I don't know. Isn't it would be just the software to develop, without any work on the hardware end?
The security would be of course up to the end user.
To be fair, software costs just as much to develop as HW. The difference is that if you're lucky, you can just roll it out as an update.
It's generally considered terrible network practice to have multiple IP addresses, but it is hypothetically possible. As an example, you can do it in Windows, buried in the advanced settings somewhere.
From our perspective, IP addresses mostly seem like arbitrary divisions that keep devices from talking to each other. From the IT perspective, the whole point of IP addresses is to allow you to split your network into subnets (sub networks). Each device can talk to the other devices on the local network, but when it wants to talk to a device outside the local network, it sends the data to the router, to figure out how to get the data where it needs to end up. Each router has a map of the other routers it is connected to, and what subnets each of them know how to talk to.
If a device uses two IP addresses simultaneously, then it means that you just connected those networks directly.
There are a variety of reasons that IT guys want to keep networks separated. Security is the most obvious of those reasons, but there are also network design concerns where bigger networks end up having more bandwidth wasted for broadcasts. It also increases the risk that a broadcast storm could take out more devices.
I worked on a project a ways back where a device had two IP addresses at the same port (I think it was a robot). One IP to talk to an upper level system via the plant network, one IP to talk to IO. We had to spend about a ton in networking HW, with much more complicated configurations than usual, to keep the IO traffic separate from the plant traffic. It would have been so much simpler and cheaper if it had a comm card we could have used, or if it had one IP address and we put a NAT router in place for the plant comms.