The FTC Rule That Bans Fake Followers, and Why It Does Not Ban Selling Real Accounts

Buying 10,000 followers to make an account look bigger can now cost you a federal civil penalty in the United States. Selling that same account, with an audience real people actually built, cannot. Those two sentences look almost identical. They are not. The difference is one federal rule, 16 CFR 465.8, and most of what circulates about it in this market is either wrong or a paraphrase of something wrong. This article quotes the rule, quotes the agency that wrote it, and marks the places where the answer is genuinely unsettled. It is not legal advice.

What 16 CFR 465.8 actually says

The rule is short. Here it is in full, from the Electronic Code of Federal Regulations:

It is an unfair or deceptive act or practice and a violation of this part for anyone to: (a) Sell or distribute fake indicators of social media influence that they knew or should have known to be fake and that can be used by individuals or businesses to materially misrepresent their influence or importance for a commercial purpose; or (b) Purchase or procure fake indicators of social media influence that they knew or should have known to be fake and that materially misrepresent their influence or importance for a commercial purpose.

That is the entire operative text. Notice what is not in it. No mention of accounts. No mention of handles, usernames, channels, or profiles. No mention of transferring or selling anything other than the indicators themselves.

The rule sits inside 16 CFR Part 465, the Rule on the Use of Consumer Reviews and Testimonials. Most of that regulation is about fake reviews. Section 465.8 is the piece that reaches social media metrics. The FTC published it on August 22, 2024, and it took effect 60 days later on a 5-0 Commission vote.

To make out a violation of 465.8(b), you need every one of these at once:

• An indicator of social media influence, as the rule defines that term.
• That indicator is fake, as the rule defines that term.
• You purchased or procured it.
• You knew or should have known it was fake.
• It materially misrepresents your influence or importance.
• The misrepresentation is for a commercial purpose.

Miss one element and the rule does not apply. That is not a technicality, it is how rules work. Every honest reading of 465.8 takes the elements seriously. Almost every dishonest summary collapses them into "the FTC banned buying and selling social media."

The two definitions that decide everything

Section 465.8 does almost none of its own work. Two definitions in 16 CFR 465.1 carry it, and if you have not read them you have not read the rule.

Indicators of social media influence

The FTC defines these as "any metrics used by the public to make assessments of an individual's or entity's social media influence, such as followers, friends, connections, subscribers, views, plays, likes, saves, shares, reposts, and comments." That list is broad and meant to be, written to cover metrics that do not exist yet.

But an indicator is a number. An account is not a number. The definition covers the metric attached to a profile, not the profile, not the handle, not the content, and not the audience underneath the metric. Nothing in Part 465 defines an account as an indicator.

Fake indicators of social media influence

This is the definition that decides cases. Fake indicators are "indicators of social media influence generated by bots, purported individual accounts not associated with a real individual, accounts created with a real individual's personal information without their consent, or hijacked accounts, or that otherwise do not reflect a real individual's or entity's activities, opinions, findings, or experiences."

Every branch points at the same thing: fabrication. A bot. A sock puppet. A stolen identity. A hijacked account. Or the catch-all, an indicator that does not reflect real activity. Put that back into the rule and read it once more. The prohibited object is a fabricated metric. Not a large metric. Not an expensive one. Not one someone paid to grow. A fabricated one.

Why the rule does not reach the sale of a real account

The strongest evidence that 465.8 targets fabrication rather than transactions is not in the rule. It is in the FTC's own Statement of Basis and Purpose, the document the Commission published to explain what it was doing and why.

During the comment period, a consumer organization asked the FTC to go further. It suggested removing "the word 'fake' from the Rule to clarify that it covers the purchase or procurement of any social media engagement . . . from both real and fake accounts unless those incentives can be disclosed to people who can view the engagement."

The Commission said no. Its answer, verbatim: "The use of incentivized indicia of social media influence is not necessarily deceptive in all cases, and it is beyond the scope of this rulemaking."

That exchange matters more than any commentary. The FTC was asked, on the record, to extend the rule past fabricated metrics into real engagement. It declined, on the ground that the conduct is not necessarily deceptive. The word "fake" survived because the Commission wanted it there.

