As the generative AI field heats up, consumer-facing chatbots are answering questions about business strategy, designing study guides for math classes, offering advice on salary negotiations, and even making marriage vows. We even put up And things are just getting started.
OpenAI’s ChatGPT, Google’s Bard, Microsoft’s Bing, and Anthropic’s Claude are some of the leading chatbots today, with more to come over the next year. In the venture capital space, generative AI deals totaled $1.69 billion globally in the first quarter. PitchBook data shows that deal value has surged 130% this year from $730 million in the previous quarter, and even though a further $10.68 billion worth of deals have been announced, the first quarter is still down. It says it’s not completed.
Two months after ChatGPT’s launch, ChatGPT surpassed 100 million monthly active users, breaking the record for the fastest growing consumer application of all time. “The adoption rate is phenomenal. Frankly, I’ve never seen anything like it. Gartner’s vice president of research, Burke, told his CNBC. “From our November 30th release to now, our inquiry volume has skyrocketed like a hockey stick. Every client wants to know about generative AI and his ChatGPT.”
These types of chatbots are built on large scale language models (LLM). LLM is a machine learning tool that uses large amounts of internet data to recognize patterns and generate human-like language. If you’re a beginner, many of the sources we spoke to agreed that the best way to start using chatbots is to dive in and try them out.
“People spend too much time trying to find the perfect prompt. 80% of them are just using the prompt interactively,” study the impact of AI on work and education. says Ethan Morrick, associate professor at the Wharton School of the University of Pennsylvania. told CNBC.
Here are some tips from the pros.
Keep data privacy in mind.
When using chatbots such as ChatGPT or Bard, the information you enter (what you type, what you receive in responses, and changes you request) may be used to train future models. OpenAI says the same thing in its own words. While some companies offer a way to opt-out (OpenAI allows this under “data control” in ChatGPT settings), chatbots’ It’s best not to share sensitive or personal data in conversations. For example, a bug in his ChatGPT in March temporarily allowed users to view parts of each other’s conversation history.
“If you don’t want to post on Facebook, don’t post on ChatGPT,” Burke said. “Please consider anything you enter into ChatGPT as public information.”
provide context.
To make the best use of your time, give your chatbot some context about how it behaves in this scenario and who to provide this information to. For example, you can write out the persona you want your chatbot to assume in this scenario: [marketer, teacher, philosopher, etc.]You can also add context like “I am a”. [client, student, beginner, etc.]This can save time by directly telling the chatbot what kind of role it should play and through which “lens” to pass information in a useful way. There is a nature.
For example, if you’re a creative consultant looking for a chatbot to help analyze your company’s logo, you can enter: “Act as if you are a graphic designer studying logo design for a company. I am the client.” thinking person. Generates an analysis of the ‘best’ company logos for publicly traded companies and why they are considered a good choice.
“If you asked Bird to write an inspirational speech, his response might be a little more general. If you ask, you’ll probably get a much better response,” Cissy Google vice president Xiao told CNBC.
Let your chatbot do all the work.
The best way to get what you need is to ask the chatbot itself for advice. Even if you’re asking what you can do as a user or how to best phrase your prompt.
“A simple question of what can be done will give most people a list of things that would actually surprise them,” Burke said.
You can also abuse the system by asking questions like, “What’s the best way to get help building a shopping list?” Alternatively, assign a prompt creation job to your chatbot. For example, “Your job is to generate the best and most efficient prompts for ChatGPT. Generate a list of the best prompts to ask ChatGPT for recipes for a healthy one-pot dinner.”
Ask for help with brainstorming.
Whether you’re looking for vacation destinations, date ideas, poetry prompts, or content strategies to go viral on social media, many people use chatbots as a starting point for brainstorming sessions.
“The biggest thing I find these helpful is that they inspire me as a user and help me learn things that I wouldn’t necessarily have thought of on my own,” said AI Research. said Josh Albrecht, CTO of startup General Intelligent. told CNBC. “Maybe that’s why they are called generative AI.
Create a crash course.
Suppose you are trying to learn geometry and consider yourself a beginner. You can start learning by asking a chatbot questions like ‘Explain the basics of geometry like a beginner’ or ‘Explain the Pythagorean theorem like a 5-year-old’.
If you’re looking for something more extensive, you can ask the chatbot to create a “crash course” for you by specifying a time period (3 days, 1 week, 1 month) or a desired amount of time. Spend time learning new skills. You could write something like, “I’m a beginner who wants to learn how to skateboard. Please make a two-week plan for him on how to learn to skateboard and kickflip.”
To extend your learning plan beyond chatbots, you can also ask for a list of the most important books on the topic, the most influential people in the field, and other resources to help improve your skill set. increase.
Don’t be afraid to leave a note and ask for changes.
“The worst thing that can happen if you’re actually trying to use ChatGPT’s output is [to] Just ask one question and then walk away,” Morrick said. “You get a very generic output. You have to interact with it.”
Sometimes you can’t choose the perfect prompt or your chatbot doesn’t produce the output you were looking for, and that’s okay. You can also tailor the information to make it even more helpful, such as asking follow-up questions such as “Does this sound uncommon?” or “Could you make the first paragraph more interesting?”
Add some salt to everything and take it.
Chatbots have a documented tendency to fabricate information, especially if the training data doesn’t fully cover the area you’re asking, so it’s important to take everything with a grain of salt. Suppose you want a biography of Albert Einstein. The chatbot might tell you that this famous scientist has written a book called How To Be Smarter, but unfortunately he hasn’t. Also, large language models are best at pattern recognition because they are trained on a wide range of internet. This means that they can produce biased output or misinformation based on your training data.
“Where information is scarce, it’s just hoaxed,” Burke said, adding, “These hallucinations are very convincing… you can’t rely on these models to always give you accurate information.” ‘ added.
Experiment and try different approaches.
Whether you’re asking a chatbot to generate a list of action items from a meeting recording or translating something from English to Tagalog, generative AI has a myriad of broad use cases. So when using a chatbot, it’s worth thinking about what you want to learn or need help with and experimenting with how well the system works.
“AI is a general-purpose technology and has many capabilities, so no matter what field you are in or what job you are doing, you will be able to do your job more than anyone else on the planet. The idea is to have a different impact on the aspects of ,” Morrick said. “It’s important to think about how you want to use it…you have to find a way to manipulate the system…and the only way to do that is to experiment.”
Subscribe to CNBC on YouTube.
