Serial entrepreneur John Pasmore founded Latimer AI after witnessing the pervasive bias in existing large language models (LLMs). He observed how casually racist outputs from AI tools were being dismissed, prompting him to create a platform designed to deliver more accurate and inclusive results, especially for Black and Brown communities. This isn’t simply about adding features; it’s about addressing a fundamental flaw in how these technologies are trained and deployed.
The Problem with Current AI Systems
The core issue lies in the data used to train LLMs. These models learn from patterns in the information they’re fed, and if that data is skewed towards overrepresented groups (historically, overwhelmingly white and male), the AI will reflect those biases. For example, when prompted for ideal leadership profiles, an LLM will often default to describing a man, perpetuating systemic inequalities. While mitigation efforts exist, the root of the problem is historical imbalance. This isn’t just unfair; it actively reinforces harm by creating tools that normalize biased outcomes.
How Latimer AI Differs
Latimer AI takes a unique approach: it combines retrieval-augmented generation (RAG) with its own curated database, alongside access to leading LLMs like ChatGPT, Claude, and Gemini. This means the platform doesn’t just rely on pre-trained biases. Instead, it actively searches for information from external sources and its own database to cross-reference and refine responses. The result is more contextually relevant, culturally fluent, and inclusive outputs.
Pasmore emphasizes that this isn’t about competing with OpenAI; it’s about ensuring that future generations aren’t conditioned to accept AI-generated narratives as absolute truth. The platform provides tiered access, including a free plan with limited interactions, and API integration for developers. Pricing for API usage starts at under 10 cents per 1,000 tokens, making it accessible for various applications.
The Human Element
The key distinction between Latimer AI and other LLMs is its attempt to inject empathy and nuance into responses. When asked about environmental racism, ChatGPT provides a clinical definition, while Latimer AI offers concrete examples and acknowledges the human toll. Pasmore’s goal is to move beyond sterile definitions and ensure the technology reflects real-world experiences.
“I want them to ask better questions,” Pasmore explains. “The whole point is to make curiosity a muscle again.”
This focus on critical thinking is paramount. AI tools have a way of feeling authoritative, and their answers can easily be mistaken for objective truth. But even advanced models like Anthropic’s Claude aren’t neutral; they’re a reflection of the biases of those who build them. Latimer AI aims to counteract this by demanding accuracy and challenging the dominant narratives embedded within these systems.
The Bigger Picture
Latimer AI isn’t just a technical solution; it’s a response to a deeper question: who controls the narrative in an age of increasingly powerful AI? Pasmore sees this as a historical corrective, a way to create a record that can’t be rewritten by biased algorithms. His platform is designed not just to generate answers but to encourage users to ask better questions, fostering a more critical and informed approach to technology.
Ultimately, the success of Latimer AI hinges on whether we can demand more from AI than just efficiency. If accuracy and inclusivity aren’t treated as technical limitations but as conscious choices, there’s a real opportunity to build a future where artificial intelligence serves everyone, not just the historically privileged.
