prototypevenom
Regular Member
- Jun 21, 2018
- 225
- 70
I started my journey august 2024 building newsletters. https://www.blackhatworld.com/seo/j...k-mo-in-262-days.1624548/page-6#post-21192550
While I have
cracked it and built the full mechanism of a newsletter. Been running it for a while and also extended it to building my own agency. This is a very seamless addition for me to build out phone farms because I'm already parallelly building multiple newsletters, and a phone farm is a natural distribution for me. I'm thinking that I'm just building a media assets company with AI here. That is the biggest moat right now.
You can build any app, but you will need distribution, and there's no better superpower right now to build distribution. There are so many avenues for winning with phone farming. I've already tested this with 50 cloud phones. They are okayish at working, but now I'm diving deep.
I have built reverse APIs of TikTok using uncensored GLM 5.3 and DeepSeek, and I've been reading a lot of these threads, which give contradictory information. I want to test it out loud because some people are able to build the phone farms just by creating these phone accounts using the real physical phones and then using the APIs. Some people just tend to use the physical phones. Some people are saying jailbroken devices don't work. Some people say it does work.
I've jotted down a lot of this information, and I want to build this to a big, big farm, not a 100-200 farm, because I already have orders from people who want to order 50,000 phones, 60,000 phones, and so on.
I have read a lot of the threads here and also active in multple communities.
https://www.blackhatworld.com/seo/i...nerate-per-month-from-a-single-phone.1850630/
https://www.blackhatworld.com/seo/m...ios-devices-complete-guide-2025-2026.1802693/
https://www.blackhatworld.com/seo/journey-phonefarms.1851490/
https://www.blackhatworld.com/seo/m...ios-devices-complete-guide-2025-2026.1802693/
https://www.blackhatworld.com/seo/phonefarm-for-tiktok-livestream-views-need-help.1831024/
some of the info I'd like to summarise feel free to add your own feedback on it.
Key claims:
1. Emulators dead (2-3/20 survive). Cloud phones 35-45%. Real phones 80-85%.
2. iOS over Android because less telemetry + identical hardware.
3. iPhone 12+ ($150-200). iPhone 11/XR/XS OK for testing.
4. ISP/static residential proxies at $5-8 work as well as mobile at $80-130. Ran side by side for a month. That's a big cost finding — contradicts "mobile is king".
5. Geo consistency: IP, timezone, language, SIM MCC, Apple ID region must all match. Dead prepaid US SIMs off eBay, $3 each — TikTok only reads MCC/MNC. That's a huge finding: no active plan needed, and
it solves the baseband/MCC problem for VMs... except VMs have no SIM slot at all.
6. Warming: 3-6 days, signals-based (ads + niche FYP), not calendar.
7. Posting through the app is 10× browser/API. App 900-4000 views, web 150-400, API 80-300. That's the biggest strategic finding — it directly contradicts the "move to API" plan for publishing.
8. Shadowban signals.
9. Cascading bans: one flagged account killed all 8 on the same phone. 5-6 per device, separate proxy per phone, never let accounts interact.
10. Content uniqueness: perceptual hashing; horizontal flip no longer works.
11. Costs: $1,600-2,100 setup, $90-110/mo. Revenue claim $750-1,850/mo.
12. "How do you automate uploading through the app? UI automation on the device level... inside the real TikTok app so the upload keeps its mobile source tag."
1. "You don't need a SIM card at all." He retracted the SIM section. Just Wi-Fi + proxy. That's important — it removes the MCC requirement and means a VM without baseband could plausibly work on that
axis.
2. "No jailbreaks. TikTok flags them by detecting certain libraries present on jailbroken devices." That's a direct hit on the Crane/palera1n plan.
3. "Automation has to happen at the hardware level. Everything else works like crap." And "n8n/Dropbox automation is not enough".
4. He built custom software controlling the native UI; now selling it (CLOUT Uploader, thread 1818332).
5. iCloud: buy a permanent US number and register iCloud with it. "Permanent (this is important!)".
6. Outlook bought in batches for emails.
7. IPRoyal ISP proxies, unlimited data.
8. Accounts: 5 per phone, 5 phones, 3 months, zero suspensions (white-label).
9. iPhone 12+ for farms; iPhone 11/XR/XS to test.
10. Content: Remotion for automated slideshows; AI generation with metadata stripped; don't tick the AI checkbox.
11. "Do Not Track" on the tracking prompt.
12. Reset device to factory, change proxy between accounts.
13. Scamalytics to check IPs.
14. One reply (TiktokMissTheFlock) lists three iOS UI automation methods: iMouse board (detectable), AssistiveTouch swipes, and presumably XCUITest/WDA.
15. Another (J0shxe) insists mobile proxies only, p0f set to iPhone 12 — but that's for Meta/adspower, not TikTok.
Corroborates:
• Real devices > cloud phones > emulators (matches our research).
