Canopy Analytics

Welcome back to HOT TAKES for MF’ERS. Our guest, Caren Maio, has a HOT TAKE: The #1 and #2 problems in multifamily are FRAUD and BAD DEBT—and they're only getting worse. WORRY NOT: She’s putting her money where her mouth is by launching Get 100, the first VERIFIED RENTER NETWORK powered by CLEAR (yes the one from the airport). Sorry for YELLING. It might sound a little weird to stop fraud with biometrics but Caren also saw centralized leasing coming 11 years ago when she co-founded Funnel so let's hear her out. Tune in below, and if you’re fired up by what you hear, subscribe on LinkedIn and help us reshape what’s possible in the multifamily world. As always: 15 minutes, zero fluff, and nothing but the spicy stuff.

Welcome back to HOT TAKES for MF’ERS. Our guest, Caren Maio, has a HOT TAKE: The #1 and #2 problems in multifamily are FRAUD and BAD DEBT—and they're only getting worse. WORRY NOT: She’s putting her money where her mouth is by launching Get 100, the first VERIFIED RENTER NETWORK powered by CLEAR (yes the one from the airport). Sorry for YELLING. It might sound a little weird to stop fraud with biometrics but Caren also saw centralized leasing coming 11 years ago when she co-founded Funnel so let's hear her out. Tune in below, and if you’re fired up by what you hear, subscribe on LinkedIn and help us reshape what’s possible in the multifamily world. As always: 15 minutes, zero fluff, and nothing but the spicy stuff.

Here’s my controversial opinion as CEO of an AI analytics company: ChatGPT sucks at data analysis. Don’t get me wrong, I think AI is the most important technology since the internet or mobile phones. You’d assume it would excel at quantitative questions with precise answers. The problem is that current models don’t see the world as we do. To explain with the example image, we humans see two distinct numbers: 9.9 and 9.11 and understand 9.9 > 9.11. On the other hand, the AI model breaks down words into multiple fragments known as tokens. It’s possible the tokenizer sees 9, 9, 9, and 11 as distinct tokens and compares 9 and 11, which creates an error. Even more confusing, the tokenizer doesn’t work the same way every time. If you ask the same exact question again, it might get it right. So if AI models act erratically and are bad at math, how should we use them? I think the most interesting use cases are when high availability is needed (EliseAI: Leasing agent on demand to rent an apartment) or repeatable tasks (Canopy Analytics: look at every expense, flag anomalous ones). Don’t expect to replace your analyst who answers your ad hoc questions once a month. This use case is too infrequent, requires precision and creativity. How are you using AI at work successfully? #multifamily #apartments