The same document draws the line again from the other direction. Asked what "fake" means, the FTC wrote: "If a social media influencer were to recommend that their followers also follow another social media account, any resulting followers of the second account would not be 'fake.'" Real people, prompted by a real recommendation, following a real account. Not fake.

The rule attacks fabricated metrics. It does not attack the existence of a large audience, the movement of an account between owners, or the fact that money changed hands. But the rule not banning something is not the rule blessing it. Section 465.8 says nothing about whether your platform's terms allow a transfer, or what the buyer does afterward. It answers one narrow question and stays quiet on the rest.

The table below applies the elements to the fact patterns that actually come up. The last column is the reasoning, because a conclusion without reasoning is just another blog post telling you what to think:

Conduct	Inside 465.8?	Why
Buying 10,000 bot followers to make your business account look bigger	Yes	Textbook 465.8(b). Fake indicators, procured knowingly, misrepresenting influence for a commercial purpose.
Running a panel that resells bot followers, views, or likes	Yes	465.8(a), selling or distributing fake indicators to people who can use them to misrepresent influence.
Selling an account with an audience real people built	No	An account is not an "indicator" under 465.1, and real followers are not "fake indicators" under 465.1(h). No element is met.
Paying a real creator to promote your account, gaining real followers	No	The FTC addressed this directly: "any resulting followers of the second account would not be 'fake.'"
A private individual buying followers with no commercial angle	No	The rule reaches misrepresentation "for a commercial purpose." The FTC wrote that such individuals "are excluded from the rule's scope."
Hiring an influencer who turns out to have fake followers	No	The FTC narrowed "distribute" in 465.1(g) to avoid this. Vicarious liability here "was not the Commission's intention."
Innocently buying a growth package that secretly delivered bots	No, if you truly did not know	The FTC added "knew or should have known" for this scenario. But "should have known" is doing real work.
Buying an account whose follower count you know is mostly bots, then trading on that number	Unsettled, arguably yes	See the next section. No court or enforcement action has reached this fact pattern.

The padded account problem, where the line gets genuinely hard

Everything above is the easy part. Here is the part nobody in this niche writes about, because it cuts against the comfortable answer.

An account has 50,000 followers. About 40,000 are bots the previous owner bought. You know the number is padded, you buy the account anyway, and then you quote 50,000 followers to an advertiser.

Read 465.8(b) against those facts. You procured fake indicators of social media influence. You knew they were fake. They materially misrepresent your influence. You did it for a commercial purpose. Every element is arguably present. That the fake followers arrived attached to an account, rather than in a panel order, is not obviously a defense.

The FTC's own commentary makes this harder to wave away. Commenters asked it to narrow "procure" to a defined exchange. It refused, explaining that "procure" carries "its ordinary, everyday meaning, that is, to obtain something," and that "any exchange of value in order to obtain fake indicators of social media influence would be 'procuring' the fake indicators." Buying a padded account is such an exchange.

The Commission also rejected an argument that would have helped a buyer here. Asked to change "can be used" to "are used" in 465.8(a), it declined, reasoning that fake indicators "are not physical products that people collect and then use later as desired. Instead, their existence is premised on and limited to situations in which they appear deceptively on a social media site. Therefore, any person or business that obtains fake indicators of social media influence is misrepresenting their social media influence."

That sentence should give any buyer of a padded account pause. On the FTC's own reasoning, the misrepresentation is not a separate step you take later. It happens because the fake number is sitting on the profile.

Now the caveats, because this is analysis and not a holding. No court has applied 465.8 to an account sale, and the FTC has not brought that case. A seller has a real argument that they transferred an account, not indicators. A buyer has a real argument that they inherited a data-quality problem. Reasonable lawyers would disagree.

But notice what this does to the popular framing. "The FTC banned fake followers, and account sales are fine" is half a thought. The accurate version is narrower and more useful: the rule does not care that you sold an account. It cares whether the numbers on it are real. If they are, you are outside the rule. If they are not, the wrapper you put them in may not save you.

Who is liable, and who is not

Section 465.8 says "anyone." That is broader than most of Part 465, which regulates businesses, and it is deliberate. The rule reaches both ends of the transaction.

Paragraph (a) covers anyone who sells or distributes fake indicators: the SMM panel, the follower shop, the growth service that quietly delivers bots. The FTC narrowed "distribute" in 465.1(g) to mean distribution "to individuals or businesses who could use the indicators to misrepresent their influence," which keeps the paragraph aimed at suppliers rather than everyone downstream.

