[HN Gopher] Disaster awaits if we don't secure IoT now
       ___________________________________________________________________
        
       Disaster awaits if we don't secure IoT now
        
       Author : mdp2021
       Score  : 63 points
       Date   : 2025-06-02 17:44 UTC (5 hours ago)
        
 (HTM) web link (spectrum.ieee.org)
 (TXT) w3m dump (spectrum.ieee.org)
        
       | HideousKojima wrote:
       | Not going to happen any time soon because there is no concern
       | about this from the consumer side, no financial incentive from
       | the manufacturing side, and no regulatory pressure from the
       | government (and I have low hopes that any regulatory solutions
       | would actually fix the problem).
        
         | Havoc wrote:
         | >no regulatory pressure from the government
         | 
         | No much, but there is some. Getting rid of default admin/1234
         | passwords was 100% a regulatory push
         | 
         | Small but important win
        
         | thawawaycold wrote:
         | What about EU's CRA?
        
           | HideousKojima wrote:
           | Doesn't really do anything to ensure the end-user truly has
           | ownership over the device and the ability to control what
           | software runs on it. 10 years of security updates is nice
           | (assuming the company making the device doesn't go out of
           | business in that time) but doesn't stop those devices
           | becoming vulnerable after that (and a truly useful device
           | will likely have more than 10 years of useful life). I don't
           | know the specifics of the CRA, but most proposed regulatory
           | solutions I've seen intentionally take control away from the
           | end-user.
        
       | spencerflem wrote:
       | I used to be more concerned with this but the longer ive thought
       | about it the more convinced I get that none of this matters.
       | 
       | Most tech gadgets are a distraction and are about as useful off
       | as on.
       | 
       | Industrial stuff sure, but if someone's internet fridge or smart
       | TV goes haywire, so what.
        
         | arcticbull wrote:
         | I think the risk isn't that your fridge is unable to suddenly
         | to phone home about your butter consumption but that it gets
         | turned into a giant botnet or joins some crypto mining pool.
         | Sensors don't have much horsepower but some of those smart
         | appliances have decent application processors.
        
         | eadmund wrote:
         | The issue is that one's fridge or light bulb becomes a lodgment
         | from which an adversary can launch LAN attacks (e.g. via
         | Ethernet) on one's more-sensitive devices, such as one's
         | personal computer (which contains more-or-less one's entire
         | life: banking details, correspondence and so forth).
         | 
         | I don't really care if the IoT device is compromised; I _do_
         | care if it is used to compromise my trusted devices.
        
         | raylad wrote:
         | This article is specifically discussing industrial Internet of
         | things. And I think it's a good and instructive article
         | actually.
         | 
         | Please read before commenting.
        
           | spencerflem wrote:
           | I did read the article. It was not very good about separating
           | the two cases (for example by saying the total # of IoT
           | devices to sound dramatic while also only listing examples of
           | industrial ones to sound more worrying).
           | 
           | I love and care pretty deeply about security, I run Graphene
           | OS & am trying to switch to Genode / Sculpt sometime this
           | year. _I care._ But I 'm coming to realize that for the
           | 'personal computing' use-case it doesn't matter much for
           | people who aren't nerds.
           | 
           | In any case, the sort of "hardware root of trust" reccomended
           | is an insidious way of going about it.
        
         | graemep wrote:
         | > Industrial stuff sure, but if someone's internet fridge or
         | smart TV goes haywire, so what
         | 
         | If large numbers of fridges or cars fail to work properly or go
         | haywire at the same time, it matters.
         | 
         | The article also mentions enabling DDOS.
         | 
         | The same with things like routers, any type of infrastructure.
         | 
         | We have increasing amounts of electricity generation and
         | storage moving to consumer devices. Its a good idea and could
         | make the system more resilient if done right, but if it relies
         | on insecure devices it would be a huge systemic vulnerability.
        
