What’s Changed: Character AI started pushing a willingness-to-pay survey to a wide slice of users in July 2026, floating optional premium tiers from $20 to $60 a month on top of the existing $9.99 c.ai+ plan. It is price testing, not a confirmed hike, but it tells you where the platform is heading. If the filter and the falling quality are your real problem, switching beats paying more.
Character AI going paid is the fear rippling through every companion subreddit this week, and the survey that started it is being read completely wrong by most people.
The prices people are screenshotting, $30, $40, even $60 a month, are not a new base rate for the app. From everything I can piece together, they are an optional add-on tier the company is testing on top of the current $9.99 c.ai+ subscription.
That distinction matters, because the panic assumes your free chats are about to vanish behind a $60 wall. That is not what the survey says. What it really signals is a company with shaky finances trying to find out how much its heaviest users will tolerate before they walk.
I have watched this platform meter, gate, and ad-wall its free tier all year, so the survey did not surprise me. What surprised me was the number.
This piece breaks down what the survey really asks, why the reaction is this loud, why $60 is not as random as it looks, and exactly what to do whether you plan to stay free or finally jump ship.
Is Character AI Going Paid or Just Testing the Waters?
Character AI is not going fully paid, but it is testing a higher-priced optional tier through an in-app survey.
The free tier still exists. The survey measures willingness to pay for a “larger model” or better service, and multiple users confirm the top price is framed as an add-on, not a replacement for c.ai+.
The way I read it, this is classic price discovery. The company shows different users different numbers to map the demand curve before it commits to anything. One user on the survey thread put it plainly: “I’m pretty sure the sixty dollars is an optional add-on to cai+.”
So no, nobody is forcing you into a $60 plan tomorrow. But a company does not survey 20 million people about premium pricing unless it is serious about charging more somewhere. Character AI’s value decline is the backdrop that makes this feel like the next squeeze rather than a one-off.
What is c.ai+: Character AI’s existing $9.99 per month subscription that removes ads, skips waiting rooms, and gives unlimited swipes on the current models.
What Does the Character AI Survey Really Ask?
The survey asks users if they would pay $20, $30, $40, or $60 a month for a larger model, with different users shown different price points.
It is A/B price testing. The high tiers appear to be an optional super-premium add-on stacked on c.ai+, not a hike to the $9.99 plan itself.
Here is where the confusion comes from. Because the survey randomizes the number, one person swears it said $30, another says $40, a third screenshots $60. They are all telling the truth.
That is how this kind of test works, and it is why the thread reads like nobody can agree on what is happening. To me, the real question the company is asking is whether a “whale” tier can exist here, the way it already does on rival apps.
The tell that it is an add-on and not a replacement came from the calmer voices in the thread. As one user noted, “it doesn’t sound like they would be increasing the price of c.ai+. This would be something additional.”
The smartest move users made was not just voting no. Several pointed out the survey has a written feedback box, and filling it with specific complaints does more than a bare rejection. Market tests get killed when the qualitative feedback is brutal, not just when the score is low.
Why Are Users So Angry About the Paywall?
Users are furious because they feel asked to pay more for an app they believe got worse.
The complaints center on price shock for a service they call barely functional, models that declined after the switch to open-source, and a content filter that still interrupts chats even for paying subscribers.
The quotes tell the story better than I can summarize it. “For 30$ I would go to shop and buy some healthy food,” read the top comment.
Another reaction cut deeper: “Ten a month is already pushing it. Thirty a month is diabolical for a barely working service.” One user did the international math and landed on “that would $85 in Australia,” noting it was more than their car registration.
The sharpest line captured the real grievance about the filter: “$60 to still get bob’d by the way.” Bob is the community nickname for the moderation system that cuts off chats. Paying more does not remove it, which means the single most common complaint would survive the price increase untouched.
What is “Bob”: The community’s nickname for Character AI’s content filter, the automated moderation that interrupts or blocks messages it flags, regardless of subscription tier.
What I keep coming back to is the sequence. The company capped free usage, users felt the models get worse, and only then did the pricing survey arrive.
That order is why people read it as monetizing scarcity they created, not selling extra value. One comment nailed the logic: “they downgrade the models and expect a reward?”
How Did Character AI Get Here?
Character AI reached this point through a year of escalating monetization: mid-chat ads in February 2026, a hard swipe cap in March, and a Charms currency that gates features free users once had.
The survey is the next step in a clear pattern, not a sudden pivot.
The timeline explains the exhaustion. Full-screen ads landed inside conversations in February and were fully rolled out by mid-April.
Then on March 18, 2026, the platform capped free swipes at roughly 400 per day per account and restricted “go-ons” and “memos” that used to be unlimited. If you want the deeper mechanics, the swipe limit breakdown covers how the cap behaves.
The currency tying it together is Charms. Once you hit the swipe cap or a thread gets long enough, you spend Charms to keep going, at rates like 175 Charms for 35 swipes.
You can now buy them on mobile, roughly $0.99 for 100 Charms up to $9.99 for 1,130. The complete Charms explainer maps every exchange rate if you want the full table.
Here is the myth I want to kill, because it drives bad spending decisions. Charms and c.ai+ do not lift the content filter, do not bypass age restrictions, and do not unlock anything the free tier blocks.
They buy usage volume and temporary ad removal, nothing more. Anyone paying in the hope of loosening the filter is buying the wrong thing.
Before: You are on the free tier, deep in a good scene, and you hit the daily swipe cap around message 60.
After: To keep regenerating you spend Charms at 175 for 35 swipes. At the mobile rate of $9.99 for 1,130 Charms, those 35 swipes cost about $1.55, and a heavy afternoon can burn three or four times that before you notice.
