INSIGHTS

Enter the Dragon: On the Potential of Artificial Intelligence in Community Management

By Patrick Cadenhead, CMCA AMS

(With attribution to the Game of Thrones television series, based on the A Song of Ice and Fire novels by George R. R. Martin.)

(Published in CAI–Orange County Chapter’s OC View, November/December 2024)

Much like Balerion the Black Dread cast his immense shadow over the Seven Kingdoms (his wingspan was so large he could engulf entire towns with his shadow), generative AI has soared into the realm of possibility, poised to unleash its transformative power upon community associations. Generative AI refers to artificial intelligence systems that can create new content, such as text, images, music, or code, by learning patterns from existing data, often mimicking human creativity. The concept of wielding such power may be both thrilling and daunting, depending on one’s perspective.

Imagine if board packets could assemble themselves, emails could instantly craft and send automatic intelligent responses, and phone calls could be managed by AI in real time. What if the agenda and management report were effortlessly generated? And what if meeting minutes were flawlessly produced, capturing every detailed motion with precision, with little or no manual effort? These ideas merely scratch the surface of possibility. Just as Aegon’s dragons bent entire kingdoms to his will, generative AI stands ready to reshape the domain of community management, offering those who can harness its power the ability to incinerate the tedious complexities of association administration with the devastating power of dragonsfire. (With attribution to the Game of Thrones television series based on A Song of Ice and Fire novels by George R. R. Martin.)

Generative AI has already demonstrated its ability to analyze PDF documents, efficiently identify their contents, and generate concise summaries. With minimal training and customization, tools like ChatGPT can take this a step further by creating detailed agendas based on the information extracted from multiple PDFs, streamlining the planning process for board meetings. These agendas can then be expanded into comprehensive management reports with just a few simple instructions, offering tailored insights and recommendations.

Moreover, financial data within PDF and other document formats can be seamlessly extracted, analyzed, and included as needed, or even transformed into visual representations like charts and pie graphs. With the potential for integration with platforms like Adobe Acrobat through API functionalities, generative AI could soon automate the entire process of assembling board packets, including organizing, formatting, and compiling all relevant documents. This technology stands on the verge of burning through administrative workflows, drastically reducing the time and effort required for board-meeting preparation.

Many of us have already experienced the power of ChatGPT to assist in crafting professional email responses. With AI models specifically trained on association-related data, a virtual assistant could be developed with intimate knowledge of an association’s governing documents, meeting minutes, and the latest community updates. To ensure accuracy and prevent confusion, each association could have its own tailored AI model, fine-tuned to understand its unique governance and operational landscape. By integrating OpenAI’s API with platforms like Power Automate, Microsoft Exchange, or Google Script, incoming emails could be automatically relayed to the correct AI model based on keywords such as names of streets and associations. The system could then generate intelligent, context-specific responses that are instantly sent via email reply, streamlining communication with homeowners, vendors, and board members.

Furthermore, leveraging technologies like Text-to-Speech (TTS) and Whisper, it’s easy to imagine AI not only handling written communication but also answering phone calls and holding real-time conversations with homeowners. With the ability to convert speech to text and vice versa, generative AI could handle complex telephone queries and provide instant, accurate responses.

The potential goes even further — AI video generation now makes it possible to create hyper-realistic avatars, whether based on real people or entirely virtual personas, with perfectly synced lip movements to live-generated text. By integrating these avatars with generative AI through API connections, we could soon see virtual managers attending Zoom meetings, engaging in real-time conversations with board members and homeowners. For those who recall the 1980s, this idea might evoke the iconic Max Headroom, the original “virtual” TV personality who captivated audiences with his digital glitchiness and charisma. Today’s AI avatars feel like the next evolution, seamlessly taking on interactions while the human manager sits back and watches with wry amusement. The future of management is here; but who is ready to “catch the wave”?

For those of us who’ve ever struggled to decipher our own notes from weeks before, frantically piecing together the minutes while prepping the next board packet, we already know the importance of the expeditious creation of the minutes and action-item list immediately after a board meeting. Now, with the power of generative AI, meeting minutes and action-item lists can be generated almost instantly after the meeting conclusion with minimal exertion. This groundbreaking capability ensures that best practices for swift minute creation are not only achievable but almost effortless, making the entire process more efficient than ever before.

Over the years, best practices for taking minutes have evolved from handwritten notes scribbled in the margins of the agenda within a hard-copy board packet to interspersing typed notes directly into management reports and processing the resulting document through ChatGPT. Looking ahead, a mobile app with API integration is envisioned, allowing board-meeting audio to be recorded on a device and instantly converted into meeting minutes and an action-item list with the push of a button.

But this process can already be achieved almost as efficiently when meeting via Zoom. While Zoom includes built-in AI functionality, such as the ability to provide a meeting summary (not yet tested), utilizing Zoom’s recording feature in conjunction with ChatGPT is the preferred approach. Pro-tip: At the beginning of your Zoom meeting, hit record. Enable captions and save the resulting transcript, which will create a text file that can be uploaded into ChatGPT. Your mileage may vary, but with some creative prompting, very accurate minutes and action-item lists can be extracted from the Zoom meeting transcripts.

Please note that the quality of the output not only reflects the native processing ability and training of the AI model in use but is also directly dependent upon the precision of the given input and the prompt provided. Preparing minutes in this manner will naturally encourage you and the board members to speak clearly and conduct the meeting in a professional manner more reflective of parliamentary procedure. As you work with generative AI, you will learn to fine-tune your prompts to obtain the exact results you seek.

In Season 8, Episode 5, “The Bells,” Daenerys Targaryen, consumed by a deep and uncontrollable fury, soared high above the city of King’s Landing on the back of the fearsome Drogon. As the beast’s mighty wings beat against the air, she unleashed a devastating torrent of dragonsfire upon the city below, incinerating buildings, streets, and its people in a relentless blaze, leaving King’s Landing reduced to ashes in her wrathful vengeance.

While some may argue she was justified in her actions, others might draw parallels to the risks of unleashing the dragon into the realm of community associations. Just as a single individual can destroy an entire city with dragonsfire, generative AI has the potential to empower one person to achieve much more than ever before. Though this article doesn’t seek to delve into possible job displacement, it raises a more profound question: In community management’s Game of Thrones, who will dare to walk through the fire and master the power of generative AI? Who will emerge as the House of the Dragon? “Dracarys!”