         | Reubachi wrote:
         | Encryption and non-repudiation to-from any deployed field
         | device back home is analogous to putting a lock and your ID on
         | your front door. It would deter the average "front door
         | unlocked" criminal of opportunity.
         | 
         | When does a device not authenticating itself and all it's comms
         | matter?
         | 
         | Mossad was able to (at great cost, and with an inept enemy)
         | hijack a telecom supply chain to embedd explosives in internet
         | radios.
         | 
         | Were there _any_ operational oversight to simply have these
         | devices call home with signed key challenge, the literal virus
         | OS /code execution/explosive would have at least been caught.
         | 
         | I only bring up THAT example to show that supply chains are as
         | exposed as a bad actor wants them to be.
        
         | Ekaros wrote:
         | Fridge could have interesting attack vectors. Say allow
         | temperature peak during evening and night and then lower it
         | before morning. Keep repeating this. You get lot of spoiled
         | food and worst case even dead people.
        
           | netsharc wrote:
           | How about home thermostats? Turn off the heating during a
           | snowmageddon, freeze some people to death. Opposite thing
           | during a heatwave.
           | 
           | Or a distributed attack, turn on all ACs of a city, and
           | overload the power grid...
        
       | hooverd wrote:
       | Secure from whom? I can envision a future where the real IoT
       | security risk is insider threat. Buried in your smart lightbulb
       | EULA there could be consent to be used as a residential proxy.
        
         | mystraline wrote:
         | > Secure from whom?
         | 
         | From the person who thought the sale was ownership. More often,
         | "sale" is 'trade green paper for a license of this physical
         | good, that they retain to do whatever with later at their
         | leisure'.
         | 
         | Look at the scam Nintendo is doing with the Switch 2:
         | 
         | Games no longer have any data other than a serial number to
         | download a game.
         | 
         | Hi tendon claims they can remotely destroy consoles they deem
         | 'modified'. Not 'removed from online play', actually full
         | digital destruction of device.
         | 
         | I support ownership, not this 'we may revoke at any time'
         | licensure.
        
           | rbanffy wrote:
           | > Hi tendon claims they can remotely destroy consoles they
           | deem 'modified'. Not 'removed from online play', actually
           | full digital destruction of device.
           | 
           | This is illegal in many jurisdictions.
        
             | mystraline wrote:
             | I would believe you, but we've seen PlayStation 3 OtherOS
             | play out. In that case, Sony sold it with 'install Linux'
             | as a feature. Later on, they poison-pilled an update and
             | killed that.
             | 
             | There was a lawsuit, in which class action plaintiffs got
             | between 9$, or $55 if you said you used OtherOS.
             | 
             | But if this was any of us, hacking Sony, we would be
             | rotting in prison. But Sony can hack millions of peoples
             | supposed property, and "here's your latte money, too bad,
             | so sad!"
             | 
             | This should have been a criminal trial, in which the
             | executives would go to prison, firmware should be rolled
             | out to reenable it, and make people actually whole.
             | 
             | So yeah, I fully expect Nintendo to remotely destroy
             | hardware, and get away with it. The concept of "you don't
             | own, but you license" is the toxic shit that allows this,
             | alongside DMCA 1201.
        
         | rbanffy wrote:
         | It doesn't need to be mentioned in the EULA.
        
       | NoboruWataya wrote:
       | For industrial devices, absolutely. For consumer devices, I think
       | it can be a double edged sword. The more "secure" IoT devices
       | are, the more locked down and difficult to jailbreak they are,
       | which is bad for consumers and just leads to e-waste when a
       | product gets abandoned by its manufacturer.
        
         | gjsman-1000 wrote:
         | Sometimes I wonder what would happen, as a beginning step, if
         | we mandated the ability to purchase unlocked bare hardware. The
         | law doesn't need to require (in the first phase) that iOS or
         | something be compatible with it, but you could at least walk
         | into an Apple Store or Nintendo Store and buy the bare
         | hardware, for your own purposes. Then we could claim it as
         | being only about choice, and allowing software competition,
         | without any plausible security risk.
        
