I work for IPinfo. We offer a residential proxy detection service, which you can check at ipinfo.io/my.

We had a previous discussion about surfacing visitor IP address resproxy status explicitly. Should we have some sort of badge or a more explicit alert to show if a site visitor's IP address is part of a residential proxy network?

Even though it is great for demonstrating the product's value, it is kind of a low-tier value. What can a user actually do when they realize their IP address is part of a residential proxy pool?

The first issue is that residential proxy SDK infiltration is massive. If you start connecting to different IP addresses and constantly check your IP address on our website, you will often see that many of those IP addresses were, at some point, part of a residential proxy pool. We provide frequency information showing how many times an IP address was observed in a residential proxy pool, with a default observation period of 7 days.

Then there is the question of what a user can actually do about it. If it is a controlled IT environment with paranoid IT admins, sure, they can actively monitor traffic and identify why their IPs are showing up in residential proxy pools. They can attempt to do something about it. But it is not easy even then.

Residential proxy SDKs can simply be baked into almost any smartphone or smartphone-derived OS that allows app installation through marketplaces. So, many residential networks are already cooked (because of android TVs). Moderate-scale NAT connections almost always see residential proxy flags, as do public Wi-Fi hotspot IPs, which we also detect.

Identifying the apps that are generating background network traffic is quite hard. You need some level of DNS monitoring or a network sniffer. Alternatively, you need router-level firewall software.

These SDKs are not always sending high-volume, constant traffic that makes them easy to detect. If you see a 100% residential proxy flag for your IP address, then they probably are. But in many cases, the traffic is intermittent and much harder to identify.

Nobody has an answer to what I should do when I see my IP address in a residential proxy pool. It has been accepted in spirit as a "consented malware" for the last few years. It is undetectable and extremely hard to remove because the SDK has been baked into apps themselves.

▲

juros 7 hours ago | parent [-]

there was a blog post linked on this thread explaining how proxy IP lists (spur, synthient, ipinfo et al) have little actionable value and introducing an alternative real-time approach to detection.

But it got flagged/downvoted into removal (twice!). Someone here has lots of HN accounts and doesn't tolerate free competition.

Disclaimer: I'm the founder and main researcher of the "flagged" company.

	▲	reincoder 6 hours ago \| parent \| next [-]
		I have been part of this community for over a decade, and in my experience the mods do take flagging and voting irregularities seriously when they are reported. If you believe there is manipulation happening on your posts, that is worth raising directly with them, since they have visibility we do not. --- On the broader point, we process 3 trillion requests last year, have more than 80 employees, and run a dedicated privacy engineering team led by an ex-cybersecurity company founder. We invest in research on new detection methods and stay closely engaged with the developer community. If there are specific gaps you see in our product, I would be glad to hear them and discuss. --- Whether any dataset, including residential proxy IP data, is valuable depends heavily on the application. Treating a dataset as invalid because it does not fit one particular model can lead to decisions on shaky ground. We are regarded as one of the more if not the most accurate IP geolocation providers, and we spend considerable effort on education, solutions architecture, and documentation so customers understand what our data can and cannot support. For example, IP geolocation, even at highest level of accuracy, is not a person identifier. It will never be a 1:1 replacement of GPS geolocation. Many of the largest companies in AI, anti-bot, fingerprinting, KYC, and CDN spaces use our data. If anti-bot systems and CAPTCHAs were fully reliable on their own, there would be less need for additional signals like residential proxy data. We do not assign a score or label an IP as good or bad. That judgment sits with the customer's own threat or analytics model. Residential proxy IPs are, by and large, mostly used in web scraping operations of many different forms. If a company sees a moderate to high amount of traffic mimicking human behavior, it can struggle to tell bot traffic apart from real users. Anti-bot mechanisms can help, but they add friction to the user experience, and they are not cheap to run at scale. Residential proxy detection data is one of the easiest zero-knowledge ways to gather intelligence. There is no need for users to solve a puzzle or for multi-page traversal to collect fingerprint data. All that is needed is the IP address. Our residential proxy data customers tend to be on the more sophisticated side of cybersecurity. Suggesting that this data is a silver bullet for all their security needs would not reflect well on their expertise or ours. We present the data as is, and from there we work with customers on the right solution for their case.
	▲	knighttt 7 hours ago \| parent \| prev [-]
		[dead]