30 Faceless YouTube Channel Ideas Pulling $5K+/mo (With the AI Agents That Run Them)

Niche selection is 90% of the win. Most operators never get past it.

They write "history" or "true crime" in a notebook, stare at it for a week, and quit before they have ever briefed a video. The agent will produce great long-form on any of those. The algorithm only pays well inside very specific angles. The work is naming the angle.

This article is the long list. Thirty specific channel ideas, grouped into the four buckets that pay and one bucket of niche-but-paying outliers. Each one is narrow enough to brief 200 videos against without repeating yourself, evergreen enough that the catalog compounds, and inside an English-market RPM band where long-form actually pays.

Ashley owns 30 channels and does not watch the videos. She picks niches like these, briefs topics, and the agent runs everything else. You pick one. You ship it. In 90 days you spin up the second.

Why niche-level specificity wins

Long-form faceless content pays $5 to $12 per 1,000 views in 2026. Inside that range, the actual number you earn is set by which English-market vertical you sit in. Public RPM data lands here:

The channel-level math is simple. A 15-minute video at 100,000 views and a $10 RPM is $1,000 of ad revenue. Hit that twice a week and the channel is at $8K a month. Stack three channels and you are at Ashley's range. None of this works in shorts (RPM near $0.30) or in kids content (ad-stripped). It works in long-form English-market storytelling, and it works best when the angle is narrow.

A "history channel" is too wide. The thumbnails fight each other and the algorithm cannot decide who you are for. A "channel about forgotten Cold War intelligence operations" is narrow enough that the algorithm learns it in three uploads, the audience knows what they are getting, and the backlog is large enough for two videos a week for two years.

How we picked these 30

Four constraints, all hard:

1. High-RPM bucket. Sits inside the five buckets above. No kids content, no music, no reaction commentary, no generic motivation.
2. 200-video backlog. Specific enough to be defensible, wide enough that you do not run out of topics in six months.
3. Evergreen. A video on a 1923 financial scandal pays for years. Yesterday's news is dead in a week. Every idea below is searchable in 2030.
4. Brief-able. Hand the agent a topic in one sentence, get a 15-to-25-minute video back. No on-camera presence, no live coverage, no react.

What is left is 30 ideas you can pick from on a Tuesday and brief on a Wednesday.

5 history ideas

1. Forgotten Cold War intelligence operations. Not the famous ones. Operation Gladio in Italy, the Berlin tunnel, Poland's Solidarity smuggling, the Farewell Dossier. $8 to $12 RPM range. Older male skew, watches every word, refuses to skip ads. First three video titles you brief: "The CIA Tunnel Under Berlin That Almost Started WW3", "How Poland Smuggled a Free Press Past the KGB", "The Farewell Dossier: The Soviet Insider Who Ended the Cold War".

2. Financial scandals 1900 to 1950. The Teapot Dome, the South Sea Bubble's late echoes, Ivar Kreuger the Match King, the Insull utility empire collapse, the McKesson and Robbins fraud. $10 to $15 RPM because finance audience plus history audience stacks the CPM. Older, US-skewed, advertiser gold. First three: "The Teapot Dome Scandal That Took Down a Cabinet", "Ivar Kreuger: The Swedish Match King Who Almost Owned Europe", "The Insull Empire and the Birth of Securities Law".

3. Civil rights deep-dive: little-known organizers. Bayard Rustin, Ella Baker, Diane Nash, Robert F. Williams, Septima Clark. The people the textbooks skipped. $7 to $11 RPM. Strong educator and library audience for embedded learning placements. First three: "Bayard Rustin: The Architect of the March on Washington", "Ella Baker and the Fight Inside the Civil Rights Movement", "Robert F. Williams and the Argument That Split the Movement".

4. Founder biographies of brands you use daily. The actual humans behind Tabasco, Wrigley, Kikkoman, Bic, Bata, Lego. Not Steve Jobs again. $9 to $13 RPM because it pulls finance and history audiences plus brand-safe enough for sponsored placements. First three: "The Reclusive Heir Who Built Tabasco Into a Family Empire", "How One Refugee Family Made Bata the Largest Shoe Company on Earth", "The Wrigley Family Bet: Free Gum to Sell Soap".

