La Fleur's Lottery World
    Twitter LinkedIn RSS
    Friday, June 9
    LOG IN: Purchases & Subscriptions
    La Fleur's Lottery World
    0 Shopping Cart
    • Home
    • Conferences
    • Magazine
    • Store
    • My Purchases
    • White Pages
    • Cart
    La Fleur's Lottery World
    0 Shopping Cart
    You are at:Home»Magazine»Magazine Feature»Are Lottery Executives Missing Out on the Benefits of ChatGPT?
    Magazine Feature

    Are Lottery Executives Missing Out on the Benefits of ChatGPT?

    April 13, 202310 Mins Read
    Facebook Twitter LinkedIn Email

    As artificial intelligence (AI) continues to advance, industries across the globe are feeling the impact of its transformative power. The lottery industry, steeped in tradition and often resistant to change, won’t be an exception, yet, to date, the industry has not embraced these advancements. A recent survey targeting more than 1,000 lottery executives explored the impact of AI services like ChatGPT, an AI language model developed by OpenAI, on their daily operations. Interestingly, only three executives found it beneficial in their work life. This article seeks to shed light on the factors contributing to this apparent disinterest, analyzing the potential gaps between the capabilities of ChatGPT and the needs of the lottery industry.”

    “Often resistant to change…” Ouch.

    This introduction was written by ChatGPT, and despite its backhanded insult, it is pretty good; especially considering it only needed a short prompt, took 15 seconds, and generated three other variations. Oh, and it was free. For those unaware of what ChatGPT is, it is a large language model, which has been trained through reading billions of words ranging from books, websites, and social media. Users interact with it through an instant messenger service, typing questions or commands. It generates responses quickly and can perform almost any written task.

    It isn’t the only AI service with the potential to disrupt the world. All the large tech companies are launching their rival versions of chatbots, like Meta’s OPT and Google’s BARD. The market is already saturated with text-to-image services, which generate an image after a user inputs a short prompt. Google researchers published a paper in January on a service, dubbed MusicLM, which can generate high-fidelity music. Microsoft showcased their new and improved Teams, which will be able to summarize meetings and answer questions; all its productivity software, including Word and Excel, will get the same treatment.

    With such fervor around the topic, La Fleur’s Magazine recently conducted a survey, asking the lottery industry if and how it is using these services. Usually, these surveys get anywhere between 300-500 responses; this survey got 27, which suggests adoption is very slow. However, out of all the respondents, only three confirmed that they have utilized these services in a work-related capacity. This article will examine the manner in which these individuals have reported using ChatGPT, as well as other use cases.

    Social Media

    The first respondent uses Canva’s text-to-image service to help create social media content, which generates an image based on a textual description. In other words, if the query was “‘A sea otter with a pearl earring’ by Johannes Vermeer,” the service would create multiple images of a sea otter with a pearl earring.

    “The results are imperfect, but that makes for some good social interaction. And sometimes the results can be surprisingly beautiful. I always mention that they were generated by AI, so people are aware of it,” the respondent said.

    Social media is one area where AI excels. Canva and Dalle’s resolution for images is still insufficient for print, but perfect for online media. For those not artistically inclined, they can describe to the computer what they want, altering their query until they find a suitable option.

    ChatGPT can also help to write posts. For instance, here are the first two results for funny Twitter posts when the Powerball is $20 million, emojis and hashtags included:

     

     

     

     

     

    It would take a brave social media coordinator to publish the first one on a Lottery’s official account, but the suggestions are undeniably humorous. For those risk averse, it can alter its output depending on how many parameters the user gives it. For instance, if you don’t want food-related tweets, simply tell it.

    Ad Agency in a Bot?

    The second respondent’s comment was a little difficult to decipher: “In 30 seconds it generates a script that would take someone days to learn how to write.”