         | rbanffy wrote:
         | In this case, you would be liable by any damage your
         | compromised devices might cause. Let's say you find updates in
         | your house battery/inverter are annoying and, then, someone
         | takes over it and cause a power surge in the neighborhood.
         | That'd not be on the hardware manufacturer, who did their best
         | to avoid such incidents, but on you, who clearly didn't.
         | 
         | If you knowingly use obsolete and insecure devices, you might
         | end up being liable for their actions.
        
           | OsrsNeedsf2P wrote:
           | Are there such examples of this happening? Or are there only
           | examples of corporations doing this and suffering no
           | repercussions?
        
         | dandanua wrote:
         | There are two types of security: one where a user is secure
         | against hackers and the other one where manufacturer has
         | "secured" its devices against users. The second type is
         | strictly stronger, and it seems companies will not be
         | interested in giving any real freedom to users, apart from
         | running a predefined and restricted set of functions. In other
         | words, you won't own a device, only the services it produces.
        
       | calmbonsai wrote:
       | This is pure academic theory-crafting. Who is going to pay for
       | this and how will it be financed?
       | 
       | Even in the industrial space, it's simply not going to happen
       | given the costs of securing wrt the business value and relative
       | risk aside from projects that have some sort of
       | community/municipal/national "critical infrastructure" domain
       | (think power generation) where a legitimate business case (think
       | in terms of MARR
       | https://en.wikipedia.org/wiki/Minimum_acceptable_rate_of_ret... )
       | can be made for safety and/or regulatory compliance future-
       | proofing and a source of cheap 3rd party (usually government)
       | capital is made available.
       | 
       | There are also huge legacy IoT investments in capital inefficient
       | low-margin businesses (think mining and ship-building) where this
       | sort of "maintenance" and retroactively applying "defense in
       | depth" makes little sense compared to wholesale generational
       | migration.
       | 
       | Nobody is going to do it until there's some sort of abrupt forced
       | exigent existential externality applied to the entire industry
       | (e.g. ozone layer depletion and refrigerants) and then they'll be
       | a single mass-migration.
        
         | ceejayoz wrote:
         | > Who is going to pay for this and how will it be financed?
         | 
         | As with climate change, "we all will" and "with great
         | difficulty, far too late".
        
         | hx8 wrote:
         | There's two very established routes for setting cyber security
         | standards.
         | 
         | * Government regulation.
         | 
         | * Insurance requirements. Mostly applies to businesses.
        
       | mystraline wrote:
       | > Guy Fedorkow is a Connection Science Fellow at MIT, a
       | Distinguished Engineer at Juniper Networks, and a contributor to
       | the Trusted Computing Group.
       | 
       | Talk about the worst corporate doublespeak - 'trusted computing'.
       | 
       | It also goes by DRM, or rental hardware, or you never actually
       | own it cause someone else retains permanent digital control.
       | 
       | There is NO trust here, only control and power in never actually
       | selling anything.
       | 
       | And since we're talking of IoT, this goes hand in hand with
       | proprietary corporate clouds, anti-FLOSS like Home Assistant,
       | rental in the form of sales, forced firmware upgrades that remove
       | previous features to gatekeep and resell what you promised.
       | 
       | I don't even need to read further. Anybody, and I do mean
       | anybody, who uses the moniker 'Trusted Computing', should be
       | ignored, blackballed, and relegated to the bin of computing.
        
         | kwk1 wrote:
         | > anti-FLOSS like Home Assistant
         | 
         | Could you expand on this?
        
           | mystraline wrote:
           | Not hard to. Go look at the wide variety of hardware that
           | does NOT work with Home Assistant, or works tenuously through
           | some heavily rate limited public web api.
           | 
           | There is a reason why for my IoT setup so far, its been
           | primarily OpenGarage, IKEA IoT (ZigBee), and existing compute
           | services I locally host.
           | 
           | Remote servers come and go. Companies come and go. If I am at
           | the forced behest of some company to 'bless' my hardware, its
           | not mine. And I think its fraud to even claim it a sale. Its
           | a 'rental as long as the company wishes'.
        
