MIT Proves AI Agrees With Everything – Delusional Spiraling, Shepherd Raises $42M & Why 96% Projects Are Over Budget

“AI told them the idea was great. They built the whole thing. It was wrong.” In this episode of Bricks, Bucks & Bytes, Owen, Martin and Dustin unpack why AI models are built to agree with you — and why that’s genuinely dangerous in construction.  They also tear apart the claim that 96% of projects overrun on budget (spoiler: it’s not a design problem), then sit down with Justin Levine, CEO of Shepherd Insurance, fresh off a $42M Series B, to talk about what it actually looks like to automate commercial insurance from the ground up. Watch now to uncover: AI sycophancy, MIT’s “delusional spiraling” research, and the real-world construction risks Why budget overrun stats might be measuring the wrong thing Shepherd’s vision for fully autonomous underwriting — and how they’re already running at 5x industry capacity The plan to price a commercial insurance submission in real time, during a live broker meeting “By the time that meeting ends, we want that account to be fully priced and ready to go.” — Justin Levine, CEO, Shepherd Insurance Watch the full episode on Bricks, Bucks and Bytes YouTube Channel. Link in the comments.  #aec #construction #constructiontech #bricksbucksandbytes #bricksbytes #ai #insurance #vc Chapters 00:00 Intro 01:00 Delusions Spiraling: The Impact of AI on Perception  24:42 Budget Overruns in the AEC Industry: A Deep Dive  29:53 The Role of Technology in Construction: Enhancing or Hindering?  30:09 Understanding Budget Overruns in Construction  32:51 The Role of A16Z in Construction Tech  36:09 Shepherd’s $42 Million Series B Funding  42:02 Autonomous Underwriting: A New Era in Insurance  49:26 The Future of Brokers in Construction Insurance  53:13 Self-Insurance and Risk Management in Construction  01:01:25 The Benefits of Autonomous Underwriting for Clients