Conversations About AI

Conference producers organize conversations for a living, and one of the joys of the job is watching an idea whose time has come explode into the public consciousness after years of quietly doing important work out of the limelight.

OpenAI launched ChatGPT on November 30, 2022. Within two months it already had over 100 million active users, making it the fastest growing consumer application ever to date. As an interactive tool and toy, it has captured people’s imaginations and made Artificial Intelligence a topic of everyday conversation from boardroom tables to breakfast tables.

As unlikely as it would have sounded two years ago, just about everyone nowadays has at least a passing understanding of how Large Language Models work, and most people have asked with varying degrees of trepidation or excitement, “But what does AI mean for me?”

It has been my pleasure to talk about Artificial Intelligence, Machine Learning, and other flavors and approaches to data analytics since long before ChatGPT took the world by storm, and now that it has, I struggle to come up with an interview I have done across all the industries and areas of responsibility we build events around that has not at least touched on AI in one way or another.

AI has to be at the top of any list of universal trends happening in the world of business right now, and so it is high time we showcase some of our content and offer a few thoughts on where different types of professionals, executives, and senior leaders see themselves at the moment, and where they believe these powerful tools are going to take them in the future.

For the sake of being entertaining, I will break this down into three sections: It’s Not as Big a Deal as People Think; It’s as Big a Deal as People Think, and It’s an Even Bigger Deal Than People Think. My hope is that by having these divisions, readers will both find validation for their own level of enthusiasm while also getting a sense of where other people stand and why. With that said, let’s get into it, shall we?

It’s Not as Big a Deal as People Think

The last ‘big thing’ that touched every company and organization of every shape and size around the world was COVID-19. In the aftermath of the pandemic —once people had a chance to catch their breath and reflect on the wild rollercoaster ride we had all endured alone together— the thing people said to me over and over was, “It didn’t so much change things as accelerate the trends that were already quietly happening in the background.”

As one example by way of illustration, the idea of working from home was not invented during the pandemic. COVID-19 just made the companies that were already doing it ahead of the curve, and the companies that had up to that point refused to entertain the idea found themselves compelled to overcome their reluctance to meet the needs of the moment.

What is happening right now with AI also feels like an idea whose time has come for widespread adoption rather than a sudden revelation, and I am not the only person to say so. We already have Machine Learning and Advanced Data Analytics and, yes, even things we were calling AI before OpenAI launched ChatGPT. When I speak to supply chain executives and manufacturing executives and finance executives, all of them can point to a decade’s worth of digital transformation initiatives that are all about finding actionable data quickly from a firehouse of information streaming in faster than any human being could ever hope to process.

In that spirit, successful and customizable large language model AI with Generative AI interfaces are an exciting next step that comes with built-in enthusiasm from parts of their organization that in the past needed a lot of persuading that technological change was good, but it is not fundamentally new.

For the people in the ‘not that big a deal’ camp, one of the biggest sticking points is we have not actually had a breakthrough in the creation of artificial intelligence as the popular imagination until recently defined it. No original thought goes into the output we receive from an LLM-powered AI. Instead, with a large enough pool of data, software now exists that is designed to give an answer that a human user will deem acceptable. That is the criteria for success, and that is the challenge the code is designed to solve. When it succeeds, the success can sometimes be downright spooky. When it fails, it’s just a confused piece of software that did not understand what it was being asked to do, and why would it? It is only ever going to be as effective as the prompt it is trying to satisfy and the data it can comb through in the search for a satisfying reply. Still, even that is an impressive next step in the world of interactive search and automated copywriting, both of which are incredibly powerful and valuable tools.

If this next step has appropriated the branding of AI as its own without actually being Artificial Intelligence, the only people who are frustrated by that are those interested in other forms of AI still under development and now pushed further out of the limelight. For the vast majority, that splitting of hairs and wringing of hands over inaccurate labels is not enough to detract from enhanced capabilities that fire the imagination and provoke real reflection on what we should do with our time if large parts of our regular duties could be automated.

Let’s talk about the people who hold that viewpoint now.

