A (formerly) loyal but now deeply dissatisfied user

Dear SeaArt team (and whoever's making these decisions),

Your warning about blocking "pornography, violence, or community guideline violations" with no refunds on energy/compute credits is starting to feel less like moderation and more like a deliberate scam to drain user wallets while hiding behind vague policies. As a paying customer, I'm fed up—and so are thousands of others based on the flood of complaints online. This isn't just poor service; it's potentially deceptive and exploitative. Let me break it down why your system is broken and why users like me are seriously considering escalation, including legal recourse.

Rampant false positives are sabotaging legitimate use—and that's on you

Your filter is a joke: swimsuits, fitness outfits, artistic nudes, pin-up art, even basic cosplay or suggestive (but clothed) poses get slapped with NSFW tags constantly. Users are reporting 70-80% of SFW generations getting blurred or blocked—check Reddit, Trustpilot, Discord, or even your own forums for proof.

Telling us to "appeal it" (and wait hours or days) isn't a fix; it's you offloading your faulty AI onto us. If this overzealous filtering is due to lazy implementation or external pressures, own it. Otherwise, it looks like bait: lure users in with promises of creative freedom, then waste their paid credits on bogus flags.

Sudden policy shifts scream bait-and-switch—bordering on false advertising

SeaArt built its reputation on being NSFW-friendly—reviews from 2024 to mid-2025 praised it as "adult content without limits." We paid for that. Now? Filters cranked to maximum, LoRAs for innocent themes (like swimsuits or fantasy art) flagged as "risky," and even old generations retroactively censored.

This isn't evolution; it's a rug-pull. If you've pivoted to appease payment processors, app stores, or investors, admit it upfront. Misleading users about what they're paying for could qualify as deceptive trade practices under consumer protection laws in multiple jurisdictions (e.g., FTC in the US, or equivalent in EU/Korea). We're not idiots—we see the pattern.

No-refund policy for your mistakes is outright predatory and unfair

When the AI hallucinates something borderline from a clean prompt, or your filter wrongly flags it post-generation, our credits vanish. No appeal success? Tough luck. Even if we win an appeal, the time and frustration are on us.

This "you pay even for our errors" model is consumer-hostile at best and potentially illegal at worst—think breach of contract or unfair terms under laws like the Consumer Rights Act or similar. Why no automatic credit restore for confirmed false positives? It feels designed to profit from failures.

"Community protection" is a flimsy excuse for artistic censorship

Blocking actual illegal content (CSAM, non-consensual material, extreme violence)? Absolutely necessary—zero argument there. But your dragnet is ensnaring everything from classical art studies, romance illustrations, fashion shoots, to mild fantasy designs. This isn't safeguarding users; it's corporate cowardice, likely driven by Stripe/PayPal/app store demands.

The result? A platform that's strangling creativity while alienating the adult-oriented creators who made SeaArt viable. If "good user experience" means punishing innocent users, you've failed spectacularly.

Look, policies are fine, but yours are arbitrary, inconsistently enforced, and disproportionately punitive to users. This has gone beyond frustration—it's eroding trust and potentially violating consumer rights. If SeaArt doesn't address this transparently (e.g., by improving filter accuracy, restoring credits for errors, and clarifying NSFW boundaries), I and many others will escalate: filing formal complaints with consumer protection agencies, leaving scathing (but factual) reviews everywhere, and yes, exploring legal options like class-action lawsuits for deceptive practices, false advertising, or breach of implied contracts.

Fix this now:

Overhaul the filter for precision, not paranoia.

Implement immediate credit refunds for false positives.

Be honest about policy changes and external constraints.

Ignoring this will only accelerate user exodus and backlash. Ball's in your court—prove you're not just another profit-over-people AI company.

Sincerely,

A (formerly) loyal but now deeply dissatisfied user