Paragraph (b) covers anyone who purchases or procures them. This is the part people miss. There is no safe side of this transaction. The agency that sold the package and the brand that ordered it are both inside the rule, and "I only bought them, I did not make them" is not a defense the text supports. "Procure" is the word that reaches the business that never touched the panel: if your agency buys the followers and bills you, you obtained fake indicators through an exchange of value.

Two groups sit outside the rule. Individuals acting without a commercial purpose, whom the Commission wrote "are excluded from the rule's scope." And people who genuinely did not know, because the FTC added "knew or should have known" after retail commenters objected that the proposal could catch a business that "innocently sold or procured a fake indicator, without knowledge or any indication that the indicator was fake."

Do not read "should have known" as a loophole. It is a diligence standard. When 40,000 followers arrive in 48 hours on an account that gets nine likes a post, the question of what you should have known answers itself.

What actually changed on October 21, 2024

Here is what almost every summary gets backwards. Selling fake followers was already illegal before this rule. The FTC had said so. A state attorney general had said so. The rule did not make the conduct unlawful. It changed what happens to you when you do it.

Before Part 465, the FTC reached fake-engagement sellers through a workaround: the "means and instrumentalities" theory, arguing that supplying fake metrics gave customers the tools to deceive, which is itself deceptive under the FTC Act. It works, but it is indirect, and a bigger problem sat behind it. In 2021, the Supreme Court decided AMG Capital Management v. FTC, which gutted the agency's ability to recover money under Section 13(b) of the FTC Act. The FTC said as much in the press release announcing this rule: "The Supreme Court's decision in AMG Capital Management LLC v. FTC has hindered the FTC's ability to seek monetary relief for consumers under the FTC Act. This rule will enhance deterrence and strengthen FTC enforcement actions." It added that "case-by-case enforcement without civil penalty authority might not be enough to deter clearly deceptive review and testimonial practices."

That is the actual story of 465.8. Violating a trade regulation rule opens the door to civil penalties in a way a plain deception claim does not. The rule did not redraw the line between legal and illegal. It moved the consequence from "stop doing that" to "stop doing that and pay." If you sell fake engagement, the risk did not become new in October 2024. It became expensive.

The penalty figure almost every article gets wrong

Search this rule and you will see the same number everywhere: $51,744 per violation. It appears in nearly every write-up, usually with no source and no date.

Here is where it came from. The FTC's final rule notice, published in August 2024, answered a commenter asking about the financial consequence of a fake review: "The maximum civil penalty is currently $51,744 per violation." Accurate when written. Not the current number.

The maximum is adjusted for inflation each January under the Federal Civil Penalties Inflation Adjustment Act Improvements Act of 2015. The current amounts live in 16 CFR 1.98, which now reads $53,088 for violations of Section 5(m)(1)(A) of the FTC Act, the provision covering knowing violations of a trade regulation rule. By its own terms, those amounts "apply only to penalties assessed after January 17, 2025." The $51,744 figure is a snapshot from before that adjustment that got frozen into every article about the rule.

That is the small correction. Here is the big one, and it comes from the same FTC sentence everyone quotes half of. The full passage reads: "The maximum civil penalty is currently $51,744 per violation, but courts must take into account the statutory factors set forth in Section 5(m)(1)(C) of the FTC Act and may impose much lower per-violation penalties. Ultimately, courts will also decide how to calculate the number of violations in a given case."

The FTC is telling you three things there, and the internet reliably repeats only the first. It is a maximum, not a rate. Courts may go much lower. And nobody has settled how you count violations, which is the entire ballgame. If each fake follower is a violation, a 10,000-follower order is an extinction-level number. If the order is the violation, it is one. The agency that wrote the rule says courts will decide, so anyone quoting you a total is guessing.

Use the number the way the FTC frames it: a ceiling that moves with inflation, applied by courts weighing statutory factors, on a per-violation basis nobody has defined. Still a serious deterrent. Not a price list.

Devumi, the case that shows what enforcement looks like

The rule has a prehistory, and it tells you who actually gets sued.

