Initial Industry Reactions
Given its extraordinary potential, ChatGPT and its successors will undoubtedly shake up many industries. An early example was none other than Google, specifically its search engine. As early as December of 2022, Google went on “Code Red,” due to ChatGPT’s ability to deliver direct answers, not just endless lists of links. According to CNET, CEO Sundar Pichai reported that he “upended the work of numerous groups inside the company to respond to the threat that ChatGPT poses.” Microsoft entered into a $10B partnership with Open AI and enhanced its own search engine, Bing, with ChatGPT. Google attempted to answer in kind, but a botched demonstration of its language model, Bard, led to a 7% drop in its stock price.
StackOverflow, the platform for a massive rated collection of programming questions and answers, banned ChatGPT, citing high volume and low quality of answers. The use of ChatGPT was banned or curtailed by Amazon, J.P. Morgan, and many universities.
Other companies responded more positively. The most obvious example is Microsoft, which partnered with ChatGPT for $10B. ChatGPT is now incorporated into Bing, its search engine, into the Azure OpenAI service, and to the Microsoft 365 Office suite. Soon, the days of interacting with a computer via mouse clicks and screen taps may be a thing of the past, giving way to conversations with applications in a similar way that we talk with each other. Github, owned by Microsoft, now offers CoPilot, which integrates ChatGPT to help programmers create code.
Stripe, the payment processing platform, already a user of GPT-3 to perform tasks such as routing tickets and summarizing user questions, asked a hundred of its employees to propose features and functionality using GPT-4 to incorporate into their platform. One idea arising from this challenge was to use GPT-4 as a virtual assistant that could summarize Stripe’s extensive technical documentation to solve a user question or issue. Other ideas addressed fraud: GPT-4 could identify potentially coordinated activity from malicious actors, or it could analyze the syntax of Discord posts to determine users previously kicked off the platform trying to ingratiate themselves back into the good graces of Stripe’s communication team.
ChatGPT’s Influence on Industries
Clearly, no one has a definitive answer of the effects short-, medium-, or especially long- term of ChatGPT. There seem to be as many opinions about ChatGPT as people talking about it, and everyone is talking. Perspectives run the gamut from open hostility (Nick Cave) to caution to open-armed acceptance (Satya Nadella).
Although there is spirited debate over how our professions will be influenced, almost everyone agrees that most industries will be affected in some way, and many will likely be disrupted. So which industries are most exposed? Ed Felton, Professor of Computer Science and Public Affairs at Princeton University, along with Manav Raj (U of Pennsylvania) and Robert Esamans (NY University) quantify and answer this question in their article, “How will Language Modelers like ChatGPT Affect Occupations and Industries?” Using a metric they call the “AI Occupational Exposure” (see the paper for the methodology), they find that the top twenty industries most exposed to language modeling include legal services and securities, commodities, investment services, and software publishers. They point out, “[t]he term ‘exposure’ is used so as to be agnostic as to the effects of AI on the occupation, which could involve substitution or augmentation depending on various factors associated with the occupation itself.”
Which brings us to the crux of the issue. Augmentation or substitution? No doubt, we will see both. Ajay Agrwal, Joshua Gans, and Avi Goldfarb, co-authors of Power and Prediction: The Disruptive Economics of Artificial Intelligence, wrote an insightful article for the Harvard Business Review, “ChatGPT and How AI Disrupts Industries.” They write:
“As AI continues to improve, more and more current jobs will be threatened by automation. But AI presents opportunities as well and will create new jobs and different kinds of organizations… These recent advances in AI will surely usher in a period of hardship and economic pain for some whose jobs are directly impacted and who find it hard to adapt — what economists euphemistically call ‘adjustment costs.’ However, the genie is not going back into the bottle… we must ask what new systems can be built with these new tools.”
Therefore, the question is not whether we will be affected – we will – but how to use the technology to augment and expand our current roles.
