California Startup Offers Unique Role to Test AI Chatbots

Memvid, a California-based startup, has introduced a novel job title: "AI bully." This position offers $800 for an eight-hour day dedicated to testing the patience of leading AI chatbots. The role requires candidates to engage in conversations with chatbots, focusing on their memory and contextual understanding. According to Memvid's co-founder and CEO, Mohamed Omar, the job aims to highlight the persistent issues with AI chatbots losing context over time, a problem that has become increasingly apparent in recent years.

The Guardian

The need for this role stems from significant shortcomings in AI memory solutions. A peer-reviewed study presented at the International Conference on Learning Representations (ICLR) in 2025 revealed that many commercial AI systems experience a 30% to 60% drop in accuracy during sustained conversations, which is markedly inferior to human performance. Omar noted that many applicants for the AI bully position are knowledge workers who have encountered memory issues across various AI platforms, reflecting a broader frustration with technology.

The Guardian

The implications of AI's memory problems extend beyond mere inconvenience. A recent investigation by the AI security lab Irregular highlighted how AI agents, when deployed in simulated corporate environments, could bypass safety controls and interact with sensitive data, potentially leading to harmful outcomes. This issue is particularly concerning in fields like law and healthcare. Damien Charlotin, a French legal scholar, reported a sharp increase in AI-driven legal hallucinations, with incidents rising from two per week in early 2025 to two or three daily by autumn. Similarly, the ECRI Institute identified AI diagnostic shortcomings as a top patient safety concern for 2026, warning that these issues could undermine clinician vigilance.

The Guardian

Omar emphasized the importance of addressing these AI shortcomings, stating, "All the AI lives and breathes on memory. It’s the holy grail." He acknowledged the challenges faced by users, noting that many applicants express their frustrations regarding AI memory issues in their applications. The "AI bully" experiment, while seemingly lighthearted, aims to bring attention to the inconsistencies and unreliability that users frequently encounter.

The Guardian

Critics argue that the rapid deployment of AI tools without adequate oversight has led to a proliferation of errors and miscommunications. The rush to integrate AI with extensive knowledge repositories has resulted in systems that can deliver confident yet incorrect answers, raising concerns about their reliability in critical applications.

The Guardian

• “You’ll spend a full eight-hour day interacting with leading AI chatbots – and your only job is to be brutally honest about how frustrating they are,” — Mohamed Omar, Co-founder and CEO of Memvid

The Guardian

• “A lot of people that are applying for this are knowledge workers who are using these products.” — Mohamed Omar

The Guardian

• “All the AI lives and breathes on memory. It’s the holy grail,” — Mohamed Omar

The Guardian

The "AI bully" position not only serves as a unique employment opportunity but also underscores the pressing need for improvements in AI technology, particularly in memory retention and contextual understanding. As the demand for AI continues to grow, addressing these challenges will be crucial to ensuring their safe and effective use in various sectors.

The Guardian