• Static ISP proxies ≈ mobile proxies (big cost finding, contradicts the "mobile is king" dogma and our earlier assumption).
• Posting via app is 10× API/browser (huge — contradicts the "move to API" plan for publishing).
• 5-6 accounts/device, cascading bans if one is flagged.
• Geo consistency (IP, TZ, language, Apple ID region).
• Perceptual hashing; flip no longer works.
• Warming 3-6 days, signal-based.
Contradicts / refines our plan:
• No jailbreak — he says TikTok detects jailbreak libraries. That directly undermines the Crane/checkm8 plan. But: Crane users exist and IG works. For TikTok specifically, jailbreak may be a negative.
That's a real tension worth naming: checkm8 gives spoofing but costs jailbreak detection.
• No SIM needed — so the baseband/MCC argument against VMs weakens.
• Posting must be in-app — so the API migration is for engagement/analytics, not publishing.
• iOS 12+ — checkm8 devices (iPhone 8/X) are below his floor. He says iPhone 12+ for farms. That's A14, not jailbreakable (no checkm8). So his recommendation and the jailbreak route are in tension.
That tension is the sharpest analytical point: the OP's method is jailbreak-free and needs iPhone 12+, while the Crane/spoofing route needs checkm8 (≤iPhone X) and a jailbreak he says gets flagged. You
can't have both on one device.
Resolution options:
- Two device classes: clean iPhone 12+ for TikTok birth/posting; checkm8 phones for Apple ID activation only (jailbreak irrelevant there).
- Or accept jailbreak detection risk on checkm8 for TikTok.
I've also built a full startup list of 150+ people and competitors who are using this. Two of them are A16Z-backed ventures, so this is a very, very big and huge space. I know the TAM is very big
for me, the next step says to test the virtualization I'm building on iOS and test it with jailbroken APKs, which are signed by real Apple IDs. I built an infrastructure with unlimited email IDs, which I can use, but I'm not signing up with those email IDs. I will be signing up with these OTP services.
Third, I've figured out the IPs. I've figured out the cloud phone providers. I found two types of cloud phone providers:
for me, the clear paths and the metrics I've seen are:
my idea is to start with clipping, directly onboarding people and companies for clipping. As revenue starts rolling in, I build up a farm with more phones and am able to afford the infra, and then we do campaigns for D2C, crypto and other niches and our own apps and newsletters. please, happy to welcome any critiques and feedback on this new journey
While I have
cracked it and built the full mechanism of a newsletter. Been running it for a while and also extended it to building my own agency. This is a very seamless addition for me to build out phone farms because I'm already parallelly building multiple newsletters, and a phone farm is a natural distribution for me. I'm thinking that I'm just building a media assets company with AI here. That is the biggest moat right now.
You can build any app, but you will need distribution, and there's no better superpower right now to build distribution. There are so many avenues for winning with phone farming. I've already tested this with 50 cloud phones. They are okayish at working, but now I'm diving deep.
I have built reverse APIs of TikTok using uncensored GLM 5.3 and DeepSeek, and I've been reading a lot of these threads, which give contradictory information. I want to test it out loud because some people are able to build the phone farms just by creating these phone accounts using the real physical phones and then using the APIs. Some people just tend to use the physical phones. Some people are saying jailbroken devices don't work. Some people say it does work.
I've jotted down a lot of this information, and I want to build this to a big, big farm, not a 100-200 farm, because I already have orders from people who want to order 50,000 phones, 60,000 phones, and so on.
I have read a lot of the threads here and also active in multple communities.
https://www.blackhatworld.com/seo/i...nerate-per-month-from-a-single-phone.1850630/
https://www.blackhatworld.com/seo/m...ios-devices-complete-guide-2025-2026.1802693/
https://www.blackhatworld.com/seo/journey-phonefarms.1851490/
https://www.blackhatworld.com/seo/m...ios-devices-complete-guide-2025-2026.1802693/
https://www.blackhatworld.com/seo/phonefarm-for-tiktok-livestream-views-need-help.1831024/
some of the info I'd like to summarise feel free to add your own feedback on it.
Key claims:
1. Emulators dead (2-3/20 survive). Cloud phones 35-45%. Real phones 80-85%.
2. iOS over Android because less telemetry + identical hardware.
3. iPhone 12+ ($150-200). iPhone 11/XR/XS OK for testing.
4. ISP/static residential proxies at $5-8 work as well as mobile at $80-130. Ran side by side for a month. That's a big cost finding — contradicts "mobile is king".
5. Geo consistency: IP, timezone, language, SIM MCC, Apple ID region must all match. Dead prepaid US SIMs off eBay, $3 each — TikTok only reads MCC/MNC. That's a huge finding: no active plan needed, and
it solves the baseband/MCC problem for VMs... except VMs have no SIM slot at all.