In January 2019, New York Attorney General Letitia James announced a settlement with Devumi LLC that her office called the first in the country "to find that selling fake followers and 'likes' is illegal deception and that fake activity using stolen identities is illegal impersonation." Devumi sold followers, likes, and views generated by bots and sock puppets across Twitter, YouTube, LinkedIn, SoundCloud, and Pinterest. Some of those fake accounts copied real people's names and pictures without consent.

One finding in that release maps directly onto the knowledge standard the FTC would adopt five years later. The AG found Devumi's practices "deceived some of the company's own customers who mistakenly believed they were paying for authentic endorsements, while many other Devumi customers knew they were buying fake activity and endorsements." Two groups of buyers, same product, different states of mind. That distinction is now the difference between violating 465.8(b) and not.

In October 2019, the FTC settled its own case, calling it the agency's first-ever complaint challenging the sale of fake indicators of social media influence. Devumi filled more than 58,000 orders for fake Twitter followers, made more than 4,000 sales of fake YouTube subscribers and over 32,000 sales of fake YouTube views, and sold more than 800 fake LinkedIn followers to marketing firms, software companies, banks, and investment firms.

The buyers were not anonymous scammers. The FTC alleged Devumi sold fake Twitter followers to "actors, athletes, musicians, writers," and to "motivational speakers, law firm partners, investment professionals, and others who wanted to boost their credibility to potential clients." The order banned the defendants from selling social media influence and imposed a $2.5 million judgment against owner and CEO German Calas, Jr., suspended upon payment of $250,000. Andrew Smith, then director of the FTC's Bureau of Consumer Protection, put it in one line: "Posting fake reviews on shopping websites or buying and selling fake followers is illegal. It undermines the marketplace, and the FTC will not tolerate it."

Note what the FTC did and did not do. It went after the supplier of fabricated metrics. It did not touch the market for real accounts, because that was never what the case was about. Real numbers, real transfers, no rule. Fake numbers, any wrapper, a rule with teeth.

Fake engagement destroys resale value on its own

Set the law aside. Regulatory risk is the second reason not to pad an account. The first is that it does not work.

The money in this market is downstream of real attention, and the payout mechanics prove it. X's Creator Revenue Sharing program requires an active Premium subscription, at least 500 verified followers, and at least 5 million organic impressions in the last three months. The gate is impressions, not followers. Bots do not read your posts. A padded account can carry 100,000 followers and fail the impressions test forever, because the followers that would have generated those impressions do not exist.

The incentive to fake it is real, which is why the rule exists. Goldman Sachs Research estimates the creator economy could roughly double from about $250 billion to about $480 billion by 2027, yet finds only about 4 percent of global creators earn more than $100,000 a year. A big follower count is the shortcut people reach for when the gap between those two facts gets uncomfortable.

It is a bad shortcut, because the buyers on the other side of the table check. An account with 50,000 followers and 12 likes a post is not hiding anything. The ratio gives it away in about four seconds, which is why anyone serious runs a follower audit and an engagement calculation before talking price rather than after. Padding only works on someone who does not look. It suppresses the engagement rate, the metric experienced buyers price off, and when it surfaces mid-deal it does not renegotiate the price, it ends the deal. The mechanics are in organic vs bought followers, and the specific tests are in how to check if an X account has real followers.

Valuation runs the same direction. Account age, posting history, and audience quality drive price more reliably than raw follower count, which is the argument in why account age beats follower count and the basis for how our valuation tool weighs inputs. Padding inflates the one number sophisticated buyers trust least while damaging the ratios they trust most. A regulatory problem stacked on a fraud problem stacked on a bad trade.

Platform rules are a separate layer from federal law

This is where people conflate two systems that have nothing to do with each other. Federal law and platform terms are different rulebooks with different enforcers and different penalties. Complying with one tells you nothing about the other.

Section 465.8 is the FTC's. Break it and you face a federal agency and potential civil penalties. Terms of service are the platform's. Break them and you face suspension. The FTC will not suspend your account. X will not fine you $53,088. The rulebooks are not consistent with each other, which is where the friction lives:

• X prohibits handle sales in its username squatting policy: "Attempts to sell, buy, or solicit other forms of payment in exchange for usernames are also violations and may result in permanent account suspension."
• TikTok covers it in section 3.2 of its Terms of Service: "Do not give others access to your account, or transfer your account to anyone else, without our permission." The same section says TikTok may revoke, reclaim, or reassign a username after 180 days without a login, or if the account is banned.
• Telegram's published terms and FAQ contain no explicit clause prohibiting the sale or transfer of channels. That is worth stating carefully, because plenty of blogs claim otherwise without citing anything. Absence of a ban is not permission, and Telegram can still act on spam, scams, or suspicious ownership changes.

