Is AI hurtful or helpful?
Is AI hurtful or helpful?
Can AI be helpful in my job? The short answer is yes, there are some tasks that AI can be really helpful with. One example, for this white-paper I asked ChatGPT “Can you make a slide deck on the helpful points of AI?” and it output the following.
Slide 1: Title
Catchy title introducing the topic of the presentation.
Slide 2: Introduction
Briefly introduce yourself and provide an overview of the presentation’s purpose.
State the benefits of AI in the workplace.
Slide 3: Automation and Efficiency
Highlight how AI automates repetitive tasks, saving time and increasing productivity.
Mention specific examples like data entry, document processing, and email filtering.
Slide 4: Intelligent Virtual Assistants
Explain how AI-powered virtual assistants can handle routine queries, providing 24/7 support and freeing up human resources.
Describe how they can improve customer service and reduce workload on employees.
Slide 5: Data Analysis and Insights
Discuss how AI can analyze large volumes of data quickly, extracting meaningful insights and patterns.
Emphasize the benefits of data-driven decision-making and improved operational efficiency.
Slide 6: Natural Language Processing (NLP)
Explain the concept of NLP and its applications.
Highlight how NLP can enhance communication, enable language translation, and aid in document summarization.
Slide 7: Enhanced Security
Discuss how AI can contribute to office security by detecting and preventing potential threats.
Mention applications like network traffic analysis and cybersecurity breach detection.
Slide 8: Intelligent Document Management
Explain how AI can assist in organizing and managing documents efficiently.
Highlight features like automatic categorization, tagging, and information extraction.
Slide 9: Workflow Optimization
Discuss how AI can analyze workflow patterns, identify bottlenecks, and provide recommendations for process improvements.
Mention how it can optimize resource allocation and enhance overall productivity.
Slide 10: Predictive Analytics
Explain how AI can utilize historical data and machine learning algorithms to make accurate predictions.
Highlight applications like sales forecasting, demand planning, and customer behavior analysis.
Slide 11: Ethical Considerations
Emphasize the importance of ethical AI implementation, including transparency, accountability, and privacy.
Slide 12: Conclusion
Summarize the key points discussed throughout the presentation.
Reinforce the benefits of AI in the office environment.
Slide 13: Q&A
Allocate a slide for audience questions and provide answers or initiate a discussion.
Not a bad starting point but the presentation is missing a number of points, for example, providing Legal or Medical research summery assistance, or improve communications. My point is as a research assistant this gives me a good starting point. You might try using this approach for your next class assignment or presentation. Using AI as a research assistant on topics you know, allows you to focus on how the content should be organized and communicated and not the basic framework.
Remember the training model – Reddit and other web sources. Those sources will have valid and incorrect information, and AI can’t tell which one is valid and which isn’t. AI is not judging the data it is learning from, it is looking how often like-minded data exists. If all web data on a topic is false, then AI will give completely incorrect information.
I urge you to tread lightly when using AI on topics you do not know, do not trust that what ChatGPT or other AI sites produce is correct, it is not a super smart brain – it is summarizing data it has absorbed and it could have easily reached the wrong conclusion.
AI training is at the point of a precocious 5-year-old who has heard a cure word. The 5-year-old who hears a curse word can’t help but to repeat it and based on adult’s feedback senses something is different about the word, so they use it a lot, until confronted and schooled by an adult.
One attorney appeared who knew the material appeared to get lazy and assumed ChatGPT was a super-brain and included the output without review and without editing is now being called in front of hte court to explain why none of the references used in his submission where valid – https://www.nytimes.com/2023/06/08/nyregion/lawyer-chatgpt-sanctions.html.
So the best way to use AI is to use it as a sounding board, consultant, research assistant, not as an expert. Even human experts can be wrong, and AI can be very wrong. Properly used it can be a significant time saver.
Here is another example, what if I am having a dinner party for 4 and some of the guests are on the Mediterranean Diet, I have some basic items in my refrigerator, but I need some ideas. In the past I would have used a search engine search, which might then I would need to calculate an ingredients list, or I can ask ChatGPT; “I have eggs, milk, spices, and vegetables please give me a spicy recipe for 4 for dinner following the Mediterranean diet”. The result is;
Here’s a spicy Mediterranean-inspired recipe for dinner using eggs, milk, spices, and vegetables:
Spicy Shakshuka with Roasted Vegetables
Ingredients:
1 tablespoon olive oil
1 onion, finely chopped
2 garlic cloves, minced
1 red bell pepper, diced
1 yellow bell pepper, diced
1 zucchini, diced
1 eggplant, diced
1 teaspoon cumin
1 teaspoon paprika
1/2 teaspoon cayenne pepper (adjust to taste)
1 can (400g) diced tomatoes
Salt and pepper to taste
4-6 large eggs
Fresh parsley or cilantro, chopped (for garnish)
Instructions:
Preheat the oven to 400°F (200°C). Place the diced zucchini and eggplant on a baking sheet, drizzle with olive oil, and season with salt and pepper. Roast in the oven for 20-25 minutes until the vegetables are tender and lightly browned.
While the vegetables are roasting, heat olive oil in a large skillet or frying pan over medium heat. Add the chopped onion and sauté until it becomes translucent, about 5 minutes.
Add minced garlic, diced red and yellow bell peppers to the skillet, and cook for an additional 5 minutes until the peppers soften.
Stir in the cumin, paprika, and cayenne pepper, and cook for a minute to toast the spices.
Add the diced tomatoes to the skillet, season with salt and pepper, and simmer the mixture for about 10 minutes to allow the flavors to meld.
Using a spoon, create small wells in the tomato mixture and crack the eggs into the wells. Cover the skillet and cook for 5-8 minutes until the eggs are cooked to your desired doneness.
Remove the roasted vegetables from the oven and add them to the skillet, gently mixing them with the tomato and egg mixture.
Sprinkle the chopped parsley or cilantro over the top for garnish.
Serve the spicy shakshuka with roasted vegetables hot, accompanied by crusty bread or pita for dipping.
This recipe incorporates the bold flavors of Mediterranean spices, roasted vegetables, and eggs cooked in a rich tomato sauce.
There is no information if this is a good tasting recipe but since the AI training data included real recipes then most likely it will. Again, this is a reference, if I have experienced that garlic bothers my guest then I can substitute another spice or I can AI for a revised recipe but my point is I am using AI as a reference tool and then working from there.
These are two examples of how ChatGPT can be used as a tool or an assistant to help people process, but only when used correctly.