         | Stefan-H wrote:
         | Are you familiar with the academic field of security and the
         | notion of trust in trusted computing? The IoT devices that is
         | being discussed in the article are for industrial control
         | systems, not necessarily your home lightbulb. The threat model
         | is different. Do you want every municipal power company to be
         | trusted to properly vet the code they are putting on these
         | devices, or do you want to trust the device manufacturer to be
         | the one who can put code on the devices?
        
           | mystraline wrote:
           | Owner is still owner, be it someone who lives in a single
           | family residence, or that of a municipality.
           | 
           | In my area, tornado sirens are unencrypted aand a simple
           | recordable and replayable frequency. The cost to add an
           | encrypted radio connection is $100k for the base station, and
           | $25k per siren. There are 80+ sirens.
           | 
           | If this were open source, then a simple computer could he
           | retrofitted to do this. But because they are highly
           | proprietary, the county would be on the hook for $2.1M just
           | to defend against an asshole with a HackRF.
           | 
           | FLOSS and open principles should matter to governments as
           | well as individuals. Trading temporary easiness for no long
           | term usability is utterly ridiculous. And you end up with a
           | doorstop in the end either way.
        
       | GuB-42 wrote:
       | Please read the article before commenting, because I find the
       | proposed solution a bit worrisome.
       | 
       | Of course we should secure IoT, but the article is about one very
       | particular kind of security: roots of trust. The idea is that
       | devices shouldn't run unsigned software, so forget about custom
       | firmwares, and generally owning the hardware.
       | 
       | There is a workaround, sometimes called "user override", where
       | the owners can set their own root-of-trust so that they can
       | install custom software. It may involves some physical action,
       | like pushing a switch, so that it cannot be done remotely by a
       | hacker. But the article doesn't mention that, in fact, it
       | especially mentions that the manufacturer (not the user) is to be
       | trusted and an appropriate response is to reset the device,
       | making it completely unusable for the user. Note that such
       | behavior is considered unacceptable by GPLv3.
       | 
       | There are some cases where it is appropriate, GPLv3 makes a
       | distinction between hardware sold to businesses and "User
       | Products", and I think that's fair. You probably don't want
       | people to tinker with things like credit card terminals. But the
       | article makes no such distinction, even implying that consumer
       | goods are to be included.
        
         | AnthonyMouse wrote:
         | Not only that, "roots of trust" and locking users out of their
         | devices is the thing that _causes_ the IoS omnishambles. The
         | foundational problem is that some company makes millions of
         | devices and then goes out of business or otherwise stops
         | supporting them, but because the users are locked out of the
         | device, nobody else can do it either. Meanwhile people continue
         | to use them because the device is still functional modulo the
         | unpatched security vulnerabilities.
         | 
         | If anyone could straightforwardly install the latest DD-WRT or
         | similar then it's solved, because then you don't have to
         | replace the hardware to replace the software, and the
         | manufacturer could even push a community firmware to the thing
         | as their last act before discontinuing support.
        
           | myself248 wrote:
           | > and the manufacturer could even push a community firmware
           | to the thing as their last act before discontinuing support.
           | 
           | This should be held in escrow before the device can be sold.
           | And the entity doing the escrow service should periodically
           | build the software and install it onto newly-purchased test
           | devices to make sure it's still valid.
           | 
           | If the company drops support, either by going out of business
           | or by simply allowing issues to go unaddressed for too long,
           | then the escrowed BSP/firmware is released and the people now
           | own their own hardware.
        
             | AnthonyMouse wrote:
             | That seems like a lot of complicated when the better
             | solution is to have the people own their hardware from day
             | one.
             | 
             | You also need the community around the device to already
             | exist on the day support is discontinued instead of needing
             | to build one _then_ around a device which is by that point
             | years old and unavailable for new purchase.
        
             | specialist wrote:
             | I really thought escrow for software would've been SOP by
             | now.
             | 
             | We made EMRs in the 2000s. Our customers required
             | everything to be placed in escrow. It seemed abundantly
             | prudent to me.
             | 
             | Maybe even prescient; our startup was bought, then murdered
             | in its crib, leaving our customers SOoL. But at least they
             | got the source.
        
