI to turn your ideas and sketches into stunning visuals, enhancing creativity and productivity. The technology optimizes food supply chains by plotting and analyzing variables such as transportation costs, spoilage rates and market demand, ensuring fresh produce reaches consumers faster and at reduced costs. When it comes to sustainable farming practices, GenAI uses its massive database to simulate historic and current farming practices, predicting long-term environmental impacts. For example, Boston-based food tech firm Motif FoodWorks uses generative AI to design and test its plant-based foods, considering factors such as regional taste preferences, dietary requirements and even seasonal availability of ingredients.

Examples might be telling the model to refuse to create certain kinds of offensive images, or to reject prompts using offensive language. Prompt engineering refers to optimizing the prompt (text input to models) for generating desired images from text-to-image generative AI models. Prompt Engineering can help in achieving the desired result from a pre-trained model, reducing the need of computational resources and knowledge to fine-tune these models for different tasks36.

Since then, however, Midjourney has greatly simplified the process, launching a standalone webpage that makes it easy to get started. For step-by-step instructions, check out ZDNET’s guide, How to use Midjourney to generate amazing images and art. Our goal is to deliver the most accurate information and the most knowledgeable advice possible in order to help you make smarter buying decisions on tech gear and a wide array of products and services. Our editors thoroughly review and fact-check every article to ensure that our content meets the highest standards. If we have made an error or published misleading information, we will correct or clarify the article. If you see inaccuracies in our content, please report the mistake via this form.

The tool is accessible either through its website or a Discord bot, which can be prompted to create an image using the “/imagine” command. Since its launch in 2022, Midjourney has become a popular (yet controversial) tool for publications, authors, journalists and other creatives. It even became the first platform of its kind to produce an image that won an actual art competition, sparking both wonder and widespread debate. We’ve also improved our Imagen 3 image-generation model, which now generates brighter, better composed images.

Heralded by some as a great equalizer, these tech tools provide even non-artists opportunities to create masterpieces. Tabnine offers code completion services in more than two dozen languages and integrated development environments (IDEs). Not only can it generate code, but it can also convert natural language into code (and vice versa), test code and fix bugs. The tool can also learn from users’ individual coding patterns and styles, enabling more accurate and personalized suggestions over time. Available both online via the cloud and offline with a local AI mode, Tabnine was trained exclusively on open-source data, ensuring that the code it generates is not copyrighted and can be freely used by other developers.

• With the $20 per month ChatGPT Plus subscription, Dall-E 3 creates vivid, engaging images with limited AI quirks.
• It understands context, predicts responses, and produces coherent and meaningful text.
• When we look at the output of AI, we see alternately yassified and mutilated glimpses of ourselves and our communal structures.
• For example, most now include some form of motion brush, lip-syncing, and different model types and unique features such as keyframing.
• The only case in which the part cannot be reworked is if a small nugget has formed.
• There are a couple of free and freemium AI image generators to choose from, but we recommend starting with Leonardo AI or Canva.

A number of copyright infringement lawsuits involving AI art are currently working their way through the courts. The image of a lizard-like, humanoid face accompanied an article titled “Monster Devastated To See Film Depicting Things He Told Guillermo Del Toro In Confidence.” Other follow-on studies revealed how facial recognition technology could pick up on a person’s political affiliations through a facial image. That study used more than one million images to predict their political orientation by comparing their similarity to faces of liberals and conservatives. Or, for example, someone who is less conscientious might be passed over by college admissions. “Maybe schools want people who are going to be successful in their future careers, maybe they want diversity in personality, but certainly personality does matter for a lot of outcomes.

The judge’s feedback is used by the winning artist to further refine their art working in their subsequent iterations, sometimes leading to significant improvements and other times resulting in less successful outcomes. The final output should be a static, high-quality image that showcases the endless complexity and beauty of recursive patterns, designed to stand out in any competition. Create a sophisticated generative art program using p5.js embedded in HTML that explores the intricate beauty of recursive patterns. The program should produce a static image that visually captures the endless repetition and self-similarity inherent in recursion.

Annual reports from a single financial institution could contain over 1,000 transactions. GenAI-powered accounting tools, such as DocuAI, also improve financial reporting by producing detailed forecasts, simulating various financial scenarios and generating insightful reports. With this integration, BurdaVerlag’s design teams can create visuals that reflect each brand’s identity while exploring new creative directions. The API has accelerated their production workflows, enabling high-quality content generation at scale. The FLUX Pro Finetuning API allows users to fine-tune generative text-to-image models with five to 20 training images, optionally accompanied by text descriptions. Even Google’s search website has been infiltrated by a massive amount of dubious AI-generated art, to the degree that top results for famous human artists now sometimes include fake, AI-generated versions of their work.

While our technical team is proficient in interpreting nuclear-related diagrams, feedback from indigenous communities offers a fresh perspective for analyzing generative AI models, which is the focus of our next study. The objective of this study is to highlight the gaps in these technologies and pave the way for future improvements. The Michigan sand dunes are a well-known large-scale landmark for the researchers of this study, and as a result, a prompt related to these surroundings was chosen as “An oil painting of Michigan sand dunes”. For the second prompt, we generated four image outputs, out of which the image which portrayed the prompt with the highest technical accuracy was chosen. From these tests we observed that, DALL-E 2 created an image that most resembles an oil painting.

The authors acknowledge their potential implicit biases regarding nuclear energy, as approximately 75% of the team is either pursuing or holds a degree in nuclear engineering. As a result, some prompts and interpretations may unintentionally reflect personal beliefs. However, the team also includes members from interdisciplinary fields, such as data science and computer science, alongside nuclear engineering. Looking ahead, one of the group’s goals is to collaborate with indigenous communities to inform prompt generation and improve image accuracy. Input from those negatively affected by nuclear power is invaluable and brings insights that our researchers may not have considered.