There are many avenues to exploitability, which makes most of this about surface area, and what types of threats you need to protect from. Just a few potential

There are many avenues to exploitability, which makes most of this about surface area, and what types of threats you need to protect from. Just a few potential weaknesses off hand: - Devices that have internal web servers, can be exploited through loading a bad web page, that makes requests forgeries to the device - Many smart home products use non wifi wireless (or even wifi) that when in range of your home network can be exploited. - The non wifi network can leak information. For example if you're detected as home or not. If I want to break into your home, it's much safer to know if someone is home. - While devices with automatic updates are badly needed, if the mechanism to provide updates can be spoofed, it can pull in a bad image (this require compounded exploits of course). Also think about a company hat goes out of business, and someone buy's up the update domains?? - If you invite a guest to your network, or a device that has been compromised before purchase (think store employee's), that can be a window into exploiting additional devices. - Some devices may use dynamic port forwarding (sorry can't remember the protocol name off hand), that may create dynamic forwarding rules in the model if the modem feature is enabled. I'm not saying this is common, but I could see people mistakenly believe that the device is protected behind a firewall when it isn't. Anyways, I think my message is, while allowing inbound connections to a smart home device significantly increases the risk, there are still many risks that exist even within your own network, that depending on the vulnerability does leave certain risks.