Why Is Character AI Charging $60 When It Used to Be Free?
Character AI is testing high prices because its unit economics are difficult and a $30 companion tier already exists elsewhere.
The company runs on an estimated $32.2 million in annual revenue against a valuation that fell to around $1 billion, and rivals like Grok Ani already charge $30 a month.
I am not defending the survey, but the business reality is not pure greed. Character AI has roughly 20 million monthly active users, down from a peak near 28 million, and its valuation dropped about 60% from $2.5 billion in 2024.
AI inference costs scale with how much people chat, not with revenue, which is exactly the difficult unit economics that Sacra has flagged for the company. Serving tens of millions of free heavy users is genuinely expensive.
The $60 number also is not invented in a vacuum. A direct competitor, Grok Ani from xAI, already requires the $30 a month SuperGrok plan with no annual discount, and CrushOn sells a top Deluxe tier at $49.90. High-priced companion subscriptions already exist and sell, so the survey is really Character AI asking whether it can join that club.
There is even data suggesting the value is there in principle. The Stanford willingness-to-accept research found the median person now needs $11.40 to give up generative AI for a month, with a mean of $124.50.
The value users get dwarfs current prices. The theory holds up. What breaks is that people do not feel that value from an app they think degraded.
| Platform | Entry price | Top tier | Removes the filter |
|---|---|---|---|
| Character AI (c.ai+) | $9.99/mo | $20 to $60/mo add-on (in testing) | No |
| Candy AI | $5.99/mo billed annually | $12.99/mo monthly | Fewer restrictions |
| Nectar AI | ~$4.99/mo annually | ~$9.99/mo plus credits | Fewer restrictions |
| Grok Ani (SuperGrok) | $30/mo flat | $30/mo, no annual plan | Partial |
What Should You Do if You Won’t Pay?
If you refuse to pay, you have two solid options: squeeze the free tier smarter, or move to an app that gives you more for less.
Free-tier survival comes down to conserving Charms and dodging the thread cap, while switching makes sense if the filter or falling quality is your real complaint.
Let me start with staying free, because most people can stretch the free tier further than they think. The tactics below are the ones I would run first before spending a cent.
- Start a new thread when you hit the message cap instead of paying Charms to continue. You lose the immediate history, so paste a two-sentence recap to rebuild the scene.
- Write a detailed persona and character definition up front. Tighter setups mean fewer bad replies, which means fewer swipes wasted regenerating.
- Buy the 20-Charm one-hour ad-free pass only when you truly need a clean session, not as a habit.
- Fill the survey’s written feedback box with specific complaints if you get it. Market tests die on qualitative feedback, so make yours count.
The honest truth is that none of that fixes the two complaints people repeat most: the filter and the sense that the models got worse. If those are your dealbreakers, no amount of Charms discipline helps, and this is where I would look at moving.
| Frustration | Why it happens | What to do |
|---|---|---|
| Filter interrupts good scenes | Moderation applies to all tiers, paid or free | Switch to a platform with fewer content restrictions |
| Replies feel worse than before | Shift from proprietary models to open-source ones | Try an app running newer companion-tuned models |
| Free usage keeps shrinking | Ads, swipe caps, and Charms gating | Move to a flat-rate plan with no metering |
| Paying more feels pointless | Premium does not remove the core limits | Compare real value before paying Character AI extra |
If you are ready to switch, the app I would try first is Candy AI. It runs closer to $5.99 a month billed annually, keeps persistent memory, and does not wall the core experience behind a filter the way Character AI does. You can see the current Candy AI pricing before deciding, then switch to Candy AI if it fits.
For people who care most about building and customizing a companion, I would look at Nectar AI as the alternative. Its entry plan sits near $9.99 a month with a credit system for images, and the customization depth is where it earns its keep.
Either one costs a fraction of the $60 tier Character AI is floating, without the filter that made “$60 to still get bob’d” the line of the week.
Before you commit to any switch, check whether c.ai+ is worth it at its current $9.99, because for some light users the existing plan still beats both the add-on and a full migration.
Frequently Asked Questions
Is Character AI going to start charging everyone?
No. The free tier still exists, and the survey tests an optional higher-priced tier, not a mandatory fee. Free chatting remains available, though it is increasingly limited by ads, swipe caps, and Charms gating.
How much is the new Character AI paywall tier?
The survey floats $20, $30, $40, and $60 per month, shown differently to different users. These appear to be an optional add-on stacked on the existing $9.99 c.ai+ plan, not a replacement for it.
Will paying for Character AI remove the filter?
No. Neither c.ai+ nor Charms bypass the content filter, age restrictions, or moderation. Paying buys more usage and removes ads, but the filter that interrupts chats stays active on every tier.
What are the best alternatives to Character AI?
Candy AI and Nectar AI are the two I would try first. Both cost far less than the $60 tier Character AI is testing, keep persistent memory, and apply fewer content restrictions than Character AI does.
Is the $60 survey price confirmed?
No. It is market research, not an announcement. Community members note the survey randomizes prices to gauge demand, and strong negative written feedback in the survey can influence whether the tier ships at all.
Quick Takeaways
- The $30 to $60 prices are an optional add-on tier being tested, not a confirmed hike to the $9.99 c.ai+ plan.
- Paying more will not remove the filter, the single most common complaint, so it fixes the wrong problem.
- Character AI’s $1 billion valuation and $32.2 million revenue explain the pricing pressure, but users feel the value dropped.
- Stretch the free tier by starting new threads and tightening personas before you spend a single Charm.
- If the filter or quality is your real issue, Candy AI and Nectar AI cost a fraction of the tested tier and skip the wall.
