Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
AI-powered photo and video models aren’t human, but they have distinct “personalities,” according to the creators who use them. The new phenomenon is a reaction to the rapidly expanding generative AI industry, highlighting how creators manage a dizzying number of choices.
Generative AI has seen a huge growth spurt over the past few years, but it won’t happen until 2025 Image by Amnesty International, video Other generative media models have taken center stage. Much like how chatbots redefined text creation, these creative AI models are changing the process of content creation and creative work, for better or worse.
Google and OpenAI have long been leaders in the AI race. Before this year, they were known for them twin and ChatGPT Chat bots. now, I see 3, Nano banana and Sora 2 They have placed the tech giants firmly at the head of the pack among innovative AI models. New artificial intelligence innovations from Adobe And innovative startups in the field of artificial intelligence, such as RunwayBoth Becca and Loma have also strengthened this area this year.
For AI companies to remain competitive in a crowded market, generative media has evolved from a niche offering to a must-have. Companies are focusing on upgrading their AI models to maintain an edge and attract new users. Enhancements typically include creating detailed, higher-resolution content, and for video, including audio and lengthening clips. The hallucinations or errors disappear with each update of the model, which is part of the reason why it becomes increasingly difficult To discover content generated by artificial intelligence.
Overall, there weren’t a lot of options for creating AI content. When content creators have to choose a model to use, it’s no longer about which model will produce serviceable results. Now, there is debate about which one would be best suited for a particular project or task. As a result, each AI model now has its own personality.
Don’t miss any of our unbiased technical content and lab reviews. Add CNET As Google’s preferred source.
Humanize AI tools with personalities
Creators use the term “personas” colloquially – AI models are not human, and therefore do not have personalities. The term really refers to a model’s ability to handle specific tasks and its reputation for excelling in certain areas. It also indicates the individual style of each model.
“Creators are humanizing these tools,” said Tiffany Kiazi. “They call it a ‘creative tool’ or a ‘detail tool’ because they’re building actual relationships with their AI. It’s not just software anymore.” (@TechTiff), founder Artificial Intelligence Flow Clubwhich teaches people how to use artificial intelligence tools. “These personas help creatives build confidence using their tools, work through creative blocks, and find comfort in workflow.”
For creatives who use AI tools every day, choosing the right model has become an integral part of the creative process, just like choosing the right camera lens or paintbrush.
“Each model interprets the world differently; some models are cinematic, others are more surreal or dream-like,” he said. Dave ClarkDirector and Creative Director at The promises of artificial intelligencean artificial intelligence production studio. “The key for me is knowing how to take my creative vision and translate it into visual language prompts that allow me to achieve the art I want.”
There is a learning curve to discovering each model’s personality. Sometimes, it’s not specified by the company that created it; It varies between photos and videos, and between different generations of the same model. Part of that is due to how the models are created.
“Part of what we learn when we train our own models is at the end of the training process, you can show the model a certain style, and the model will more or less fit, or adapt to that style and essentially take on the personality,” said Alexandru Kostin, vice president of generative AI at Adobe. “So we see very stubborn models doing it. Others are trying to be more neutral.”
The training data used to create and refine the model also plays a role in developing the basic style of each model. For example, Adobe’s Firefly models are trained using licensed Adobe Stock images, which is why content generated with Firefly often has a stock-like appearance. (Costin said the company is working on fixing this to create more realistic output.)
What is the personality of each AI image and video model?
I’ve spent a lot of time with these AI models, and the creators I spoke with had similar ideas and experiences to mine when it comes to the personality of each model. Here are some of the most popular models’ personalities.
- Google I see 3 (Video): Cinematic, natural motion, high quality
- flow (Video): It excels in realism, especially in human features
- Runway (Video): A complete creative studio, great for those who need hands-on control
- Sora (Video): Useful for thinking, exploring, and memes for Sora social networking application
- Mid-flight (Photo and Video): The most creative models, best for artistic or stylized works
- Google Nano Banana (Image): Best for personality consistency, good for e-commerce and social media work
- Adobe Firefly Image Model 5 (Image): Commercially safe results for professional work
You’ll notice distinct personalities among the chatbots as well. ChatGPT is known for its affectionate and personable tone (sometimes Very annoyingly), while Claude is a preferred search tool, while Gemini is a convenient option for Google users. However, the different personalities of AI-powered photo and video models – styles, aesthetics, innate preferences, etc. – become more immediately apparent.
Although you can create almost any scene with AI-powered image and video generators, they are not the “all machines” that chatbots can be. Creators who use creative AI tools for professional work often need to leave a specific piece of content up to them. Understanding the personality of each model is crucial.
Benefits of using multiple models
The idea of jumping between AI models and programs may not seem appealing at first, but there are benefits to expanding your AI repertoire.
Clark and his team used a variety of artificial intelligence models for the new design Short film His name is brought out by my friend Zev. This hybrid filmmaking method, as Clark calls it, involved the team using AI tools like Adobe Firefly, Google’s Veo 3.1, and Luma’s Ray3, as well as traditional Adobe software, including Photoshop and Premiere Pro.
“By mixing multiple models, you get creative scope and precision, as if you had a team of specialists,” Clark said. “We can envision the story world much earlier, iterate faster, and make stronger creative choices before we step onto set.”
Some creators are loyal to specific AI tools and platforms and may be reluctant to expand. The idea of loyalty to AI is a bit misleading, Kiazi said. The creatives who get the best results are “tool-aware and goal-oriented.”
“The real benefit of a multi-model workflow is that you’re not forcing one tool to do everything. You’re taking advantage of the actual strengths of each model. This is not only more efficient. It gives you better results because you’re using the right tool for each specific part of your project,” Kiazi said.
Developed personalities
The concept of AI models with personalities is relatively new, thanks to the recent boom in models available to creators. But they are not fixed labels; A model’s reputation and personality can change over time. As new updates roll out, models that were previously known to be bad at a particular task may be improved.
This trend is another sign that AI is playing an increasing role in creative work. This is not true for all creators, as there are many who are against AI and do not want to use it. But for those interested, there were no more options.
Creating distinct personas for AI-powered photo and video models is one way or solution to help them choose the right tool to achieve better results – without wasting a lot of time and money on AI tools that are not a good fit.
Although generative media models have improved a lot, they are still not perfect. Adapting to the strengths and weaknesses of each model is intelligent workflow design, Kiazi said. Remembering that AI models are just tools is also important, Clark said.
“It’s the artist’s human expression — our personality and creative point of view — that really drives results,” Clark said. “It’s not about replacing the traditional process, it’s about expanding what’s possible and bringing imagination closer to the screen than ever before.”
(Disclosure: Ziff Davis, CNET’s parent company, in April filed a lawsuit against OpenAI, alleging that it infringed Ziff Davis’s copyrights in training and operating its AI systems.)