What Is Artificial Intelligence Really and When Will It Matter for Events?
Skift Take
Artificial intelligence (AI) and machine learning technology have come a long way in recent years, and they have a wide range of applications. However, there are some common misconceptions about what AI actually is, having often become synonymous with automation or, more generally, thoughtful computer programming.
In the event industry in particular, more and more event tech offerings now market AI features. For example, 46 percent of participating apps in EventMB’s Event App Bible 2021 report reported that they offer AI-powered attendee matchmaking, which is currently one of the most common AI-related claims in event platforms.
However, many of these features are powered by sophisticated algorithms that may not necessarily be AI in the strict sense. So, what does AI really mean, and when can it be used in events?
What Is Artificial Intelligence?
AI is a relatively broad term that is generally defined as the ability of a computer or machine to simulate human intelligence or behavior. Machine learning, which is a subset of AI, is considered a more precise term to describe one of its main current uses: to evaluate data and learn to make decisions for itself over time.
This is distinct from non-AI algorithms that — while they can also be quite complex — simply execute specific instructions as defined from the outset in the initial programming (barring any bugs).
According to Simon Clayon, Chief Ideas Officer at RefTech, one of the hallmarks of AI is precisely its ability to learn and get better at its task over time as it receives feedback. However, this criteria is not always met in features that are marketed as AI.
An example of this would be a matchmaking algorithm that matches attendees and provides recommendations based on predetermined criteria, but doesn’t improve on its own using data it collects as the event goes on. As Clayon explains, “this wouldn’t be AI because it wouldn’t be using a neural network. It wouldn’t have learned how to do that itself; it would have been programmed.”
However, since many eventprofs may not be making this distinction between ‘real’ AI and sophisticated automation, they are not necessarily being misled. If most people have come to associate AI with complex algorithms in general — for better or worse — then they will likely be getting what they expect.
Jim Sharpe, CEO of Aventri, notes that definitions of AI range from machines simulating human behavior to actually being able to mimic the problem solving functions of the human mind, which is a much higher bar to clear. The working definition that he uses is as follows: “Are we empowering people to have better experiences based on the capabilities of machines to inform decision making?” And when it comes to the event industry, his answer is yes.
Can AI Be Used For Events?
Clayton argues that true AI, defined as machine learning, has many use cases across industries, but its applications for events are currently very limited. This is due in part to the limited data and the lack of feedback loops in most event scenarios.
Sharpe, on the other hand, explains that if machine learning is the criterion for AI, there are already existing event tech features that are leveraging it. Aventri’s networking engine is one example as it includes a feedback loop — if attendees accept the suggestions that the tool makes, the system learns and gets more sophisticated.
He adds that while the data being used to improve the system may still be relatively limited, it isn’t necessarily limited to a single event, so the system can collect and learn from data from multiple events being held on the platform. This process may be improved by event planners standardizing the parameters they use to define their audience and the criteria used in the recommendation engines, but there are likely other incentives to do that as well (such as increasing users’ familiarity with the process and the labels being applied).
In terms of the future evolution of AI for events, Sharpe sees multiple applications. Networking recommendations are one aspect of events that AI will continue to improve, and not just for virtual to virtual connections, “but also for live to virtual, or live to live. Being able to tell an attendee that someone at the coffee bar 500 yards away is someone they should meet would be a powerful tool.”
Another area that will benefit from AI, according to Sharpe, is overall event intelligence. Platforms will be able to collect a lot of data on what attendees are doing at events, including which sponsors they’re interacting with and for how long, and suggest how planners or CMOs can improve future events.
Finally, Clayton explains that one of the main uses of AI that is currently being deployed for events is facial recognition, which Sharpe also believes will continue to be leveraged and improved to ensure the safety and security of events.
Is the Term “Artificial Intelligence” Overused?
Both Clayton and Sharpe agree that the term AI has been overused across all industries, not just the event industry. As a result, it’s not always used to accurately describe what a product or feature can actually do. But whether that constitutes a problem for marketing purposes really depends on whether those making purchasing decisions are being tricked.
Overall, it’s worth it for eventprofs to understand the true meaning of AI because it’s not going anywhere. There are several AI-powered features in development to keep an eye on, and having a bit of deeper knowledge about the technology will make it easier to have intelligent conversations with providers about their capabilities.
That said, while Clayton cautions against false marketing, he notes that a technically inaccurate description does not necessarily mean that a feature isn’t valuable or useful. The most important thing is that planners understand what the feature does, and from there, they can evaluate it on its own merits and determine whether it fits their needs.
“What I would recommend event planners do is think about the core features that they need at their event and test those features through discussions with the providers, demos, etc. At the end of the day, whether or not it’s technically AI is not as relevant as whether or not it’s solving the problem and meeting customer needs.”
– Jim Sharpe, CEO, Aventri
IN CONCLUSION
We will definitely be hearing more about AI in the near future, so it is worthwhile to understand exactly what it refers to. While it has become an overused marketing buzzword, planners should keep in mind that while not all “AI-powered” features may necessarily be leveraging AI or machine learning, they may still be useful and shouldn’t be discounted purely due to the potential inaccuracy of the nomenclature.