Building Trust While Utilizing Generative AI

Katie Fiechter

Artificial Intelligence has become quite the buzzword this year– especially when talking about the generative AI chatbot ChatGPT. We’ve heard about generative AI being used for customer service, product descriptions, content creation, and even producing artwork, but how legitimate is it? And how can businesses utilize this new technology while remaining authentic? 

What is Generative AI?

Generative AI is a type of AI that can create images, videos, audio, text, and even code through learning from data that already exists, resulting in new and seemingly endless possibilities. Take Adobe for example, they’ve released Adobe Firefly– a generative AI made specifically for creators. The idea is for creators to give Firefly a simple text prompt and it will create fully customizable content that will come from the company's own content library. Adobe’s goal with Firefly is to help creators expand upon the natural creativity and ideas they already have and provide practical advantages. Many different iterations of generative AI are being created like Soundraw which can generate royalty-free music, or Podcastle, an audio recording and editing platform that creates studio-quality recordings. The things you can do with AI feel endless. 

How does this impact creators? 

Since AI learns from data that already exists, is it really producing unique, original content? How are original creators compensated if AI is pulling elements from images they’ve created? Adobe is promising that creators will be compensated and is even working on a “do not train” tag which would allow creators to opt-out of having their work used to train Firefly. This is an extremely gray area for generative AI since it can be argued that AI is learning like a human does– taking in information that already exists and producing a new version or variation. 

To no surprise, lawsuits for AI-generated imagery are already popping up which is difficult with something so new and rapidly evolving. The main argument is that artists' work is being scraped from the internet and AI tools are learning from these pieces without the artists' consent, bringing up the issue of copyright infringement and compensation. Getty Images has even filed a lawsuit over copyright infringement by Stability AI (a text-to-image model). Getty Images claims that Stability AI used Getty’s copyrighted images, associated text, and meta-data to train its AI tool without permission or compensation. In some cases, the images that Stability AI produced would remove the Getty watermark. 

As of now, works created by AI are not protected by copyright because the fair use law permits the use of copyrighted material under certain circumstances without needing permission from the owner. These types of lawsuits are new for everyone and with AI rapidly evolving, it will be some time before there are resolutions to these cases, as well as proper rules and regulations enacted to avoid similar problems in the future.

How are businesses taking advantage of Generative AI?
Over three-quarters of businesses (77%) have already adopted or implemented some type of AI solution in their organization. Many businesses are starting to recognize the power of AI and how impactful it can be to combine human insights with technology to produce something new at a rapid pace. Generative AI has proven itself to be a valuable asset across a variety of different industries and use cases.

Some businesses are leveraging generative AI for:

  • Content Creation: Businesses are using generative AI to automate content creation. AI-powered chatbots can generate responses to customers, and AI-generated copy can create product descriptions, ad copy, emails, and even blog posts. 

  • Design: Generative AI can create unique product designs. For example, creating new clothing designs, logos, or marketing materials.

  • Personalization: Generative AI algorithms can analyze customer data and use it to generate personalized product recommendations, offers, and discounts. 

  • Image and Video Processing: Generative AI can generate realistic images and videos which could help businesses create digital content rapidly, without the need for additional resources and equipment. 

  • Customer Service: Businesses can use generative AI to create chatbots that can interact with customers and provide support for basic questions. 

Overall, generative AI is helping businesses automate their processes which will help improve their operations and offer better service to their customers. 

How does a brand balance leveraging AI & remaining authentic to its audience?

A few months ago Levi’s announced that they will start using AI to generate diverse models wearing their clothes to make their brand feel more inclusive. This move by Levi’s received a lot of public backlash, not welcome for a brand that prides itself on being the “purest wearable form of authentic self-expression.” Not only does this hinder the relatively limited opportunities for real-life models of color and plus-sized models to get hired, but it also goes against the company's core values. This is a good example of a brand that is still trying to find a good balance between leveraging this revolutionary technology, but also remaining true to themselves and true to what their customers expect from them. Instead of Levi’s utilizing AI to generate diverse models wearing their clothes, maybe AI would be better suited for helping designers create and ideate new clothing items or assist with implementing personalization across their website.

The best way for businesses to take advantage of AI while remaining authentic is to treat it as a tool that will enhance the work that they are doing as opposed to a replacement that will push people out of their jobs. In fact, 84% of businesses running AI applications do so with an in-house team. When utilized with human talent and insight, AI can help take people down a path they may not have considered, help foster new ideas, and help speed processes along which would be extremely valuable for time-sensitive work.

When it comes to generative AI, it is not a perfect piece of technology and it is not the end all be all. There is still immense room for error since it is scraping data across the internet, a place where misinformation runs rampant. AI can be effectively leveraged as long as there is fact-checking happening to compensate for the human error that exists in the information AI is learning. There is a responsibility on those who are utilizing AI. And businesses that understand and acknowledge this and incorporate a formal review process will build more trust with their customers through their transparency.