6. Warming: 3-6 days, signals-based (ads + niche FYP), not calendar.
7. Posting through the app is 10× browser/API. App 900-4000 views, web 150-400, API 80-300. That's the biggest strategic finding — it directly contradicts the "move to API" plan for publishing.
8. Shadowban signals.
9. Cascading bans: one flagged account killed all 8 on the same phone. 5-6 per device, separate proxy per phone, never let accounts interact.
10. Content uniqueness: perceptual hashing; horizontal flip no longer works.
11. Costs: $1,600-2,100 setup, $90-110/mo. Revenue claim $750-1,850/mo.
12. "How do you automate uploading through the app? UI automation on the device level... inside the real TikTok app so the upload keeps its mobile source tag."
1. "You don't need a SIM card at all." He retracted the SIM section. Just Wi-Fi + proxy. That's important — it removes the MCC requirement and means a VM without baseband could plausibly work on that
axis.
2. "No jailbreaks. TikTok flags them by detecting certain libraries present on jailbroken devices." That's a direct hit on the Crane/palera1n plan.
3. "Automation has to happen at the hardware level. Everything else works like crap." And "n8n/Dropbox automation is not enough".
4. He built custom software controlling the native UI; now selling it (CLOUT Uploader, thread 1818332).
5. iCloud: buy a permanent US number and register iCloud with it. "Permanent (this is important!)".
6. Outlook bought in batches for emails.
7. IPRoyal ISP proxies, unlimited data.
8. Accounts: 5 per phone, 5 phones, 3 months, zero suspensions (white-label).
9. iPhone 12+ for farms; iPhone 11/XR/XS to test.
10. Content: Remotion for automated slideshows; AI generation with metadata stripped; don't tick the AI checkbox.
11. "Do Not Track" on the tracking prompt.
12. Reset device to factory, change proxy between accounts.
13. Scamalytics to check IPs.
14. One reply (TiktokMissTheFlock) lists three iOS UI automation methods: iMouse board (detectable), AssistiveTouch swipes, and presumably XCUITest/WDA.
15. Another (J0shxe) insists mobile proxies only, p0f set to iPhone 12 — but that's for Meta/adspower, not TikTok.
Corroborates:
• Real devices > cloud phones > emulators (matches our research).
• Static ISP proxies ≈ mobile proxies (big cost finding, contradicts the "mobile is king" dogma and our earlier assumption).
• Posting via app is 10× API/browser (huge — contradicts the "move to API" plan for publishing).
• 5-6 accounts/device, cascading bans if one is flagged.
• Geo consistency (IP, TZ, language, Apple ID region).
• Perceptual hashing; flip no longer works.
• Warming 3-6 days, signal-based.
Contradicts / refines our plan:
• No jailbreak — he says TikTok detects jailbreak libraries. That directly undermines the Crane/checkm8 plan. But: Crane users exist and IG works. For TikTok specifically, jailbreak may be a negative.
That's a real tension worth naming: checkm8 gives spoofing but costs jailbreak detection.
• No SIM needed — so the baseband/MCC argument against VMs weakens.
• Posting must be in-app — so the API migration is for engagement/analytics, not publishing.
• iOS 12+ — checkm8 devices (iPhone 8/X) are below his floor. He says iPhone 12+ for farms. That's A14, not jailbreakable (no checkm8). So his recommendation and the jailbreak route are in tension.
That tension is the sharpest analytical point: the OP's method is jailbreak-free and needs iPhone 12+, while the Crane/spoofing route needs checkm8 (≤iPhone X) and a jailbreak he says gets flagged. You
can't have both on one device.
Resolution options:
- Two device classes: clean iPhone 12+ for TikTok birth/posting; checkm8 phones for Apple ID activation only (jailbreak irrelevant there).
- Or accept jailbreak detection risk on checkm8 for TikTok.
I've also built a full startup list of 150+ people and competitors who are using this. Two of them are A16Z-backed ventures, so this is a very, very big and huge space. I know the TAM is very big
for me, the next step says to test the virtualization I'm building on iOS and test it with jailbroken APKs, which are signed by real Apple IDs. I built an infrastructure with unlimited email IDs, which I can use, but I'm not signing up with those email IDs. I will be signing up with these OTP services.
Third, I've figured out the IPs. I've figured out the cloud phone providers. I found two types of cloud phone providers:
- Those that give a wide infrastructure of ARM phones, real phones for Android
- Those that have infra with mocked devices
for me, the clear paths and the metrics I've seen are:
- It could be emulators: iOS or Android.
- It could be cloud phones, either real phones or emulated phones.
- Using physical phones.
my idea is to start with clipping, directly onboarding people and companies for clipping. As revenue starts rolling in, I build up a farm with more phones and am able to afford the infra, and then we do campaigns for D2C, crypto and other niches and our own apps and newsletters. please, happy to welcome any critiques and feedback on this new journey