None of that is federal law, and none of it changes what 465.8 says. A transfer can be permitted by the FTC's rule and prohibited by the platform's terms at the same time.

Courts, meanwhile, have started treating accounts as property. In JLM Couture, Inc. v. Gutman, decided January 17, 2024, the Second Circuit held that disputed social media accounts "should be treated in the first instance like any other form of property," then set out the framework: "the analysis of social-media-account ownership begins where other property-ownership analyses usually begin, by determining the account's original owner. The next step is to determine whether ownership ever transferred to another party. If a claimant is not the original owner and cannot locate their claim in a chain of valid transfers, they do not own the account." You can read the full opinion here.

Hold that against the platform view and you get the real tension in this market. Paula Brillson of the Digital Law Group described the platform side to Sherwood News: "You have no right to a username. You're a guest. You're leasing space and that is your ID." A court applying property law and a platform applying its terms can reach opposite conclusions about the same account. That is why a documented chain of transfer matters.

What this means before you pay

Fake metrics are one species of a larger problem, and that problem is getting worse. According to the FTC's April 2026 data spotlight, people reported losing $2.1 billion to scams that started on social media in 2025, about eight times the 2020 figure, and nearly 30 percent of everyone who reported a scam loss said it started there. The Commission's own footnote gives the full series:

Year	Reported losses to scams originating on social media
2020	$261 million
2021	$789 million
2022	$1.2 billion
2023	$1.5 billion
2024	$1.9 billion
2025	$2.1 billion

The FTC adds the caveat that matters: "since most scams aren't reported to the government, the real losses are likely much higher." It cites research finding only about 4.8 percent of mass-market fraud victims complained to a government entity or the Better Business Bureau. An account that looks influential is a better vehicle for a scam. That is the harm 465.8 aims at.

The practical takeaway from the rule is not legal, it is procedural. Section 465.8 turns "I did not know the followers were fake" into a claim you have to support, because the standard is what you knew or should have known. Checking is now the difference between a defensible position and an indefensible one. Before money moves, on any account, on any platform:

• Run the follower base against the engagement it produces. A follower audit and an engagement calculator answer the only question the rule cares about.
• Check reach independently. A shadowban check tells you whether an audience that looks real is actually being served.
• Look at growth shape, not just growth. Real audiences accrete. Purchased ones arrive in steps. The method is in our due diligence checklist.
• Ask the seller directly, in writing, inside the deal, whether any followers were purchased. A seller who will not answer in writing has answered. Keep that record: if the metrics later turn out to be fabricated, the difference between a buyer who asked and one who did not is the entire "should have known" question.

This is also, plainly, why listings here get reviewed by a person before they go live. Every listing on PlayerSells is manually checked for bots, fake followers, shadowbans, and past suspensions, on X, Telegram, and TikTok alike. A marketplace that lets padded accounts through is not just selling a bad asset, it is handing both sides a problem the FTC has now written a rule about. Rejecting those listings costs us a fee. Approving them would cost our buyers more.

Escrow does the rest. Your payment sits locked until you have inspected the account and confirmed it matches the listing. If the followers are not what the listing said, that is what a dispute is for, and it is decided on the evidence in the deal record. No proof, no release. The mechanics are in how it works, and the fee schedule is on the pricing page.

The same distinction shapes the promotion services side. A promoted tweet, thread, or retweet from a real account with a real audience produces real followers, and the FTC addressed that exact fact pattern: any resulting followers "would not be 'fake.'" A panel that sells you 10,000 bots produces fake indicators. One is reach. The other is a number.

Frequently Asked Questions

Is it illegal to buy followers?

In the United States, yes, if the followers are fake and you are doing it for a commercial purpose. 16 CFR 465.8(b) makes it an unfair or deceptive act to purchase or procure fake indicators of social media influence that you knew or should have known were fake and that materially misrepresent your influence for a commercial purpose. A private individual with no commercial purpose sits outside the rule, because the FTC wrote that such individuals "are excluded from the rule's scope." That is a narrow exclusion, not a general one.

