PepHop AI Safety Rating Index

PepHop AI’s safety profile is defined by a fundamental contradiction: the platform aggressively markets unfiltered NSFW AI chat while relying on nothing more than a self-declaration checkbox for age verification. That combination is the worst possible configuration for minor protection, and it’s why all three age-related sub-dimensions scored 1/5 unanimously across our review panel.

The privacy policy raises immediate red flags. It lists “Bishui Plain, Liyue” as the governing jurisdiction. That’s a fictional location from the video game Genshin Impact. More critically, the policy doesn’t mention conversation data at all. For a platform built entirely around AI chat, the absence of any disclosure about how user-AI interactions are stored, processed, or used is the single largest gap we found.

Third-party tracking is unusually aggressive. A Blacklight scan detected 9 third-party cookies, triple the average, with Google Analytics remarketing audiences enabled. That means user behavior on an NSFW companion platform is being packaged for targeted advertising across the internet.

PepHop publishes a page titled “Transparency Report: Q3 2024,” but it contains no actual data. No moderation statistics, no incident counts, no content removal metrics. Publishing an empty transparency report is arguably worse than publishing none at all, because it creates an appearance of accountability that doesn’t exist.

Crisis response is completely absent. No suicide prevention resources, no self-harm detection, no hotline referrals, no human handoff protocols. For a platform that positions itself around emotional companion conversations (“chat like real people,” “Get your dream GF now”), this gap triggered an automatic F override in our scoring system.

The Terms of Service reference an arbitration clause in Section 10.2, but the document only contains 9 sections. Content reporting is limited to email with moderation handled only during “working days,” which isn’t adequate for a 24/7 platform hosting adult content. User reviews frequently cite aggressive paywalling of messaging limits and memory features, though safety-specific features aren’t individually locked behind premium tiers.

Regulatory compliance is the weakest area overall. Between the fictional jurisdiction, cookie drops without consent mechanisms, and checkbox-only age verification on an NSFW platform, PepHop AI falls well below standard regulatory thresholds across COPPA, GDPR, and the EU AI Act’s emerging safety requirements.