It’s as Big a Deal as People Think

Having mentioned ChatGPT a few times now, let’s put a little distance between that specific tool and what we are going to talk about when referencing conversations we have had with industry thought leaders. ChatGPT is an incredible proof of concept that the public can engage with freely as a technology demonstrator, and while it does have commercial applications, many of the weaknesses of ChatGPT —that it can hallucinate; that its original pool of reference data is dated, problematic, and potentially open to legal dispute; that moving forward new iterations will be drawing information from data pools tainted by earlier AI-generated content— will not apply when a company uses the framework as a template to creates its own Large Language Model AI, hooks it up to a Generative AI user interface, and lets its employees engage with its data in a way that only a select few have ever had the training or access to do before.

Our pharmaceutical, biopharmaceutical, and medical device manufacturing executives and the service and solution providers who work with them are seeing a way to rationalize and engage with the data they have already been collecting for compliance and QA/QC purposes to make a step-change in their productivity, profitability, and product development. What is the future of medicine going to look like if an LLM is chewing through confidential proprietary internal data, and people can ask questions that connect dots automatically and propose next steps in a conversational list of suggested instructions?

For our supply chain, manufacturing, and food safety executives, AI is often about democratizing internal data and getting it into the hands of frontline workers. While senior leaders have had access to enormous pools of data and powerful analytical tools for many years now, how often has the output of those tools been readily available to someone on a shop floor who is troubleshooting a problem or trying to improve a process in real time? Generative AI means that person can ask simple questions through a mobile device and be given actionable answers in a way that before would need to go up and down the chain of command to and from the person with access and training to make the most out of the tools and data. What does that do to productivity and job satisfaction?

For our Finance and Sustainability senior leaders, data has been so much of their job for so long, they look at LLMs not so much as a major change in their processes, but as generating a groundswell of support and understanding for what they have been trying to do with less well-known and less-powerful tools for years. They have always had the metrics, and now they can put it into a Large Language Model and ask it impactful questions that will then help them communicate what their efforts offer the larger organization in a way everyone can understand. As powerful as the data analytics aspect is, how much more powerful will AI be as a communication tool helping them tell the story of their data to others?

For the senior leaders of Human Resources, AI can help address some of the biggest challenges they face on a day-to-day basis. Talent Attraction, Talent Development, and Talent Retention all see positive impacts from AI-driven automation of repetitive grunt work, freeing HR professionals to spend more of their time on the actual people part of the job.

Imagine training an LLM with what makes a good fit for a given role, and letting AI sort through job applicants, flagging only the best matches for human review? Imagine a Generative AI that could talk to current workers on employee engagement issues in a way that looks nothing like the surveys of old. What does that do to response rates and creating genuine insights into how people feel, what they want, and how their company can best support them personally and professionally? Once AI is a new standard rather than a novelty, how much higher will retention rates be for companies that are recruiting, listening, engaging, and developing people with the guidance of advanced data analytics through a conversational interface?

It will be what divides being an Employer of Choice and being seen as a backwards-looking company. It will be the difference maker.

It’s an Even Bigger Deal Than People Think

There are three big arguments that promoters of AI make in private conversation that I have not seen emphasized in blog posts like this one yet.

First, possibly the most important single thing ChatGPT has done so far in the world of business is getting the people who are the least interested in Digital Transformation excited about putting a new tool onto the very top of the toolbox. From older frontline workers who have always had a natural conservatism about automation to C-Suite executives trained to focus on the bottom line in quarterly increments, it has always been a struggle for advocates of digitization to inspire the passion and secure the resources they need to turn pilot projects into the company-wide processes that will deliver the kind of ROI that has long been promised but few have yet realized.

ChatGPT has changed that. Unlike some of the earlier iterations of digitization, people know what ChatGPT is, understand it enough to be enthusiastic about it, and there is no fear factor that it is too expensive or too complicated. If anything, the zeitgeist is maybe misunderstanding the possibilities in the other direction where they expect free and easy to be deliverable at the snap of the fingers. Wherever in the middle the truth of the matter may lie, the older frontline worker is being offered a tool they want to use. They ask it questions, and they get answers that make a real difference to their work immediately. They have not been replaced. They have been freed and empowered to do more of the tasks that actually matters to them. One demonstration of how it works makes all the normal pushback go away. Generative AI is the kind of ‘it sells itself’ improvement change management experts dream about.

