I just tested ChatGPT from OpenAI. My immediate reaction after five minutes is that the marketing profession, business world and society are not even close to ready for what is about to happen as a result of rapid advancements in artificial intelligence.
Everything we think we know about communications, creativity and strategy is going to be redefined and reimagined in the months (and years) ahead.
The closest comparison I can find in my mind to understand the magnitude of the moment would be to rewind to June 2007 when the iPhone was introduced. What is about to be built in the generative AI space is basically the equivalent of the emergence of the app ecosystem, which Apple says facilitated $643 billion in billings and sales worldwide in 2020.
With generative AI, we're talking about trillions of dollars in value and wealth creation over the next decade. And the barriers to entry (e.g. access to the tech, technical skills required to use AI, compute power, energy costs) are rapidly falling.
The rest of this post is the direct transcript (zero edits) of my first conversation with ChatGPT:
* * * * *
Me: What is ChatGPT?
ChatGPT: ChatGPT is a chatbot that uses the GPT-3 (Generative Pretrained Transformer-3) language model to generate human-like responses to user inputs in natural language. It can be trained on specific topics or domains to provide more personalized and accurate responses. ChatGPT is designed to improve the natural language processing capabilities of chatbots and enhance the user experience of conversational AI.
Me: How can marketers use ChatGPT?
ChatGPT: Marketers can use ChatGPT to improve their customer service and engagement by providing personalized and accurate responses to customer inquiries. ChatGPT can be integrated into a company's customer service platform to handle routine or frequently asked questions, allowing human agents to focus on more complex issues. It can also be used in marketing campaigns to engage with customers in a more conversational and interactive way. Additionally, ChatGPT can be trained on specific products or services to provide detailed information and answer potential customer questions.
An excerpt from my conversation with ChatGPT.
Me: How can writers use ChatGPT?
ChatGPT: Writers can use ChatGPT to help generate ideas and inspiration for their writing. By providing ChatGPT with prompts or starting sentences, writers can get creative and unique suggestions for their stories, articles, or other written works. ChatGPT can also be used to help with editing and proofreading by providing suggestions for alternate words or phrases. Additionally, ChatGPT can be trained on specific writing styles or genres to provide more tailored suggestions.
Me: Will ChatGPT make current conversational agents on company websites obsolete?
ChatGPT: It is unlikely that ChatGPT will completely replace current conversational agents on company websites. While ChatGPT has advanced language processing capabilities and can provide personalized and accurate responses, it is still a machine learning model and may not be able to handle all types of customer inquiries or complex issues. Human customer service agents will likely still be necessary for handling more complex or sensitive issues. Additionally, ChatGPT may not be able to fully replicate the human-like interactions and emotional intelligence of current conversational agents.
Me: How could ChatGPT change the way we search for information? Could it replace Google Search?
ChatGPT: ChatGPT could potentially change the way we search for information by providing more personalized and conversational responses to our queries. Instead of typing out keywords and phrases, we could simply ask ChatGPT a question in natural language and receive a detailed and accurate response. However, it is unlikely that ChatGPT could fully replace Google Search, as it does not have the same level of access to vast amounts of information and data. Additionally, ChatGPT is designed for more conversational interactions, while Google Search is primarily geared towards providing relevant web pages and results.
Me: How will all these advancements in generative AI, specifically language and image generation, impact the economy and jobs?
ChatGPT: Advancements in generative AI, specifically language and image generation, are likely to have significant impacts on the economy and the job market. On the one hand, these technologies can improve productivity and efficiency by automating tasks that were previously performed by humans. This can result in cost savings and increased competitiveness for businesses. On the other hand, there is a potential for job displacement as certain tasks and roles become redundant. It will be important for businesses and individuals to adapt and develop new skills in order to remain relevant and competitive in the changing job market.
Me: Do you see generative AI having a net positive effect on creative professionals, such as writers and designers?
ChatGPT: It is difficult to predict the exact impact of generative AI on creative professionals, such as writers and designers. On the one hand, generative AI can provide writers and designers with new tools and inspiration, allowing them to create more unique and creative works. It can also assist with tasks such as proofreading and editing, freeing up more time for the creative process. On the other hand, generative AI may also pose a threat to the job market for these professions, as it has the potential to automate certain tasks and potentially create competition for human creatives. It will be important for writers and designers to adapt and incorporate generative AI into their work in order to remain relevant and competitive.
Me: Is AI creative?
ChatGPT: AI has the ability to generate creative output, but it does not possess the same level of creativity as a human. AI can be trained on large amounts of data and can generate novel ideas and output based on this training. However, it lacks the ability to truly innovate and create something truly original and novel, as it is ultimately limited by the data it has been trained on and the algorithms it uses. AI can assist and enhance human creativity, but it cannot fully replicate the human creative process.
Me: What steps should marketers take to understand the power and potential of large language models (LLMs) in general, and GPT-3 (and ChatGPT) in particular?
ChatGPT: Marketers interested in understanding the power and potential of large language models (LLMs), such as GPT-3 and ChatGPT, should take the following steps:
- Research and understand the capabilities and limitations of LLMs, such as GPT-3, in terms of natural language processing and generating human-like responses.
- Explore the different applications and uses of LLMs in marketing, such as customer service and engagement, content creation, and personalization.
- Test and experiment with LLMs, such as GPT-3 and ChatGPT, to understand their capabilities and potential for marketing purposes. This can include training the models on specific topics or domains, and testing their responses to various inputs.
- Monitor and stay up-to-date on the latest developments and advancements in LLM technology, as well as any potential challenges or limitations.
- Collaborate and engage with other marketers and experts in the field to learn from their experiences and insights.
Me: Do you have any final thoughts that could give people some peace of mind about AI and the near-term future?
ChatGPT: While AI has the potential to bring many benefits and advancements, it is important to recognize and address the potential challenges and limitations. It is important to carefully consider the ethical and societal implications of AI, and ensure that its development and use is aligned with our values and goals. It is also crucial to invest in and support research and development in AI, in order to continue advancing its capabilities and addressing its limitations. By doing so, we can harness the power of AI for the benefit of all, and mitigate any potential negative impacts.
* * * * *
You can try it for yourself at https://chat.openai.com.
IMPORTANT DISCLAIMER About ChatGPT and Language Generation
As the mass market of professionals and leaders start to experiment with generative AI technology, many will make the mistake of assuming you don’t need the human in the loop, especially for language generation.
Shortcuts will be taken, and inaccurate content will be published by marketers and writers who don’t understand how the underlying technology works.
For example, since the ChatGPT outputs don't cite sources, you have to be very careful about names, locations, dates, numbers, "facts," etc. What may seem like authoritative and correct content, can easily include false information. Not by intention, but by design. It's just how the AI learns and produces content, at least in its current form.
Comprehension of what AI is and what it is capable of doing is critical for companies to get this right. We need more industry leaders experimenting and sharing, but we also need more leaders who fully grasp the power and potential of AI to be use for good, and for evil.
This stuff is going to move very fast, and the business world is not even close to prepared for what's to come in 2023.
If you want to be ready for what's next, we have a new Piloting AI for Marketers online course series that was built as a step-by-step learning journey for understanding and adopting AI.
Paul Roetzer is founder and CEO of Marketing AI Institute. He is the author of Marketing Artificial Intelligence (Matt Holt Books, 2022) The Marketing Performance Blueprint (Wiley, 2014) and The Marketing Agency Blueprint (Wiley, 2012); and creator of the Marketing AI Conference (MAICON).