How ChatGPT Might Be Used Across Industries
In his article, “The Age of AI has begun” Bill Gates says,
“As computing power gets cheaper, GPT’s ability to express ideas will increasingly be like having a white-collar worker available to help you with various tasks. Microsoft describes this as having a co-pilot. Fully incorporated into products like Office, AI will enhance your work—for example by helping with writing emails and managing your inbox…Eventually your main way of controlling a computer will no longer be pointing and clicking or tapping on menus and dialogue boxes. Instead, you’ll be able to write a request in plain English… In addition, advances in AI will enable the creation of a personal agent. Think of it as a digital personal assistant: It will see your latest emails, know about the meetings you attend, read what you read, and read the things you don’t want to bother with. This will both improve your work on the tasks you want to do and free you from the ones you don’t want to do.”
ChatGPT will also assist software development. In a sense, this is a natural extension of the evolution of software development. We’ve gone from manually wiring giant computers to writing micro-instructions in assembly to writing instructions in a “human like” language – originally C, then Pascal, then a whole host of languages, leading to extremely user friendly languages used today, such as Python – to connecting open source libraries of pre-existing routines to accomplish nearly any task. Next up: Copilot, which uses ChatGPT to help build a code. Taking this a step further, is it a stretch to imagine speaking your desires to an interface and it building software routines or visualizations for you on the fly? Note, however, even in the scenario, there’s a human being running the show. You need someone to tell the computer what to do, and it’s only common sense wisdom to have human beings validate and verify the results.
Industry Examples for ChatGPT
Let’s ideate on what ChatGPT might do for business. For a starting point for this exercise, I typed the prompt, “What can I do with ChatGPT?” into ChatGPT’s interface. It responded with the following.
“As an AI language model, I can assist you with a wide range of tasks, including:
1. Answering questions: You can ask me any question you have, and I'll do my best to provide you with an accurate and helpful response.
2. Generating text: I can help you write essays, reports, articles, and more. Just tell me what you want to write about, and I'll generate some text for you.
3. Translating text: I can help you translate text from one language to another.
4. Offering recommendations: I can offer recommendations on books, movies, TV shows, and more based on your preferences.
5. Providing general knowledge: I can provide you with general knowledge on a wide range of topics, such as history, science, and culture.
6. Entertaining you: I can tell you jokes, riddles, and stories.
Overall, I'm here to assist you in any way I can with my language processing capabilities.”
Let’s expand on a couple of these items. “Answering questions” implies that ChatGPT can summarize text, which in fact, it can. A possible use case might be to summarize steps of a process, say customer service phone trees (IVR), to pass along to the next step. For instance, it’s frustrating to talk to several bot or human agents and repeat your problem over and over. Wouldn’t it be nice if a summary of previous chats was passed to the next agent so that they already know what’s going on?
Another of ChatGPT’s capabilities is generating text. An obvious example is marketing text. Here is ChatGPT’s response to “Write text to advertise the benefits of a cafe that offers a variety of soups and sandwiches, flatbread pizzas, and delicious pastries. Mention the benefits of a low-cost coffee membership in which you can get a free coffee or tea every day.”
“Welcome to our cozy café, where we pride ourselves on serving up a variety of delicious soups, sandwiches, flatbread pizzas, and pastries to satisfy your cravings. Whether you're in the mood for something savory or sweet, we've got you covered.
Our soups are made fresh daily with wholesome ingredients, and our sandwiches are crafted with artisanal bread and premium meats and cheeses. Our flatbread pizzas are a customer favorite, with a crispy crust and mouth-watering toppings that are sure to please.
But that's not all. We also offer a low-cost coffee membership that gives you a free coffee or tea every day. Imagine starting your day with a steaming cup of our house blend coffee or a soothing cup of tea, all while enjoying the warm and inviting atmosphere of our café.
And let's not forget about our pastries. From flaky croissants to rich chocolate muffins, our selection of pastries will satisfy your sweet tooth and leave you wanting more.
So why not come by and see what all the fuss is about? Our friendly staff will be happy to welcome you and help you choose from our mouth-watering menu items. And with our coffee membership, you can enjoy your favorite hot beverage every day without breaking the bank. We look forward to serving you soon!”
If I didn’t already patronize a cafe fitting that description three to five times a week, I would now!
Final Thoughts
Clearly, ChatGPT can be used to level up any number of business and technological needs. It will be interesting to see the creative ways that businesses find to use it to help them in the marketplace.
____________
This post is part of a series about the explosion of AI happening right now. Check out our previous post from our Decision Science team about the Google-ChatGPT showdown.