This is another installment of On Tech: AI, a pop-up newsletter that teaches you about artificial intelligence, how it works, and how to use it.
A few months ago, my colleagues Cade Metz and Kevin Roose explained the inner workings of artificial intelligence, including chatbots like OpenAI’s ChatGPT, Microsoft’s Bing, and Google’s Bard. Now we are back with a new mission: to help you learn to use artificial intelligence to its full potential.
People from all walks of life—students, programmers, artists, and accountants—are experimenting with the use of artificial intelligence tools. Employers post job openings for people who know how to employ them. Very soon, if you haven’t already, you will have the opportunity to use artificial intelligence to streamline and improve your work and personal life.
As a personal technology columnist for The New York Times, I’m here to help you discover how to use these tools safely and responsibly to improve many aspects of your life.
I am going to spend today’s bulletin talking about two general criteria that will be useful in various situations.
Then, in the next few weeks, I’ll give you more specific advice for different aspects of your life, such as parenting and family life, work, organization in your personal life, learning/education, creativity and shopping.
Some common sense caveats to get you started:
-
If you are concerned about privacy, omit personal information such as your name and place of employment. Tech companies say that your data is used to train their systems, this means that other people may see your information.
-
Do not share confidential data. Your employer may have specific guidelines or restrictions, but in general, entering trade secrets or sensitive information is a very bad idea.
-
Hallucinations: Chatbots are based on a technology called “Large Linguistic Models” (LLMs), which gain their capabilities by analyzing vast amounts of digital text pulled from the internet. Many things online are wrong, and chatbots can repeat those falsehoods. Sometimes, while they’re trying to predict patterns from their huge amount of data from their training, they can make things up.
The instructions or ‘prompts’ of gold
ChatGPT, Bing, and Bard are some of the most popular AI chatbots. (To use ChatGPT, you’ll need to create an OpenAI account, and it requires a subscription to use its more advanced version. Bing asks you to use Microsoft’s Edge browser. For Bard, you must have a Google account.)
Although they seem easy to use—you type something in a box and you get answers!—asking the questions the wrong way will result in generic, unhelpful, and sometimes completely wrong answers.
It turns out that there is an art to typing the words and delimiting the precise frames to generate the most useful answers. I call them the instructions (or, in English, prompts) of gold.
Variants of the following strategies have been used by people who are making the most of chatbots:
“Act as if.” If you start your prompt with these magic words, the robot will emulate an expert. For example, if you type “Act like you’re a tutor for college entrance exams” or “Act like you’re a personal trainer,” the robots will be modeled after people in those occupations.
These instructions provide additional context for the AI to generate its response. In reality, artificial intelligence does not understand what it means to be a personal tutor or trainer. Rather, the prompt helps artificial intelligence to draw on specific statistical patterns from your training data.
A weak instruction without guidance will generate less useful results. If all you type is “What should I eat this week?”, the chatbot will give you a generic list of meals for a balanced diet, such as stir-fried turkey with a side of colorful vegetables for dinner (which, to me, sounds to something without much interest).
“Tell me what else you need to do this.” For more personalized results—for example, health tips for your particular body type or specific condition—invite the bot to request more information.
In the personal trainer example, an instruction might be: “Act like you are my personal trainer. Create a weekly exercise regimen and meal plan for me. Tell me what else you need to do this.” At that point, the robot could ask you for your age, height, weight, dietary restrictions, and health goals to create a meal plan and exercise routine for you to carry out for a week.
If you don’t get good answers on your first try, don’t give up right away. Better still, in the words of Ethan Mollick, a professor at the Wharton School of the University of Pennsylvania, he treats the bot as if it were a human intern: “When it makes a mistake, point it out and ask it to do better.” Be forgiving and patient and you are likely to get better results.
Thread your chatbot’s conversations
After you learn how to use the instructions, you can make your chatbot more useful over time. The key is to avoid treating the chatbot like a web search and starting with a new query each time. Instead, keep multiple conversation threads open and add instructions over time.
This strategy is easier with ChatGPT. Bing requires you to restart your conversations every so often, and Bard doesn’t make jumping between conversation threads that easy.
Natalie Choprasert, an entrepreneur from Sydney, Australia who advises companies on how to use Artificial Intelligence, uses ChatGPT as a business coach and executive assistant. She holds separate conversations for each of these features.
In the business coach thread, he shares information about his professional experience and the goals and problems of the company. In the case of the executive assistant, he shares information about his schedule, such as the clients he is meeting with.
“He builds up and trains properly, so when I ask him a question later, he’ll be in the right context and give me answers that are close to what I’m looking for,” Choprasert said.
Choprasert shared an additional gold command that he has trained his assistants to be very useful: Apply a frame of reference. recently read clockwork, a book on business creation. When he asked ChatGPT, the business coach, to advise him using the framework of clockworkwas delighted to see that she could incorporate principles from the book into an action plan to expand her business.
Brian X. Chen is a consumer technology columnist. He reviews products and writes Tech Fix, a column on solving technology-related problems. Before joining The Times in 2011, he reported on Apple and the wireless industry for Wired. @bxchen
Source: NYT Espanol