Datacenter blocklists still look like the default proxy defence in a lot of fraud stacks. This week's capture window was softer on volume, but the mix did not flip: newly seen proxy exits were still almost all off datacenter space. The datacenter slice held at 2.2%, the same share as the last posted recap, still in the flat ~2% band of late October.
Between 9–15 November 2026 (UTC), the IP99 risk feed captured 9,914,335 proxy exit addresses it had not seen in the previous 7 days (−8% vs the previous 7 days). Source page: ip99.com/en/proxy-exits (generated 16 November 2026 UTC). The public page rolls daily over the last 7 complete UTC days; this is the window it reports today, continuing the series after the 5–11 November recap (the two windows overlap on 9–11 November).
Against that last posted week (10,026,346 addresses), volume is only down about −1.1%. The page's official −8% compares 9–15 November with 2–8 November, not with the overlapping posted window. A large part of that official step-down is compositional: the comparison window still includes the heavier days of 3 November (1,642,470) and 4 November (1,752,678), which had already rolled out of the last posted recap. These figures describe what entered the 7-day mirror, not worldwide proxy supply.
What those addresses sit on
| Usage type | Addresses | Share |
|---|---|---|
| Residential broadband | 7,951,392 | 80.2% |
| Mobile network | 1,245,122 | 12.6% |
| Gateway / NAT egress | 425,455 | 4.3% |
| Datacenter | 214,585 | 2.2% |
| Education network | 49,101 | 0.5% |
| Unclassified | 28,680 | 0.3% |
About 63% were rated high-risk at capture; the rest still carried a proxy-class signal at a lower risk level.
IP99 counts an address as "captured" when it enters the feed with a proxy-class signal (proxy, VPN, dial-up pool, cloud phone, or hijacked shared proxy) and was not in the live 7-day risk mirror. Addresses that go quiet and return can be counted again. The page describes what 60 upstream collection sources delivered — not the total number of proxies in the world. An address is counted once per window. No address lists are published.
Why 2.2% still is not "mostly datacenter"
If your defence is mainly an ASN or hosting blocklist, you are still mostly looking past the problem. Residential broadband alone is 80.2% of this week's capture window. Mobile adds another 12.6%. Together that is 93% of newly seen proxy exits sitting in access networks where real customers also live.
The datacenter row is unchanged at 2.2% versus the last posted recap (5–11 Nov), after easing from 3.2% (4–10 Nov) and 3.6% (29 Oct–4 Nov) toward the 2.0% of late October. A hosting-only blocklist still misses most of the supply-side traffic that shows up as residential or mobile exits.
Residential share eased slightly from 80.7% to 80.2%, with mobile (12.2% → 12.6%) and gateway / NAT egress (4.0% → 4.3%) taking the difference — small mix moves inside the same access-network story.
Usage type answers "what kind of network is this," not "is this address currently sold as an exit." Rotating residential proxies and dial-up pools are designed to sit in DYN / MOB space. Detecting them needs supply-side observation with an age attached, not a permanent network-type label. Product notes: rotating residential proxies, detect proxy / VPN.
Which networks show up most
Top ASNs in this week's capture window (by distinct address count):
| ASN | Addresses | Share |
|---|---|---|
| AS4134 | 914,073 | 9.2% |
| AS4837 | 691,730 | 7.0% |
| AS9808 | 340,319 | 3.4% |
| AS6057 | 255,175 | 2.6% |
| AS36903 | 213,898 | 2.2% |
AS4134, AS4837, and AS9808 remain the three largest contributors. The top five is unchanged from the 5–11 November recap: AS6057 still fourth, AS36903 still fifth (share 2.0% → 2.2%). AS55836, fifth in the 29 Oct–4 Nov recap (2.2%), remains eighth this week (1.8%, up from 1.6% last time). Large access networks appearing here are a clue about where proxy supply is being sold from — not a claim that every customer on those networks is a proxy. Ban a whole ASN and you take out legitimate home and mobile users. Prefer per-address evidence. ASN pages: ip99.com/en/asn/.
Day-to-day shape
New-to-mirror captures that day, and how many were high-risk at capture:
| Day (UTC) | Captured | High-risk at capture |
|---|---|---|
| 2026-11-09 | 1,371,503 | 883,005 |
| 2026-11-10 | 1,467,002 | 940,244 |
| 2026-11-11 | 1,217,179 | 695,544 |
| 2026-11-12 | 1,576,485 | 1,000,400 |
| 2026-11-13 | 1,606,501 | 994,471 |
| 2026-11-14 | 1,422,246 | 896,773 |
| 2026-11-15 | 1,253,419 | 804,951 |
Volume was down on the official comparison (−8% vs 2–8 November); upstream sources sat at 60 (56 in the last posted recap). Daily capture stayed in a 1.22–1.61M band, lightest on 11 Nov (1,217,179) and 15 Nov (1,253,419), heaviest on 13 Nov (1,606,501). That is still not a shape that makes a static list look "caught up."
How to use this without over-blocking
- Branch on evidence, not network type. Residential is where your customers live; do not hard-block on DYN / MOB alone.
- Step up, don't stop. A fresh proxy-class signal on signup, a coupon claim, or a login from a new device justifies a second factor or a delay — not an immediate ban of a whole building's shared egress.
- Read the age.
evidence_state: "stale"means the address was in supply before and is not now; treat it as a weak prior. Re-query at decision time rather than caching a verdict for days. Score0withevidence_state: "none"is not a clean bill of health — it means no current verifiable evidence is on file. - Keep the timestamp. Store
computed_atwith the decision so a disputed block can be traced back to when the observation was made.
IP99 by ThreatHunter publishes the free weekly aggregate so you can see the mix without buying a feed. The score on a single lookup is the freshness and strength of verifiable risk evidence — not a probability that the address "is an attacker." Do not treat IP99 scores as identical to every other ThreatHunter dataset you may have seen elsewhere; this article only claims what the public proxy-exits page and free /v1/ip document.
Check one yourself
Free, no signup:
curl https://ip99.com/v1/ip/8.8.8.8 Look at evidence_state (active / stale / none), risk.score, and the proxy-class signals. Verified 16 November 2026, 13:05 UTC: public DNS infrastructure here still shows a hosting network attribute with score 0, evidence_state: "none", and usage_type IDC — a network fact, not a proxy verdict. The same lookup is available to agents as an MCP tool at https://ip99.com/mcp (ip_risk_lookup). Field notes live on the API docs.
Try it on ip99.com, keep an eye on this week's proxy exits, and wire the free OpenAPI at ip99.com/openapi.json for Postman or codegen.