           | oezi wrote:
           | The issue is as much companies going out of business as
           | consumers buying devices from shit companies.
           | 
           | We need schemes which enforce security and which make long
           | term economic sense. I would require software escrow for all
           | companies to ensure a bankruptcy doesn't mean all software is
           | lost.
        
         | freedomben wrote:
         | I completely agree, and it disappoints me greatly to see
         | articles like this because it advocates doing what the
         | manufacturers already want to do so they can increase their
         | control. It's the exact type of article that people who want to
         | lock things down will use to discredit or disagree with someone
         | advocating a more ethical approach.
        
         | wmf wrote:
         | I would add that root of trust secures against rare, advanced
         | attacks like the "evil maid" or supply chain attacks. You
         | should worry about that _after_ you 've already secured against
         | basic vulnerabilities which 90% of IoT devices have not done.
        
         | Stefan-H wrote:
         | How user antagonistic changing code on IoT devices should be is
         | highly dependent on the threat model for the devices. I'm happy
         | to trust home users to flash their lightbulbs and door locks
         | (though the company might not see that as acceptable to their
         | brand reputation if their lock is compromised nonetheless), but
         | I would prefer not to trust the hundreds of IT departments and
         | engineering teams to properly vet the code they are flashing
         | onto industrial control systems when lives are at stake -
         | centralized authority and accountability with high visibility
         | on the code base that is flashed to the devices is what is
         | needed there.
        
       | jerf wrote:
       | You'll know full-on no-engineer-required AI is here when you can
       | point an AI at an IoT device and say "hack it", walk away for 30
       | minutes, and come back to a hacked device.
       | 
       | I'm not even being sarcastic. Most of them aren't that hard to
       | hack now as it is; I know a guy who broke at least two devices in
       | under an hour each because that's how bad they are. A piece of
       | junk that goes out today that maybe still flies under the radar
       | and nobody bothers to hack it isn't going to fly under the radar
       | in a world where there's 10, 20, 50 times more "software
       | engineering" power in the world, in the hands of a lot more
       | people. In 5 years those things are going to be a nightmare for
       | their owners, for their manufacturers, for all kinds of people.
        
         | AnthonyMouse wrote:
         | Finding a novel exploit against a device is kind of a reach
         | right now, but if you give it a device with public unpatched
         | vulnerabilities that were present in the AI's training data,
         | emitting code to exploit them given arbitrarily many attempts
         | is the sort of thing it can already do.
        
       | jandrewrogers wrote:
       | A challenge with IIoT is that the standards processes are
       | consistently driven to the lowest common denominator. It is
       | partly driven by cost sensitivity but also by the reality that
       | many of these hardware companies are unsophisticated at software
       | development and have no realistic way of becoming sophisticated.
       | 
       | Adding to this that the scale requirements for implementing these
       | standards varies by about 10 orders of magnitude between the
       | smallest and largest companies, which is such a large scale
       | difference that it becomes qualitative with respect to standards.
       | 
       | This creates divergent incentives. Many companies want to
       | optimize standards for the cheapest thing that will check a box,
       | even if the implementation is ineffective in practice. The
       | minority of companies that actually care about robust and
       | efficient implementation often find this isn't feasible (and in
       | some cases impossible) within the constraints of what ends up in
       | many of these standards, so they ignore the standards since they
       | are strictly worse than whatever non-standard thing they end up
       | doing.
       | 
       | Wash, rinse, repeat. I have participated in IIoT standardization
       | efforts for almost two decades and it is the same vicious cycle
       | irreconcilable requirements every time.
        
