anyone heard of AKITA? Security device for smarthome devices

[steinooo]

I wondered what people's thoughts were about this? https://www.kickstarter.com/projects/akita/akita-instant-privacy-for-smart-homes

I thought it was interesting, however am sure there are ways to achieve the same effect with plugins for pfSense etc.

I have some smart home stuff, and TBH the video did freak me out a bit about the lack of interest i've paif my own network security. I suppose it's like the old adage of a builders own house though!

 

[Caged]

Sounds like a snake-oil band aid for poorly secured IoT devices. It's just a way to sell their services to clean up any problem - it's like an Indian tech support scam but you pay for the device that requests the call.

The only way for IoT devices to be secure is for the device and the backend platform to be designed and supported by a capable team, and regularly patched. I would also put all the IoT stuff in its own subnet firewalled off from the rest of my stuff, with no inbound access (also turn UPnP off).

You could easily spend more cash on security services to try and make fundamentally insecure devices something approaching safe, or you could just toss out the devices that cause problems and make a mental note to blacklist that vendor. Sticking to products from companies that operate bug bounties, are popular enough to have had researchers prod at them etc. is the best way to go here.

 

[steinooo]

Sounds like a snake-oil band aid for poorly secured IoT devices. It's just a way to sell their services to clean up any problem - it's like an Indian tech support scam but you pay for the device that requests the call.

The only way for IoT devices to be secure is for the device and the backend platform to be designed and supported by a capable team, and regularly patched. I would also put all the IoT stuff in its own subnet firewalled off from the rest of my stuff, with no inbound access (also turn UPnP off).

You could easily spend more cash on security services to try and make fundamentally insecure devices something approaching safe, or you could just toss out the devices that cause problems and make a mental note to blacklist that vendor. Sticking to products from companies that operate bug bounties, are popular enough to have had researchers prod at them etc. is the best way to go here.

pretty good hard truths there. Athough it is likely a pretty basic thing for some of you posters and security people, I quite liked the concept (which, in my sphere has similarities to Microsoft's Advanced Threat Analytics) how it analyses patterns of traffic and if it's out of kilter then it blocks it. The skeptic in me picked up on the fact he said "known IP addresses" which makes me wonder whether you need to give all devices fixed IPs which is a pretty heavy requirement of use.

 

[Caged]

It just sits on the Wi-Fi and picks up on things that seem to be scanning the network (perhaps pretending to be certain devices to act as a honeypot) and then sends deauth frames. I wouldn't bother.

 

[Ev0]

Be interested to look into this a little more to see what it actually does under the covers given it’s just plugged into the network rather than receiving any logs or network traffic to analyse like your usual SIEM/NBAD/IDPS type stuff.

There’s also various DNS services that can be used to help in this area, Quad9 in particular and think OpenDNS also has some filtering/categorisation functionality that could be used to help out around requests to places that probably shouldn’t be contacted by an ‘iot’ device.