Is it illegal to sell a social media account?

16 CFR 465.8 does not prohibit it. The rule reaches fake indicators of social media influence, meaning metrics generated by bots, sock puppets, stolen identities, or hijacked accounts. An account with an audience real people built is not a fake indicator, and the FTC declined a specific request to extend the rule beyond fake metrics into real engagement. That answers the federal question only. Platform terms are separate, they vary, and they are enforced by suspension rather than by law.

What is the fine for buying fake followers?

The maximum civil penalty is $53,088 per violation under 16 CFR 1.98, applying to penalties assessed after January 17, 2025. Most articles still quote $51,744, the figure in the FTC's August 2024 rule notice before that inflation adjustment. Treat both as ceilings, not rates. The FTC itself notes courts must weigh the statutory factors in Section 5(m)(1)(C) and "may impose much lower per-violation penalties," and that courts "will also decide how to calculate the number of violations in a given case."

Can I get in trouble for buying an account that has fake followers?

Possibly, and this is the least settled question in the rule. If you knowingly buy an account whose follower count is fabricated and then trade on that number commercially, the elements of 465.8(b) are arguably all present, since the FTC defines "procure" as simply "to obtain something" and says any exchange of value to obtain fake indicators counts. No court has decided this fact pattern and the FTC has not brought the case. The practical answer is the same either way: audit before you pay, and keep the record of having done it.

Does the FTC rule ban buying engagement from real people?

No. The Commission was asked to do exactly that and refused. A consumer organization asked it to remove the word "fake" so the rule would cover engagement "from both real and fake accounts," and the FTC responded that incentivized indicators are "not necessarily deceptive in all cases" and the question was "beyond the scope of this rulemaking." It also confirmed that if an influencer recommends their followers follow another account, the resulting followers "would not be 'fake.'" Separate FTC endorsement disclosure rules can still apply to paid promotion.

Does 465.8 apply to individuals or only businesses?

Both. The rule says "anyone," which is broader than most of Part 465. The limiting element is not who you are, it is the commercial purpose requirement. A business padding its account is covered. An agency procuring the padding for a client is covered. An individual with no commercial angle is excluded. Someone building a personal brand to attract clients or sponsorships is acting for a commercial purpose, whatever their job title says.

Is buying a Telegram channel or TikTok account legal?

The FTC's fake-follower rule does not address it, because a channel is not an indicator of social media influence. TikTok's terms are explicit that you should not transfer your account without permission. Telegram's published terms and FAQ contain no equivalent clause, which is not the same as approval. The mechanics and risks differ by platform, and we cover the Telegram process in how to buy a Telegram channel safely.

How do I check if an account has fake followers before buying?

Start with the ratio between audience size and the engagement it produces, because that is the hardest thing to fake and the easiest to test. Run a follower audit, calculate the engagement rate, look at whether growth accrued gradually or arrived in steps, and check whether the account is actually being served with a shadowban check. Then ask the seller in writing whether any followers were purchased, inside the deal record where the answer is preserved. The full sequence is in our due diligence checklist.

The short version

16 CFR 465.8 bans fabricated metrics. It does not ban the transfer of a real account with a real audience, and the FTC declined a direct invitation to make it do so. Both halves of that sentence matter, and most coverage drops one of them.

What the rule really did was change the price of faking it. The conduct was already illegal before October 21, 2024. Now it carries civil penalty exposure the FTC did not have when it took Devumi apart with an indirect theory.

The market had already priced it in. Buyers audit. Engagement ratios do not lie. Payout programs gate on impressions, not followers. Fake engagement failed as a strategy before the FTC wrote it down, because everyone downstream of the number checks the number. The rule just added a federal agency to the list. None of this is legal advice, and the padded-account edge is genuinely unsettled. If you are structuring transactions that depend on it, get a lawyer, not an article.

If you are buying, audit before you pay and keep the deal in escrow until the account matches the listing. Start with the follower audit, then browse the marketplace. If you are selling, the fastest route to a clean sale is an account that survives an audit, which is the whole argument for never touching a panel. List it free and let the numbers hold up. To see why the escrow flow is built the way it is, read how escrow protects every deal or the safety guide.