The simplest yet most pertinent way to define social networking platforms would be as a shared public space where ideas, identities, and relationships get blended in layered ways. This unprecedented medley of technology and human tendencies has given rise to vast governance challenges that social networks are grappling with. The CEO of Soul App, Zhang Lu, has taken a unique path to battle these issues and fight nagging problems such as misinformation, online fraud, and cyberbullying that are choking social ecosystems.
The recently released 2025 Q3 Ecosystem Security Report from Soul sheds light on how the social networking app integrates AI technology, systematic governance, and user education into a three-dimensional model that not only safeguards its ecosystem but also nurtures a culture of empathy, participation, and trust.
The industry-wide trend is to let technological advancements handle the heavy lifting pertaining to platform security. Although this approach does work, it isn’t without its pitfalls. One of the most glaring issues is that it turns safety and security into a service that is the sole responsibility of the social networking app.
Another jarring issue is that since it does not involve users in the safety and security process, there is a looming risk of over-moderation, which can create an environment entirely too sterile to foster user engagement.
Considering these factors, Soul Zhang Lu’s team decided to challenge the traditional method by embedding co-governance into the community DNA of the platform. So, users weren’t just mute service recipients but active participants in the safety and security matrix of the app. This not only gave the safety protocol greater reach but also made the ecosystem more vibrant and appealing to the users.
Soul Zhang Lu’s 2025-Q3 Security Report revealed that the platform’s “Content Recommenders” and “Discipline Officers” programs now include over 78,000 active members. Together, they form a citizen-driven moderation network that identifies inappropriate behavior, recommends high-quality content, and encourages a sense of accountability within the community.
In addition to this, every day, 51,000 users act as in-app recommenders. By casting 22,000+ votes, these moderators help to produce nearly 2,900 high-quality posts. Moreover, Soul Zhang Lu’s team has turned technology into a “foundation of trust” of sorts.
Artificial intelligence is used to bolster the efficiency of the overall security system. Behind every layer of community interaction lies a robust AI-driven risk management system designed to detect, predict, and neutralize threats before they reach users. The Q3 report shed more light on the effectiveness of this system with actual figures. For instance:
- Soul App’s AI image anti-fraud model achieved an 80% reduction in false positives while improving coverage to over 90%, ensuring that users are protected from scams without disrupting authentic engagement.
- Soul Zhang Lu’s persona recognition model improved its accuracy to over 70%, enabling early identification of accounts that pose deception risks. This is a form of anticipatory governance instead of merely reactive.
Along with their proactive approach to on-platform security management, Soul App’s team also wanted to do their bit to protect public interest in the digital ecosphere. To achieve this goal, the platform’s cybersecurity division routinely collaborates with law enforcement and regulatory bodies to share data and assist in investigations. In the third quarter alone, Soul Zhang Lu’s team provided over 100 leads on harmful content, assisted in 80 verification requests, and participated in three police-enterprise cooperation programs.
Another unique part of Soul’s overall safety strategy is fostering a healthier community culture. Soul App recognizes that the tone of its community is defined as much by what it allows as by what it encourages. So, by combining AI moderation and human participation, the platform fosters a clear, civil, and inclusive communication environment.
To this end, in Q3, the app intercepted 4.39 million pieces of violating content, reviewed 15.78 million related comments, and took disciplinary action against 306,000 violating accounts. Additionally, group chat moderation has also become a focal point of the platform’s governance. With 5,600+ voice messages and 26,000 text messages intercepted daily, Soul helps users enjoy real-time interaction while significantly reducing exposure to toxicity or abuse.
Simply put, with this hybrid approach, Soul Zhang Lu has achieved what can only be termed “always-on” governance. The best part is that it mirrors the platform’s belief that digital civility should be the norm, not the exception.
Another remarkable aspect of Soul App’s safety framework is its emphasis on education as a step towards prevention. The company views education as the third and equally essential pillar of its “Technology + System + Education” strategy.
So, significant efforts are invested in cultivating a more informed user base. For this, Soul App regularly produces digital literacy content through its official channel, “Discipline Captain”. According to the report, this account released 59 educational posts covering key safety topics in Q3.
The topics covered ranged from preventing obscenity to tackling online harassment and from recognizing misinformation to preventing image misuse. This educational content from Soul Zhang Lu’s team reached a record 510,000+ users.
Soul also has a strict policy against cyberbullying. The platform recognizes that the emotional well-being of the app’s users is just as important a part of the security matrix as structural safety. So, the platform’s algorithms are capable of detecting and blocking aggressive interactions in real time. In Q3, Soul protected an average of 3,127 users per day from online harassment. Additionally, it handled 2,100 private chat incidents daily and sent 215,000 reminders.
These and the other figures from Soul’s Q3 Ecosystem Security Report are impressive to say the least, and they clearly point to a standout framework capable of delivering real-time results. Also, the information published in the report underscores the fact that the efficacy of Soul Zhang Lu’s system is attributed to the platform’s unique approach of balancing automation and humanity.