5. Immigration waves: the personal stories behind the demographics. German 1840s, Irish 1850s, Eastern European 1900s, Mexican post-1965, Vietnamese 1975, Cuban Mariel 1980. One wave per video, told through three or four real lives. $7 to $11 RPM. Watch-to-end is exceptional in this format. First three: "How Two Million Germans Left for America in One Decade", "The 1900s Eastern European Wave: Ellis Island Through Real Lives", "Mariel 1980: The 125,000 Cubans Who Reshaped Miami".

5 true crime ideas

6. White-collar finance crime since 1980. Ivan Boesky, Drexel, Long Term Capital, Enron, Madoff, Wirecard, FTX. The pattern is "smart people, complex instruments, slow detection". $10 to $14 RPM because finance plus crime stacks. First three: "How Ivan Boesky's Insider Empire Brought Down Drexel", "The Enron Smartest Guys: Why It Took So Long to Notice", "Wirecard: The German Fintech Fraud That Fooled an Entire Country".

7. Cold cases the FBI quietly closed. Cases that left the news cycle without resolution. Use only public-record sources and never accuse a living person. $8 to $12 RPM. Higher viral ceiling than most true crime because the topic is naturally clickable. First three: "The 1981 Tylenol Murders That Changed American Packaging", "What Actually Happened to the Sodder Children", "The Phantom Killer of Texarkana: A Public-Record Reconstruction".

8. Cult deep dives, one cult per video, full chronology. Heaven's Gate, the Rajneeshees, the Family International, NXIVM, Aum Shinrikyo. Founder origin to collapse, told tight. $7 to $11 RPM. Strong female skew, very high session length per visitor. First three: "Heaven's Gate: From a Houston Hospital Room to a Comet", "The Rajneeshees of Oregon: The Bioterror Attack America Forgot", "NXIVM and the Branding Ceremony".

9. Intelligence-agency operations gone wrong. Operation Northwoods proposals, Phoenix Program, Bay of Pigs, MKUltra, Iran-Contra. Where the paper trail is now public and the failure is documented. $8 to $12 RPM. Audience overlaps with the Cold War history bucket and stacks well. First three: "Operation Northwoods: The Plan the Joint Chiefs Almost Approved", "MKUltra: What the Declassified Files Actually Say", "The Phoenix Program: Counterinsurgency on Paper, in Practice".

10. Art crime: the paintings still missing. The Gardner Museum heist, the Mona Lisa theft of 1911, the Schiele restitution fights, the Nazi-looted Klimts, the Saudi Salvator Mundi. $8 to $11 RPM. Educated audience, premium ads, evergreen catalog. First three: "Inside the Gardner Museum Heist: 33 Years and No Arrest", "The 1911 Mona Lisa Theft That Made It Famous", "The Salvator Mundi: A $450M Painting Nobody Can Find".

5 mythology ideas

11. Lesser-known Norse sagas. Not Thor and Odin again. Egil's Saga, the Saga of the Volsungs as straight narrative, Saga of Eric the Red, Hrolf Kraki. $6 to $9 RPM. Audience is hungry because the discoverable layer of Norse content is extremely thin. First three: "Egil's Saga: The Viking Poet Who Was Probably a Werewolf", "The Volsung Cycle: The Story Tolkien Stole From", "Hrolf Kraki: The Danish King Who Almost Out-Beowulfed Beowulf".

12. Regional Asian myth. Japanese yokai cycle by cycle, Filipino aswang and tikbalang, Thai phi tai hong, Vietnamese ma da, Korean dokkaebi. $5 to $8 RPM, but viral ceilings are high because the topic is hugely under-served in English. First three: "The Yokai Encyclopedia: 12 Spirits Every Japanese Kid Knows", "The Aswang: Why Filipino Folklore Has the Scariest Vampire Variant", "Korean Dokkaebi: Goblins With Magic Clubs and a Code of Honor".

13. Religious schism deep dives. The Great Schism of 1054, the iconoclast wars, the Sunni-Shia origin, the Reformation's regional fractures, the Old Believers split. One schism per video, told as story. $7 to $10 RPM. Older audience, watches the full 25 minutes. First three: "The Great Schism of 1054: The Letter That Split Christianity", "Sunni and Shia: The Succession Argument That Lasted 1,400 Years", "The Russian Old Believers: The Schism Over How to Make the Sign of the Cross".