    “Script” could technically mean an advertising script. Even if this is a forced interpretation, here is what ChatGPT created for a 30-second commercial pitch:

    “Take a Shot at the $20 Million Jackpot – It’s Cheaper Than Therapy!” – This commercial could highlight the stress and anxiety of everyday life and suggest that buying a Powerball ticket is a more affordable alternative to traditional therapy. It could feature shots of people dealing with stressful situations, such as a traffic jam or a difficult boss, and then buying a Powerball ticket and daydreaming about a private island the next time they’re in that same situation. The commercial could end with a humorous tagline like ‘Win big or at least feel better trying!’”

    A lottery promoting its product as a therapy substitute in a commercial is inappropriate. However, the concept is funny and was generated in under 15 seconds. It’s just as simple to request five or 10 suggestions instead of just one. Even if none of them are perfect, one of them could be the starting point for a fantastic idea.

    Dalle’s interpretation of ChatGPT’s commercial pitch. Admittedly, it is two images stitched together as it had difficulty creating an angry boss and happy employee. This is a example of a bad text-to-image graphic, but it fits the theme of the article. Why does the employee have one eye and the boss have curled fingers? I don’t know, but I am not going to be the one to question our new robot overlords.
    Excel and Coding

    However, the respondent probably meant coding, which ChatGPT does as well. It can code in Python, JavaScript, Java, C++, Ruby, PHP, Swift and more. It can identify syntax errors, code formatting issues, and potential logic errors in the code. It can suggest improvements to the code’s readability, efficiency, and maintainability. It cannot do a full code review, as that takes a deeper understanding of what the program is trying to accomplish, but it is still impressive. For lotteries trying to create their own apps, it could help take months off development time.

    Now this application isn’t useful to everyone. However, while most people do not have to write code, many people must write Excel formulas and macros, which it can also produce.

    Training

    The third respondent, Terri Rose, Deputy Executive Director, Brand Management & Communications, North Carolina Education Lottery (NCEL), said: “I have used it to develop training, search the internet more accurately and to help with Excel formulas.”

    NCEL recently hired a new communications manager who is new to the lottery world. “The lottery industry is not always the easiest thing to explain to people,” Rose said.

    Their training manuals are outdated and need to be updated for new commission members and staff, but the task is time-consuming. To simplify the process, Rose began using ChatGPT to answer common questions, such as the types of games the lottery offers and the meaning of various acronyms.

    Some of it was wrong, for instance, such as identifying SCEL as an acronym for Scientific Games. But in other cases, it was fantastic. “We asked for information on Pick 3 in North Carolina. We really started getting great information; we could use this word for word, cut and paste it right into a training document,” Rose said. “It still needs some human intervention. It can’t do everything for you.”

    Other Suggestions

    It will likely also have applications the industry hasn’t tried yet. For instance, here is a suggestion for a new product:

    “Chain Reaction”—In this mechanic, the player scratches off a covering to reveal a set of symbols or numbers. If the player reveals a winning combination, they can continue to play on a mobile device. If they win again, the symbols or numbers will “explode” and reveal new symbols or numbers. The new symbols or numbers that are revealed could create additional winning combinations, leading to even more prizes. This creates a chain reaction of exploding symbols or numbers and potential wins. To make this mechanic even more exciting, the ticket could feature bonus symbols or numbers that offer higher payout values or bonus prizes when matched. (Admittedly, I had to sift through 15 or so suggestions before I found an original concept—at least to my knowledge).

    The capabilities of ChatGPT extend beyond the suggestions above. For individuals, it can check for grammatical errors, write emails, edit resumes, and even create press releases or winner stories. There is potential for the technology to be utilized in other areas that have not yet been explored. No idea should be disregarded. For example, even something as time-consuming and complex as RFPs could be aided by ChatGPT in the future.

    AI-ght, I’m Scared

    The current iteration of ChatGPT and other AI services won’t be stealing anyone’s job. They still make way too many mistakes. However, it’s the trajectory, and the speed at which it is improving, that should cause some trepidation. The most recent iteration of GPT, the technology behind the chatbot, which was released on March 14, is a huge step forward.

