Industry Insights

Background image of user typing on a calculator with floating interface elements surrounding them

Pushing the Limits of Creativity: A Roundup of Common Generative AI Applications

March 1, 2023

Generative Artificial Intelligence (AI) is not just a fad, but a significant technological advancement that is poised to become ubiquitous across various industries. Today, most companies are already utilizing some form of generative AI, ranging from chatbots and virtual assistants to image and text generation tools. The likes of ChatGPT (for text), DALL-e2 (for images), OpenAI Codex (for code), Synthesia (for videos), and MusicLM (for music) are leading the way in the new era of creative expression.Despite the benefits offered by these AI-powered tools, there are concerns about their potential misuse and impact on human creativity. It is worth noting that the content generated by these tools still requires manual oversight and proofreading as they can make mistakes or even hallucinate information that isn't real. This blog will delve into the capabilities of each generative AI application area and the implications of its use.

Text Generation / ChatGPT

AI has immense potential to revolutionize the way we interact with customers and the efficiency of document creation and written text. Chatbots powered by AI can handle a wide range of problems and provide genuine benefits to customers. Additionally, tools like ChatGPT can assist in drafting and proofreading documents, providing suggestions for changes that can significantly enhance writing. However, while text-focused generative AI can be an excellent tool to augment a writer's work, there is still a risk associated with relying too heavily on it. Recent reports of fully made-up, or hallucinated, papers and links in ChatGPT output, which looked genuine but were not, highlight the importance of having clarity on when and how it's acceptable to use AI. In fact, publishers of thousands of scientific journals have banned the use of ChatGPT amid concerns that it could pollute academic literature with flawed and even fabricated research.

Image Generation / Dall-e2

To generate an image using DALL-e2, users provide a text prompt describing the image they want to create. The prompt can range from a simple description, such as "a pink sofa with yellow pillows," to a more complex one, like "a surreal landscape with floating islands and a blue sky." DALL-e2 then uses its neural network to generate an original image that matches the description as closely as possible. The generated image is an interpretation of the prompt based on the vast amount of data that DALL-e2 was trained on. This technology is particularly useful for creative and marketing teams that need a very specific image to fulfill a request without going through the hassle of setting up a custom photo shoot. However, it's worth noting that it's hard to predict what will be generated from any given prompt. In many cases, as with writing, experts take the image generated by the AI process and then apply their own tweaks. This approach allows for the creation of images that meet the exact specifications needed while still including human creativity and refinement.

Code Generation / OpenAI Codex

While the process of generating code is similar to generating human language, the applications are entirely different. Code generation AI, such as OpenAI’s Codex, can be highly effective for translating code from one language to another or creating new code for well-defined problems, saving developers significant amounts of time and effort. Moreover, coding is an area where AI can have a big impact, as it enables developers to increase productivity by automating repetitive and mundane coding tasks. However, there are concerns in the industry regarding the use of code generation AI, particularly around the confidentiality of data and intellectual property. The AI generators may claim rights to store and use any code uploaded for model training, which means that proprietary data, fields, and commentary might be accessible to others. Despite these concerns, code generation AI is gaining momentum and building acceptance.

Movie Generation / Synthesia

Taking image generation to the next level, generating video content is a more complex and intricate process that requires greater sophistication from AI. While movie generation AI is currently not as advanced as image generation AI, we can expect to see significant improvements in the future as technology continues to evolve. Tools like Synthesia have the potential to revolutionize the creation of marketing videos, eliminating the need for actors or influencers and allowing for seamless integration of corporate content and logos within the generated content. While human oversight and editing are still necessary to ensure high-quality video output, AI-generated videos provide a fast and cost-effective way to create content without requiring a large production team. As AI technology continues to improve, we will see an increasing number of use cases for movie generation AI across a wide range of industries.

Music Generation / MusicLM

Google has recently announced the development of MusicLM, an AI-powered tool that generates music based on text input. Using deep learning algorithms, MusicLM can analyze text and create original music based on the emotions and themes present in the text. The tool can produce high-quality compositions in a range of musical styles, including classical, jazz, and pop. While MusicLM is not the first AI-powered music generation tool, its integration with Google's existing music streaming and recommendation services could give it a significant edge in the market. However, there are concerns over the potential for AI-generated music to devalue the work of human composers and performers. Due to legal concerns, Google continues to work on the MusicLM AI and is exploring ways to address the legal and ethical issues before scheduling a future release.

Paving the Way for a New Era of Innovation and Progress

The future of generative AI is exciting, with new advancements being made every single day. While it is necessary to remain vigilant about the potential risks associated with these tools, there's no denying the significant benefits they may offer. As millions of people continue to explore the possibilities presented by generative AI, there will be further integration of AI into people’s personal and business lives. The recent rise of generative AI should be a major, defining moment in the evolution of technology.This post was developed with input from Bill Franks, internationally recognized thought leader, speaker, and author focused on data science & analytics.