In Blog 1, we defined some of the differences between AI & Gen AI. Generative AI interacts with Natural Language and can generate responses, as opposed to AI, it is narrower in scope and limited to specific tasks. I’d like to explore a few topics that have come up in conversations—whether from my colleagues or personal life—that warrant a discussion here.
There are three main concerns I hear about Gen AI.
- Privacy and Security
- Job Security
- Digital Work and Copyright Infringement
Notice a theme? It seems there is a bit of a worry that this new technology could take something away from us. Let’s take a closer look at each of these concerns.
Privacy and Security: The Digital Age Dilemma
Privacy and security are not new topics of discussion. As a digital native, privacy concerns aren’t something I’m unfamiliar with. I personally don’t have too much of an issue with it because it’s all I’ve ever known, but I understand why others have kept security issues top of mind.
Explaining privacy and security as it relates to Gen AI, the first thing that comes to mind is the big data case study with Target and predictive analytics back in 2012.
- Based on a young woman’s search and purchase history, Target predicted she was pregnant.
- She was sent coupons to her house where she lived with her parents for baby items.
- Her father was unaware of any pregnancy and complained to Target.
- It turned out that the data analysis Target had done was correct, revealing the pregnancy before the family knew.
The Target Case predates Gen AI, but based on Gen AI, being a probabilistic model; it’s doing this same thing on a much larger scale. If you’re concerned about privacy and security then you should check on your personal digital footprint, what your contract says for sharing data with other companies, and things of that nature.
The European Union continues to lead the regulatory charge for digital rights. In 2016 the EU enacted GDPR (General Data Protection Regulation). GDPR is a series of protections for your personal data. The EU is working to enact newer legislation for Gen AI to protect us from bad actors.
EU AI Act
- Categorizes technologies by risk, Unacceptable (social scoring), High Risk (regulating advertising), Limited Risk (transparency requirements), and Minimal Risk (unregulated)
- Artificial Intelligence-generated content needs to be labeled
- AI users need to demonstrate the model is without bias
- AI users must tell clients and consumers they are using AI
- It will be ~ 1 ½ years until the act is ratified into law
Job Security: The Human in the Loop
This is probably the most pervasive concern. And I understand why, it’s scary when you type a few sentences into a model and you get something back that is impressive… so impressive, it makes you wonder. It’s crucial to understand that while AI may change how we perform our jobs and how we work, it’s more than likely going to create more opportunities than it will take away.
From a historical point of view, one of the most disruptive technologies ever was the telegraph. It might not be obvious at first, but it was. For all human history, the way we communicated was only as fast as a horse could ride. Then suddenly, we could communicate at near the speed of light. While there were jobs that were lost, horse riders and couriers, the new technology created so many more opportunities.
AI is changing how we work, but not every job will be as dramatically impacted as others. Paul Baier, Jimmy Hexter, and John J. Sviokla wrote an amazing piece in the Harvard Business Review, Where Should Your Company Start with GenAI? In it, they introduce an idea called the WINS framework. It’s a matrix that measures how quickly your industry will be impacted based on the digitization of Words, Images, Numbers and Sounds (WINS). If You’re In the Crucible – your level of digitization is high and the percentage of WINS work is high – then you are likely to be greatly affected by Gen AI, for example, a marketer or a lawyer. If you’re on the other side of the matrix, in the balcony, then you’re not going to have to worry about learning this technology as quickly – for example, an auto mechanic.
Table 1 WINS Framework by Paul Baier, Jimmy Hexter, and John J. Sviokla
**Created with Claude 3.5 Sonnet Artifacts**
If you are In the Crucible, you likely already have started experimenting with Gen AI and are familiar with tools like Chat-GPT, the most popular Gen AI tool available today. Chat GPT is an incredible asset for the work you do, but it’s not an all-encompassing technology. It can’t replace human skills like developing and executing complex strategies, it doesn’t have any emotional intelligence, and it can’t think creatively. There needs to be a human in the loop writing the prompt and checking the work.
Humans in the loop will be the basis for how we use gen AI, which will impart a retraining of labor. Workers, some more than others, will need to be retrained in how they work. If you don’t retrain this muscle of labor with AI as a part of the process and workflow, then you will be left behind. Humans and AI together in harmony will make us more effective.
Digital Work: All Boats in the Harbor Rise
I’ve spoken with people about digital work and being worried about the quality and what’s human and what’s AI. This tech will not pull people away from that work, but instead, it will democratize creativity and allow more people to share their imagination who otherwise couldn’t.
For example, I’ve always thought of myself as having a creative imagination but felt that I lacked the skills to bring what’s in my head onto paper, canvas, or screen. Now with some creative AI tools like Dalle, Midjourney, and Runway to name a few, more people like me can create and share our imaginations and ideas resulting in the baseline for quality creative thought to be higher. All boats in the harbor will rise.
A democratization of creativity like this is much like when social media exploded and allowed regular people to share creatively with a larger audience. What previously would have required a director, writer, or producer, could now be accomplished by a regular person with an idea and a YouTube channel.
In 2022, there was a controversy with Gen AI artwork at a digital art show in Colorado. A gentleman won the first prize by submitting a largely AI-generated piece he called “Théâtre d’Opéra Spatial”. He used a complex prompt in Midjourney, some fine-tuning in Photoshop, and upscaled the image with another AI tool. Now he did win the art show but when it came out it was generated by AI, he wasn’t able to copyright the work because it didn’t have enough human authorship. The problem persists to this day. It’s the Wild West right now with this tech. The US Copyright and Patent Office is still working on how to handle generated art and what the models have been trained on. I look at the piece and think that this tool really can raise the bar of human creativity, which in turn makes the world a better place.
Figure 1 Théâtre d’Opéra Spatial, an AI-generated image that won first place at the Colorado State Fair annual art competition
©2022 Jason M. Allen
That was a lot… there are a lot of challenges to deal with. Some have even carried over from Internet 2.0. but with these challenges come a bounty of opportunities. What can you do today to be a part of the Gen AI revolution?
- Embrace continuous learning: stay informed about the rapidly changing landscape of AI.
- Focus on human skills: Emphasize emotional intelligence, creativity, communication, and complex strategy.
- Collaborate with AI: USE IT, look for ways to add it into your workflow if you’re allowed. Make mistakes, learn from them and do it again, fast.
- Adopt an “AI-first” Mindset: Allie K. Miller is a top LinkedIn voice on Gen AI, she said in a lecture one day that she has a sticky note on her desk that reads, “Can I use AI?” This “AI-first” mindset will help you professionally.
All that being said, if you do these things, you will not only be a part of this Gen AI Wild West, but you’ll thrive as a professional with a new power tool in your tool belt. So, start today if you haven’t… after all, Gen AI is the dumbest it will ever be today. Imagine what it will be like tomorrow.