Exploring AI & Data Analytics in Modern Events

Panos Moutafis is the CEO of Zenus, a company that uses a cutting-edge mix of AI and anonymous image capture to offer a contactless look at attendee analytics. It can decipher sentiment and deliver data analysis, be it at an event or a retail location. At Event Tech Live Moutafis moderated a panel with Dax Callner from Smyle (a creative agency that uses Zenus’s tool for delivering event metrics), Emily Miller from The Founders Forum Group, and Mary Ann Pierce from MAP Digital, to talk about the increasingly critical role of data in events. Advanced techniques using AI, facial recognition (done ethically), and data analysis offer the events business a better understanding of the participant experience, so that planners can offer improvements. Click on the photo below and you can watch some highlights of the conversation. Incidentally, look at how they elongate the life of the event by creating these easy-to-retrieve snippets at the same time that they are using the content to capture new leads who may have not been able to be physically present. You’ll need to give them your email address to unlock the video. 

If your company is thinking of implementing facial recognition systems, just be sure that it’s for the right reasons. When it comes to cautionary tales, this story reported by the NY Post takes the prize. Over Thanksgiving weekend, Kelly Conlon, a personal injury lawyer from Bergen County, N.J., was chaperoning her 9-year-old daughter’s Girl Scout troop on a trip to Manhattan to see the “Christmas Spectacular” at Radio City Music Hall. Security guards pulled Ms. Conlon aside, telling her that she could not enter. It turns out that the lawyer's face profile was recognized on the Music Hall’s naughty or nice list. The owners of Radio City Music Hall and other venues such as Madison Square Garden are the very powerful Dolan family, and Conlon’s law firm is involved in a lawsuit against them. (I sense another lawsuit pending and it’s a glimpse of the darker side of facial recognition at event venues.)

No one likes to talk much about it, but striving for diversity, equity and inclusion at events (and life in general) is really hard work. That goes double if your event is in the tech-ier segments of an industry. For years, my company managed speakers at high-tech events. Looking for a woman or person of color to participate on a panel was often a challenge; worthwhile, but a challenge. And not coincidentally, if you did manage to find female speakers, there was a much higher likelihood that you’d lose them as the event date grew near. Like it or not, women still shoulder most of the family emergency situations that invariably arise. 

Many companies have DEI initiatives, and speakers at conference sessions are starting to reflect the upside of those efforts, but it’s still a hard slog. The tech industry, for a variety of reasons, is still clearly dominated by men. Statistics show that year after year, women are underpaid and underrepresented. Women make up only 25% of computer-science-related jobs, according to the Pew Research Center, and are even less represented in engineering, making up just 14% of the workforce. At C-suite levels women occupy less than 20% of the top tech jobs in the U.S. A survey by Built In found that one in four of the companies it surveyed said their teams were more than 70 percent White. 73% of respondents said there were no Black leaders on their executive teams. And 39% of women and BIPOC employees surveyed said that they didn’t feel that their voices and perspectives were included in the decision-making process at their jobs. This year’s Lean In report indicates that women have less opportunity to climb the corporate ladder (referred to as “the broken rung”), are increasingly leaving the workplace, and want a more flexible work culture.

That doesn’t mean gender and racial equity at events is impossible, or that it’s not a battle worth fighting. When your panel’s speakers reflect a variety of backgrounds – gender, ethnicity and differently-abled – you’re bound to get a deeper, more realistic look at your subject matter. And your audiences will appreciate the diversity.

Here are some tips that can help your programming. 

• Look at the bios and headshots of your speakers and make sure they reflect the diversity you want to see.
• Get in the habit of using appropriate language. The American Psychological Association has developed helpful inclusion language guidelines.
• And remember that inclusion is not only a matter of gender and race. Details such as closed captioning, wheelchair access, and other tools for differently-abled people are a must.

If you want to geek out on the progress being made to ensure that everyone has a place at the table, then read the article recapping some of Google’s accessibility work. This fall Google and Marriott Bonvoy announced Project Neu (Neu is short for neurodivergent) to address the needs of event attendees with different needs, including things like “quiet places” with less stimulation and “fidget spinners” to help with distraction. 

The tent continues to grow, but it takes commitment. Miguel Neves, the Editor-in-Chief of Skift Meetings, details his take on how meetings have evolved both in sustainability and in DEI. “An evolved industry – one that thinks beyond itself – is better prepared to play a critical supporting role in dealing with today's global challenges,” he writes.

Generative AI Goes to the Movies
D-ID lets you create AI-generated videos from a single image.

D-ID works well for creating training videos, explainers, and welcome videos. You select an image as your talent, add text, and choose a voice. D-ID will turn that into a movie.

MUMs the Word
Trust us, Google is not sitting by twiddling its thumbs while AI generators are making searching the web more conversational every day. MUM, which stands for Multitask Unified Model, is Google’s project model for working to make searching the Internet more like having a conversation. Popular Science explains how your next search could be phrased like this: “I’ve hiked Mt. Adams and now want to hike Mt. Fuji next fall, what should I do differently to prepare?” That’s a lot to unpack. Hiking tall mountains, preparation and training are all part of the query. It requires training models that understand not just words, but the relationships between very specific word usage. We’re not there yet. The question is will Google get there first?

Become the Sherlock Holmes of AI
If you flunked the test at the top of this newsletter then you need some instruction in how to tell AI from human speak. It’s not an exact science, but AI leaves a trail of clues. MIT Technology Review one good clue is to look for typos because humans are more likely to make them than AI-generated text. (However, AI’s can be asked to add typos and other mistakes, as was done in the NYTimes 4th grade challenge.) Another clue is to look for too many “the,” “it,” or “is” instead of other words. That’s because the models predict the next words in the sentence. Surely, someone is working on an AI skill to detect AI-generated content. That will be a challenge, because AI continues to improve its language skills, but I'm not so sure that’s the case for humans.

Following the Money
While tech platforms are finding their footing in the post-pandemic normal, there are signs that large event companies are seeking to round out their offerings with digital tools. Events.com received a $100M capital investment from Global Emerging Markets (GEM) to expand its services, ostensibly through acquisitions of tools.

Looking Back to Look Forward
From looking at the changing makeup of event attendees (including things like planning for higher “no-shows” at the event) to the more last-minute nature of life today, to the adoption of digital-first tools, the events industry is re-emerging, new learnings in tow. Bizbash does a nice job of capturing the sentiments in a look back at what 2022 taught us.