    For example, GPT-4 passed a simulated Uniform Bar Exam (Bar) with a score in the top 10% of test takers; in contrast, GPT-3.5’s score was in the bottom 10%. The Bar consists of essays, multiple choice questions, and performance tests, which include writing a legal memorandum and drafting a settlement. It is infamous for how difficult it is, causing many an aspiring lawyer to study in insolation for months to pass it.

    It took GPT three months to go from flunking the Bar exam to acing it.

    History has proven time and time again humans do not have a good record against machines, especially over a long period of time. In 1996, the then World Champion, Garry Kasparov, played a chess match against a computer (called Deep Blue), winning 4-2. In 1997, they played a rematch; Deep Blue won 3-2. Today, chess computers have so far outpaced their human counterparts that it is impossible to beat one. Chess grandmasters, like Hans Niemann, must undergo thorough metal detector tests before tournaments to prove they aren’t hiding a supercomputer on… or in their body. Over a long enough timeline, there is no single profession or job that will be insulated from AI’s relentless progress. Whether that is a good thing or not remains to be seen.

    On one hand, at the risk of being quixotic, AI could usher in a utopia. On the other hand … well, it wasn’t good for humanity when SkyNET became self-aware. Regardless of the outcome, this technology has the potential to become the next Industrial or Digital Revolution. While organizations may need to be mindful of proprietary or confidentiality concerns, individuals should still experiment with it when safe to do so, whether for brainstorming or overcoming writer’s block. As this technology continues to develop, it is likely to become an essential tool for all employees. By familiarizing oneself with AI now, individuals will be better equipped to maximize its potential in the future.

    Or, as ChatGPT said: “The true power of AI lies not in its ability to replicate human intelligence, but in its capacity to augment it. As we work alongside intelligent machines, we can unlock new levels of creativity, productivity, and innovation.”

    Like this content? Subscribe for more interesting lottery news and data!

    * indicates required
    Share. Facebook Twitter LinkedIn Email
    Previous ArticleThe Roundup: On ChatGPT, AI, & the Workforce
    Next Article MORE GREAT GEORGIA LOTTERY SCRATCHERS COMING TO PLAYERS FROM GEORGIA-BASED SCIENTIFIC GAMES

    Related Posts

    Paper to Pixels: How the Oklahoma Lottery Said Goodbye to Paper Playslips

    April 20, 2023

    How Texas Lottery Harnessed iHeart’s Power to Promote Cash Blitz

    April 20, 2023

    The Roundup: On ChatGPT, AI, & the Workforce

    April 13, 2023

    Comments are closed.

    Subscribe

    La Fleur's will respectfully update you twice a month on lottery industry news, data reports, and conferences.

    Reflections on the First Decade of iLottery in the US: Moti Malul, NeoGames S.A.

    Exploring the Ultimate iLottery Toolbox with Lianne Paturel, Pall Palsson & Shannon DeHaven, Pollard Banknote

    IGT Platinum Sponsor Presentation: The Cloud Has Arrived- Jesse Saccoccio, IGT

    About
    About

    Your source for lottery industry news, data, and conferences.

    We're social, connect with us:

    Twitter RSS LinkedIn
    Contact Information

    5157 Ijamsville Road

    Ijamsville, MD 21754

    301-610-6070

    byron@lafleurs.com

    Popular Posts

    Reflections on the First Decade of iLottery in the US: Moti Malul, NeoGames S.A.

    June 8, 2023

    Exploring the Ultimate iLottery Toolbox with Lianne Paturel, Pall Palsson & Shannon DeHaven, Pollard Banknote

    June 8, 2023

    IGT Platinum Sponsor Presentation: The Cloud Has Arrived- Jesse Saccoccio, IGT

    June 8, 2023
    Copyright © 2023 TLF Publications.
    • About Us
    • Returns Policy
    • Privacy Policy
    • Terms and Conditions

    Type above and press Enter to search. Press Esc to cancel.