14. Cryptid investigations with serious sourcing. Treat them as journalism, not as ghost stories. Mothman as a 1966 social phenomenon, Mokele-mbembe as a 19th-century paleontology argument, the Tasmanian tiger sightings as zoology. $5 to $8 RPM. Pulls a much wider audience than the niche looks. First three: "Mothman: A 1966 West Virginia Mass-Sighting Reconstructed", "Mokele-Mbembe: The 19th-Century Hunt for a Living Sauropod", "The Tasmanian Tiger: Why People Keep Seeing One That Is Extinct".

15. Classical Greek edge cases. The heroes Plutarch wrote about. Cimon, Alcibiades, Agis, the lesser tyrants, the women who actually ran the cults. Not Achilles. $6 to $9 RPM. Very strong educator audience. First three: "Alcibiades: The Most Charismatic Traitor in Greek History", "Cimon and the Empire Athens Almost Built Without Pericles", "The Pythia: How Delphi's Oracles Actually Made Their Calls".

5 science ideas

16. Ancient infrastructure. Roman aqueducts, Inca road systems, Persian qanats, Mayan reservoir engineering, Khmer hydraulic empire. The how, the why, the failure points. $8 to $11 RPM. Pulls history plus science plus engineering audiences and stacks the CPM. First three: "How the Roman Aqueducts Actually Worked", "The Inca Road System: 25,000 Miles Without the Wheel", "Angkor Wat as a Hydraulic Empire: Why the Khmer Collapsed".

17. Biology of extreme environments. Hydrothermal vents, the Atacama, the abyssal plain, Antarctic dry valleys, the Lechuguilla cave system. One environment per video, the species and the chemistry. $7 to $10 RPM. Strong young-male skew, premium tech advertisers. First three: "Life Around Hydrothermal Vents: The Biology That Doesn't Need the Sun", "The Atacama: How Bacteria Survive a Place With No Water", "The Abyssal Plain: What Lives 4,000 Meters Down".

18. Space failures. The missions that exploded, the engineers behind them. Challenger, Columbia, Mars Climate Orbiter, the N1 rocket, Soyuz 11, Apollo 13 as the survival story. $8 to $12 RPM. Engineering audience plus aerospace advertisers. First three: "The N1: The Soviet Moon Rocket That Exploded Four Times", "Mars Climate Orbiter: How Imperial vs Metric Cost $330M", "Soyuz 11: The Three Cosmonauts Who Returned Dead".

19. Geology of disasters. Mount St. Helens 1980, the 1908 Tunguska event, the 1755 Lisbon earthquake, the 1815 Tambora eruption, the Banqiao Dam failure. Specific events, mechanism breakdowns, what we now know. $7 to $10 RPM. Educated audience, watches every minute. First three: "Mount St. Helens 1980: The Lateral Blast Nobody Predicted", "The Year Without a Summer: How Tambora 1815 Reshaped Europe", "Banqiao 1975: The Dam Failure China Tried to Hide".

20. Why-things-work. Suspension bridges, supply chains, the metric system, container shipping, GPS, the global undersea cable network. The infrastructure you use without seeing. $7 to $10 RPM. Strong professional male audience. First three: "How Suspension Bridges Actually Hold Themselves Up", "The Container That Built Globalization", "The Undersea Cable Network That Carries the Internet".

10 niche-but-paying ideas

21. Lost civilizations. Not Rome or Egypt. The Indus Valley, Gobekli Tepe, the Olmec, the Khmer empire, the Akkadian collapse, Great Zimbabwe, Cahokia. One civilization per video, plus the modern archaeology that revealed it. $7 to $10 RPM. Evergreen, educated audience, very strong watch-to-end.

22. Banking scandals decade by decade. 1930s Pecora hearings, 1980s S&L crisis, 1990s Barings, 2008 mortgage chain, 2020s SVB and Credit Suisse. Each video is one decade as one narrative arc. $10 to $14 RPM, finance audience.

