It’s Time to Get Up to Speed on Generative AI
If you’ve been living on planet Earth since the generative artificial intelligence phenomenon known as ChatGPT launched late in 2022, you’ve probably heard the term “AI” more than your own name this year. Along with OpenAI’s next wave of generative AI called Bing chat and Google’s BARD, these new technologies are promising (or threatening depending on your perspective) to change nearly everything about how we live and work.
As a business owner or leader (even in a non-technical field) do you need to understand and use these new tools? The unequivocal answer is yes, or risk being left behind with typewriters and carbon paper. When ChatGPT went public, it reached 100 million users in two months and became the fastest-growing app in history. The faster you can demystify it and explore its uses, the quicker you can apply it to grow your business or advance your career.
A couple key definitions
The broad category that ChatGPT and its fellows fall under is generative artificial intelligence or generative AI. “Generative AI is a technology that can create content, including text, images, audio or video when prompted by a user,” according to the U.S. Government Accountability Office. It creates (generates) those responses using algorithms that are typically trained on open-source information, such as text and images from the internet. An algorithm in this sense is an exact list of step-by-step instructions in hardware or software routines.
If you’ve played around with these technologies, they can feel like a magic box generating cool content in seconds. What they’re actually doing is looking at patterns and relationships across massive amounts of data and then creating new content similar to, but not identical, to the content it’s been trained on.
Karim Lakhani, Ph.D., a generative AI researcher, Harvard Business School professor and co-author of Competing in the Age of AI, explained it this way on the HBR IdeaCast podcast from Harvard Business Review. He said the “AI plumbing” has been around for a long time through machine learning algorithms that can make strong predictions and recognize patterns. Google, Netflix and Amazon are all based on these technologies. With so much more infrastructure in place today, we’ve moved to another level. “Instead of just doing a prediction about the next sales thing or the next HR person you want to hire, let’s make predictions based on all of human text, all of human graphics that are out there, and get these systems to train themselves such that they can predict the next word.” Or the next pixel in a graphic image.
But the technology can also spread misinformation called hallucinations. Generative AI can even be used to intentionally create false or misleading information or phishing emails. Sometimes, it makes up source citations out of thin air.
Using it at work
While early predictions about AI bots replacing humans make alluring headlines, most experts suggest the technology will function more like a reliable partner to enhance how work is done, allowing humans to perform more sophisticated activities.
Lakhani doesn’t see machines replacing humans, but rather humans with machines replacing humans without machines. “Now you have this copilot that knows everything, but remember when you are the captain and you have a copilot, it’s your responsibility to keep the copilot in check,” and to know both its superpowers and its limitations, he advises.
Getting down to the nitty gritty in What every CEO should know about generative AI, McKinsey & Company report, “Generative AI can be used to automate, augment and accelerate work.” They point to several key functions generative AI can perform in organizations, such as classifying, editing, summarizing, answering questions and drafting new content. “Each of these actions has the potential to create value by changing how work gets done at the activity level across business functions and workflows,” according to McKinsey.
Lakhani supplies even more specific examples. An accountant might use generative AI to explain the data in a massive spreadsheet. A legal assistant might query it about applicable law for a certain case. He used it himself to clean up an article he had written for a professional journal. A colleague of his is using it to automate polite emails to decline speaking engagements. You might record a zoom call, enter the transcript from the call and request a summary. In a factory setting, it could help detect errors and automate processes.
Generative AI can be used to detect fraudulent transactions, categorize audio files from customer calls based on satisfaction levels, generate content and correct grammar, create content in multiple languages, assemble a highlight video from hours of footage, write and debug code, develop graphics in a desired style, craft a speech on a specific topic, write a song or compose a poem.
Still, remember you’re in charge. Generative AI can hallucinate answers that are plausible but untrue. And its content can be a bit vanilla. But it can also provide an excellent starting point for thinking humans.
Implications for leaders
It’s fun to ponder these new applications, but experts emphasize that leaders must exercise broad strategic thinking to ensure this shiny new thing is used in responsible ways that support organization goals.
If you are a tech-oriented company or in a tech-focused role, you may need to push to be an early adopter. Others may reasonably take more of a wait-and-see approach as you actively learn more about it.
McKinsey points to a range of legal and ethical concerns: bias built into algorithms and training data, intellectual property rights, privacy, inaccurate and intentionally false information, and the need for humans to maintain oversight.
Leaders must develop guardrails around where generative AI tools can work most effectively and prioritize its uses in their settings. They will also need to train the workforce on how to optimize prompts, understand AI limitations, provide ongoing oversight and integrate AI into critical workflows. “Fostering a culture of self-driven research and experimentation can also encourage employees to innovate processes and products that effectively incorporate these tools,” McKinsey advises.
The future may have gotten here faster than we expected — but regardless of your field, the time is now to explore generative AI with both prudence and vigor.