In this episode, Dale and Alan discuss the recent AI Day from nPlan and the highlights of the event. They focus on GraphGen, a model that enables self-serve reporting in the construction and projects world. They also talk about Auto Report, which automates the process of creating lengthy reports, and Agent Barry, which can interact with different tools and generate reports. They explore the future of reporting, the role of AI in improving project outcomes, and the challenges of data sharing and controlling the narrative. They explore the idea of boundaries and silos in data sharing, highlighting the need to respect legal boundaries and competitive advantages. They also discuss the potential of peer-to-peer data networking, where access to data is shared without sharing the actual data. The conversation then shifts to the topic of GraphGen and its application in project planning and scheduling. They discuss the practicalities of using GraphGen, including the input process and the iterative nature of generating schedules. They also touch on the validation of schedules and the inclusion of different scheduling methodologies. The conversation concludes with a discussion on the limitations of data sets and the importance of feedback and improvement in AI models.

Takeaways

🧻 GraphGen is a model that enables self-serve reporting in the construction and projects world.

🧻 Validation of schedules is important, and feedback is crucial for improving AI models.

Chapters

01:54 Enplan's AI Day: Introducing GraphGen and Auto Report

21:59 Challenges of Data Sharing in the Construction Industry

28:26 Data Sharing and Boundaries

30:00 Exploring Peer-to-Peer Data Networking

32:26 Automating Project Scheduling with GraphGen

34:11 The Practicalities of Using GraphGen

39:50 Validating and Adjusting Generated Schedules

53:57 GPT-4.0 and Hallucinations