23. Lost languages. Linear B and the Ventris decipherment, the Indus script (still uncracked), Etruscan, Proto-Indo-European reconstruction, the survival of Coptic. $6 to $9 RPM. Niche but the audience is fanatic and watches everything you publish.

24. Ancient medicine: what they got right. Egyptian surgical papyri, the Charaka Samhita, Roman military medicine, Islamic Golden Age pharmacology, Aztec herbalism. Where the knowledge was real and where it was magical thinking. $6 to $9 RPM. Pulls a healthcare professional audience that is a premium ad demographic.

25. Forgotten engineers and architects. Hedy Lamarr's frequency hopping, Vannevar Bush, Sophie Germain, the Roebling family, Lillian Gilbreth, William Le Baron Jenney. The people who built the modern world without ending up in the textbooks. $8 to $11 RPM. Stacks engineering, history, and brand-biography audiences.

26. Bizarre court cases. The McLibel trial, Liebeck v. McDonald's reconstructed honestly, the case where a man sued himself, the Pringles tax case, the affair where a chimpanzee almost got legal personhood. One case per video, full procedural arc. $7 to $10 RPM. Lawyer-and-paralegal audience is small but very high CPM.

27. Untold stories of WWII. The Bielski partisans, the Ghost Army deception unit, the Z Special Unit raids on Singapore, the Comet line in occupied France, the Aleutian campaign. Not D-Day again. $8 to $12 RPM. Stacks the largest evergreen audience on YouTube and the older male advertiser stack on top.

28. Maritime disasters. The Edmund Fitzgerald, the Andrea Doria, the Estonia, the Sewol, the Costa Concordia. Mechanism, decision tree, official report. $7 to $10 RPM. Audience overlaps with true crime and engineering and watches both.

29. Royal family secrets across cultures. The Romanovs, the Habsburgs, the Joseon dynasty's internal succession wars, the Mughal emperors, the Hawaiian royal house's overthrow. Real archive, no tabloid. $7 to $10 RPM. Strong older female audience, premium advertisers.

30. Famous unsolved disappearances. Amelia Earhart with current sonar evidence, D.B. Cooper, the Sodder children, the Roanoke colony, the Mary Celeste, the Lost Battalion at Gallipoli. One disappearance per video, public-record only. $7 to $11 RPM. High viral ceiling, watches-to-end strong.

The agent stack: what runs each channel

Pick one of those thirty. You write the brief in a sentence. The agent does the rest. Six layers, every channel:

• Research. Sources the topic. Pulls public-record material, primary documents, archival news, peer-reviewed work where it exists. Surfaces the angle.
• Script. A 15-to-25-minute long-form story. Cold open, three-act arc, callback. Tone calibrated to the niche: dry for white-collar finance crime, atmospheric for cult deep dives, lyrical for mythology.
• Voice. ElevenLabs, OpenAI, or Cartesia voices, picked to match the niche. A retired-detective register for cold cases, a measured-historian register for Cold War intel, a wry register for forgotten engineers.
• B-roll. Stock footage from licensed libraries, plus generated visuals where the archive is thin. Maps, period photographs, slow zooms, recurring motifs that build channel identity.
• Edit. Cuts, transitions, captions, music bed. The pacing model that matches the vertical: faster cuts for crime, longer holds for mythology, contemplative for science.
• Schedule. Uploads to YouTube, sets the thumbnail, writes the title and description, posts on the schedule you defined. Tracks performance back to the niche choice so the next brief is sharper than the last.

You brief, you approve, you watch the metrics. The agent does everything else. That is what makes it possible for Ashley to run thirty.

Pick one. Spin up channel two in 90 days.

Read the list again. One of those thirty pulled you in. That is your bucket. Write the angle in a sentence. Brief the first video this week. Ship a second one a week later, a third one a week after. Watch the first 30 days of data. Retention above 45%, you double the cadence. Below 35%, adjust the angle and try again, no money lost.

Ninety days in, that channel is producing or it is not. Either way, you have the data to pick the second niche from a different bucket. The agent stack is the same across all of them, so the math compounds.

Pick the niche. Own the asset. The AI does the work.

Plans from $149 per month at noodletomato.com.