A.I. in the workplace; navigating your benefits and risks

Extra Good’s goal is to explore relevant and useful topics, like this one on AI, for nonprofit organizations in our blogs, reports, and articles. By joining our mailing list, you can easily stay updated on our content offerings. 

A hot topic in the workplace, and rightfully so, is the use of AI. According to Forbes Advisor, a staggering 97% of business owners believe that ChatGPT will benefit their businesses. One in three businesses plan to use ChatGPT to create website content, while 44% aim to generate content in multiple languages. While many bosses dream of the cost-saving aspects of AI, including staff reductions, many workers are having nightmares, afraid their jobs may be on the chopping block.  Let’s take a deeper look into AI and the benefits and challenges this technology powerhouse brings to the table.

Starting with the basics: Artificial Intelligence vs Generative AI; what’s the difference? For the answer, I went directly to the source, Chat GPT, that is: 

Q. Me: What is the difference between AI and Generative AI?

A. Chat GPT:

AI, or artificial intelligence, is a broad field encompassing the development of machines or systems capable of performing tasks that typically require human intelligence. This can include a wide range of techniques, such as machine learning, natural language processing, computer vision, and more. AI systems can be designed to perform specific tasks based on predefined rules or by learning from data.

Generative AI is a subset of artificial intelligence that specifically focuses on creating or generating new content, such as images, text, music, and videos. Generative AI uses techniques like neural networks to produce original content that often mimics human-created output. These systems can create content autonomously after being trained on large datasets, allowing them to generate realistic and novel outputs.

In essence, while AI refers to the broader concept of machines performing intelligent tasks, generative AI is a specialized area within AI that focuses on creating new content or data.

While Generative AI has been shown to be quite impressive in the areas of content creation, healthcare applications, art, and other creative platforms, the idea that Generative AI can stand alone, meaning without human input, is hotly debated. Why is that? Well, once again, I asked Chat GPT.

Q. Me: What does generative AI have difficulty doing?

A. ChatGPT:

Generative AI, despite its incredible capabilities, still faces challenges in several areas:

1. 1. Understanding Context: While AI models have improved in understanding context, they can struggle with nuanced or complex contexts in language or situations, leading to generated content that might lack coherence or accuracy.
   2. Consistency and Coherence: Maintaining consistency and coherence in generated content remains a challenge, especially in longer pieces like essays or stories. There might be inconsistencies or lapses in logic within the generated content.
   3. Ethical Concerns: Generative AI raises ethical issues, particularly regarding generating fake content, deepfakes, or biased outputs. Ensuring responsible use and preventing misuse of this technology is a significant challenge.
   4. Generating Novelty: While generative models excel at mimicking existing patterns and generating content based on learned data, producing truly innovative or novel content remains a challenge. Originality beyond learned patterns is an ongoing area of research.
   5. Data Dependence and Bias: Generative models heavily rely on the data they’re trained on. Biases present in the training data can be reflected in generated content, perpetuating societal biases or inaccuracies if not addressed carefully.
   6. Fine-grained control: Providing precise control over what is generated can be challenging. For example, in image generation, ensuring specific attributes (like age or gender) are accurately controlled in the generated images remains an ongoing challenge.
   7. Resource Intensiveness: Training and running generative AI models can be computationally expensive and resource-intensive, limiting their accessibility and practicality for smaller organizations or individuals with limited resources.

Addressing these challenges requires ongoing research and development to improve the capabilities of generative AI while mitigating ethical concerns and biases, ultimately aiming for more reliable, controllable, and ethical use of this technology.

There you have it, straight from the source. Generative AI can be a timesaving first step in producing content, but it should not be the final step.  Additional research, validation, nuance, and other factors are essential in creating sound output. The Open Worldwide Application Security Project (OWASP) warns using AI chatbots, without proper oversight, can make organizations vulnerable to publishing misinformation or offensive content that results in reputational damage or even legal action.

Technology of all sorts, including Generative AI, are fabulous tools when used correctly; however, they are just that, tools. So, sleep well, knowing that while you may not be able to spit out obscure facts in a nanosecond or possess amazing grammar skills, you still have the upper hand in AI, and that is, you are human.

Copywriters and content developers may be particularly interested in using Generative AI as a drafting tool. Alan Tarr at “How to Write Copy” recently went in-depth on this topic. Take a look at his perspective on AI in the copywriting world.