Reddit Deploys AI to Block Generative Engine Optimization Spam

Reddit Inc. is deploying advanced artificial intelligence and large language models to combat generative engine optimization, or GEO. This stealth marketing practice involves brands planting fake user reviews and discussions to manipulate AI chatbots such as ChatGPT and Gemini into citing specific products as trusted community opinions.

To counter these efforts, Reddit implemented automated systems that now block 23 million spam views and identify approximately 25,000 suspicious posts and comments daily. Between January and March 2026, the platform reduced user exposure to spam by 20% compared to the first quarter of 2025 and revoked nearly 2 million inauthentic votes every day. The company maintains a hybrid defense strategy, relying on AI for initial detection while utilizing community moderators, who managed over 52% of content removals between July and December 2025.

These measures target AI-focused marketing firms like ReachLLM and Profound, the latter of which reached a valuation exceeding $1 billion in February. Shanzila Ahmed, founder of ReachLLM, acknowledged that while some content is removed by Reddit's improved tools, her firm continues to produce content at regular intervals to secure AI citations for clients. Research from Cornell Tech has confirmed that strategically placed user-generated content can successfully manipulate the outputs of AI research tools.