       | fidotron wrote:
       | This would be out of the frying pan and into the fire.
       | 
       | The only long term viable approach for IoT security is to not
       | allow these devices on the Internet in the first place. Have the
       | WiFi Access Point, or some other gateway, act as the broker for
       | all information, and the default is each device sees nothing
       | until given permission. *
       | 
       | Whenever this comes up people raise the point that this won't
       | work because it disincentivizes making devices to slurp data, but
       | it's not like that ecosystem actually exists at all, with the
       | exception of smart TV which hardly counts as IoT. Consumer IoT
       | hasn't taken off because consumers are rightly paranoid about
       | bait-and-switch and being left with useless devices in the walls
       | of their homes.
       | 
       | * This is roughly what https://github.com/atomirex/umbrella is
       | trying to head towards, hence seeing if a $50 AP can act as a
       | media SFU, and learning it totally can.
        
         | jamesgeck0 wrote:
         | Apple had a decent start on this. An always-on Apple TV can
         | connect to a Zigbee hub and provide remote access to devices
         | through HomeKit. But it seems like Zigbee isn't winning out as
         | the dominant standard. Most of the recent HomeKit improvements
         | have been related to the wifi-based Matter.
        
           | fidotron wrote:
           | Industrially there is a decent amount of LoRa (and relatedly
           | lorawan) about too, and that fills a similar role to Zigbee
           | in the comparison.
        
         | CalRobert wrote:
         | It's definitely an issue - I was at a heat pump company working
         | on data, and as much as the work was interesting (and I liked
         | doing something related to sustainability), there was a core
         | question of "wait, _why_ do we need to gather data from
         | everyone's thermostats again?"
         | 
         | Though as we move towards things like virtual power plants and
         | more integrated systems with home batteries, etc. the use case
         | is clearer.
        
         | rightbyte wrote:
         | > The only long term viable approach for IoT security is to not
         | allow these devices on the Internet in the first place.
         | 
         | Ye it is about that simple. IoT don't need the I. Given how low
         | my trust is for vendors I wouldn't even be happy with a
         | separate no internet wifi since the devices can hook up to some
         | other wifi.
        
           | fidotron wrote:
           | Exactly. The wifi these devices should be on should have no
           | access to anything except the broker in the gateway device.
           | 
           | Certainly no multicast or anything like that.
        
         | heraldgeezer wrote:
         | >Have the WiFi Access Point, or some other gateway, act as the
         | broker for all information, and the default is each device sees
         | nothing until given permission. *
         | 
         | Not everywhere is going to have WiFi. A SIM can use a private
         | APN that selects the PGW/IP/Network range and from the ISP
         | usually has a VPN to your network. Does not go "over the
         | internet" at all.
         | 
         | This is (usually) how industrial IoT, connected cars etc work.
         | Shipping containers cant rely on WiFi.
        
       | brookst wrote:
       | I was positive this should have had a (2012) but sure enough it's
       | a new article.
       | 
       | "Security is the 's' in IoT" was an old joke back then. Still a
       | problem but hardly a new one.
        
       | nobodyandproud wrote:
       | This is a technical "solution" which ignores a market and legal
       | problem: There's no incentive to keep things secure, because the
       | accountability falls on the rank-and-file, but rarely if ever on
       | the actual leaders that all but require insecure incentives.
       | 
       | We have Boeing-level incidents daily, except that it can be swept
       | under.
        
       | jmclnx wrote:
       | >How to secure IoT devices
       | 
       | Do not buy them and avoid built in ones at all costs.
       | 
       | Maybe at some point newer ones can be secured, but I doubt it.
       | 
       | Nevermind thousands have already be installed. People cannot be
       | bothered to update their phones or computers, those updates are
       | rather easy. Who will be bothered to update their IOTs ? Plus
       | 2038 is coming quickly, good luck to people who have 32bit IOTs.
       | Do 64bit IOTs even exist ?
       | 
       | >it's up to system integrators, who are responsible for the
       | security of an overall service interconnecting IoT devices, to
       | require the features from their suppliers, and to coordinate
       | features inside the device with external resilience and
       | monitoring mechanisms
       | 
       | So we give integrators access to our devices ? How is that
       | security ? Also what if these integrator's company is purchased
       | or goes out of business ?
        