As for a company’s C-Suite? From every conversation I have had over the last year, there is near-universal enthusiasm for this suite of technologies and capabilities. Everyone has heard of it. No one wants to be left behind. The digital transformation advocates within their companies are now being bombarded with questions about how to incorporate these new tools into existing systems, or even whether the existing infrastructure should be wholly replaced by LLMs and Generative AI. For the people who have been working in this space for years, it is a very exciting time to try new things with full confidence that the support and enthusiasm level are sky high with no danger of coming back down to earth any time soon.

Second, technological solutions only really find success when they answer the needs of an organization. Popular and cool are not nearly as impactful as solving something that desperately needs fixing. There is an enormous issue happening right now that AI is almost ideally suited to address: The workforce of the later half of the 2020s and the early 2030s will be fundamentally different than the workforce of yesteryear.

By most estimates, 10,000 Baby Boomers are retiring each and every day. Meanwhile, Generation X and the Millennials came out the far side of COVID-19 lockdowns with a different outlook on work-life balance and what they want from their employers than they had held before. At the same time, Generation Z is entering the workforce in increasing numbers with a completely different philosophy for how they want to navigate their careers from the generations that came before. Institutional Knowledge is going to be increasingly rare and valuable from now on. People are going to move from job to job despite the best efforts of Talent Retention programs. Boring repetitive tasks are going to cost companies not just time and money and the lost potential of more productive work, but also the talent who understandably would prefer to be doing something else with their time. Meanwhile, even when new people do join a team, there are going to be fewer and fewer mentors and senior staff to bring them up to speed and make them excellent at what they do.

An AI working from company data that can interact and engage with people of all ages and experience levels to help them do their jobs better while also taking automatable tasks off their plates to let them spend most of their time doing what attracted them to the job in the first place is not just going to be a nice-to-have capability in the future. It will increasingly be the difference between winning and losing the war for talent. The stakes are that high, and not enough people are talking about how Artificial Intelligence is the solution to the hard truth that more and more companies are going to be relying on fewer and fewer people who actually have long-term real-world experience in their current positions; LLM AI with a Generative AI interface is going to provide the continuity and connective tissue that will keep teams running and processes operating as increased turnover accelerates and is cemented in as yet another facet of the solidifying New Normal of Work.

Third and finally, not enough people are pointing out ChatGPT is just the tipping point of something that is only going to accelerate from here on out. While the tech sector has a long-standing tradition of regular layoffs, 2023’s and 2024’s layoffs hit traditional search and non-LLM AI projects particularly hard as every major tech company is now either doubling down on OpenAI’s approach or is desperate to catch up with OpenAI’s breakthrough in both technology and marketshare.

We are only at the very start of what LLM-powered Generative AI is going to offer the working world. For anyone who is still thinking it’s not that big a deal, or it’s only just as big a deal as people say, you haven’t seen anything yet. Customizable LLMs chewing through company-specific data is going to fulfill every promise Digital Transformation evangelists have ever made, and then make some new ones on top of that.

Get ready for a wild ride, because it’s coming.

Geoff Micks
Head of Content & Research
Executive Platforms

Geoff joined the industry events business as a conference producer in 2010 after four years working in print media. He has researched, planned, organized, run, and contributed to more than a hundred events across North America and Europe for senior leaders, with special emphasis on the energy, mining, manufacturing, maintenance, supply chain, human resources, pharmaceutical, food and beverage, finance, and sustainability sectors. As part of his role as Head of Content & Research, Geoff hosts Executive Platforms’ bluEPrint Podcast series as well as a weekly blog focusing on issues relevant to Executive Platforms’ network of business leaders.

Geoff is the author of five works of historical fiction: Inca, Zulu, Beginning, Middle, and End. The New York Times and National Public Radio have interviewed him about his writing, and he wrote and narrated an animated short for Vice Media that appeared on HBO. He has a BA Honours with High Distinction from the University of Toronto specializing in Journalism with a Double Minor in History and Classical Studies, as well as Diploma in Journalism from Centennial College.