         | pixl97 wrote:
         | Yea, it's not security and it's not how this shit works in
         | practice either.
         | 
         | "You by device that does a, b, and c"
         | 
         | Device updates.
         | 
         | "You now have device that does a, b, and d"
         | 
         | You ask what happened to c?
         | 
         | "You need to buy a new device to get feature c"
        
       | leptons wrote:
       | I can't take any article about IoT security seriously if it
       | doesn't mention the inability to secure the internet connection
       | with SSL. We're practically forced to use HTTP on some devices
       | instead of HTTPS, because you can't issue an SSL cert to a local
       | IP address.
        
         | foobiekr wrote:
         | The correct answer here is an attestation server and an 802.1AR
         | device certificate; this should be used to attest to a remote
         | server returned via DHCP or at a well known address (this
         | sounds evil but a well known address and specific VLAN is the
         | most widely deployed untrusted environment solution that
         | actually works) which uses a long lived certificate for which
         | the fingerprint is TOFU'd by the device. The device can use
         | this attestation and mutual challenge to obtain a certificate
         | (if it needs one) from an authority that other entities trust
         | (which is necessarily not the 802.1AR CA).
        
       | gigel82 wrote:
       | It's so disappointing from IEEE to be arguing against user's
       | freedom to own their devices.
       | 
       | Supply chain attacks concerns are infinitesimally smaller than
       | manufacturers writing buggy, unmaintained software / firmware.
       | 
       | We should instead argue for open software / firmware, or at least
       | local control only (if desired). Also solves the problem of
       | billions of e-waste devices whenever manufacturers go under or
       | shutdown their servers, or moves them behind new paywalls /
       | adwalls.
        
         | chickensong wrote:
         | They aren't arguing that, they're promoting defense in depth.
        
       | foobiekr wrote:
       | Industrial IOT is mostly secured through network segmentation.
       | There is a lot to hate about this fact.
       | 
       | However, by career I deal with a lot of embedded devices.
       | Embedded is where the hell begins. Embedded plus standards is
       | pretty bad. SCADA and others. Embedded plus standards plus
       | vendors .. now you are doomed. The half-assedness of embedded
       | systems security is worse than any normal coder can imagine.
       | 
       | Basically the IIOT domain is the one place that remains where
       | network segmentation and encirclement and so on are the only
       | things we have that work. The worst best option.
        
       | raminf wrote:
       | Any post about IoT security that doesn't mention or link to
       | Shodan (https://www.shodan.io) is missing a lot of context. It's
       | way worse than you think.
       | 
       | Also, with tools like Chip Whisperer
       | (https://www.newae.com/chipwhisperer) the physical security of
       | the hardware root of trust needs to be reevaluated.
        
       | mikewarot wrote:
       | Imagine if electrical devices were locked down in this manner.
       | You'd only be able to use appliances that the outlet
       | manufacturers approved. If they went out of business, you'd have
       | to have your house wiring stripped and replaced. The local
       | utility would also get a vote.
       | 
       | It's obvious to me that this model is unacceptable.
       | 
       | A better choice is to ensure that internet stacks are secure in
       | hardware, not software. This is how power is distributed.
       | Internet connectivity could be secured the same way.
       | 
       | Then you could use anything you want without worrying about
       | confused software or hackers, in the same way you can plug in
       | anything to any outlet without worrying about burning the house
       | down.
        
       | sublinear wrote:
       | > ... does anyone really want their refrigerator to automatically
       | place orders for groceries?
       | 
       | Yes, actually. I also want the groceries delivered to my door
       | using the services and stores of my choice depending on the items
       | and prices, and for it to integrate seamlessly with my Home
       | Assistant instance without any funny business regarding the API
       | or needing to install yet another bullshit app I didn't ask for.
       | 
       | This is the real reason we don't have nice things. The
       | implementation is always botched by businesses who don't do a
       | thorough job and finish their product. There's clearly a huge
       | chasm between what the customer and the IoT business would
       | consider an MVP. Do it right or don't do it at all.
        
       ___________________________________________________________________
       (page generated 2025-06-02 23:02 UTC)