Imagine a 17-year-old Nigerian teenager declaring he’s going to build a rival to ChatGPT from scratch. Sounds crazy, right? Meet Okechukwu Nwaozor, the founder of OkeyMeta, who’s doing exactly that. And this is the part most people miss: he’s not just dreaming—he’s already built a working AI model that’s publicly available for testing. But here’s where it gets controversial: can a self-taught developer with limited resources truly challenge a Silicon Valley giant like OpenAI? Let’s dive in.
When I first heard Okechukwu’s story, I couldn’t help but chuckle. A first-time founder, fresh out of secondary school, claiming to take on a $500 billion company? It seemed far-fetched. Especially when he mentioned raising just ₦2.7 million—a fraction of OpenAI’s $40 billion funding. But during our call, Okechukwu’s determination was undeniable. He’s no stranger to skepticism; when he announced his project on Facebook, the comments ranged from doubt to outright mockery. Yet, he pressed on.
Here’s the kicker: OkeyMeta’s chatbot, OkeyAI, isn’t just a concept—it’s a functional product. When I tested it, I was skeptical it was truly built from scratch. But OkeyAI confidently assured me it was: ‘I am OkeyAI 4.0 DeepCognition, created by OkeyMeta Ltd, right here in Nigeria.’ It even credited its team, including Okechukwu and his co-founders. This wasn’t just a ChatGPT clone; it was something entirely homegrown.
So, how did a 17-year-old decide to build a large language model (LLM) from scratch? Surprisingly, it wasn’t the hype around ChatGPT that inspired him. Instead, it was his fascination with Google’s search algorithm years earlier. ‘I kept wondering how Google gives you so many relevant results instantly,’ he recalls. This curiosity led him down a rabbit hole of AI, and by 2022, he was gathering data to train his first model. He even designed his own Transformer architecture, SSAILM, which he claims makes OkeyMeta the first truly African-origin LLM.
But here’s the controversial part: Okechukwu insists his model is built from the ground up, unlike competitors like Awarri’s N-Atlas, which he says is based on Meta’s Llama 3. ‘We discovered N-Atlas wasn’t created from scratch,’ he asserts. This distinction matters to him, even if his model is smaller than global giants. Is he right to draw this line? Or is fine-tuning an existing model just as valid? Let’s discuss in the comments.
Despite his progress, Okechukwu faces massive challenges. OkeyMeta’s user base is growing—nearly 1,000 users and 8,000 developers—but visibility remains low. ‘Our posts only get one or two likes,’ he admits. And while the platform offers unique features, like unlimited memory (compared to ChatGPT’s 1,000-token cap) and reduced hallucination, scaling is expensive. Renting GPUs from Google costs $100 per month, and his ₦2.7 million funding won’t last forever. He needs capital, mentorship, and infrastructure to survive.
Yet, Okechukwu’s vision is bold: he sees OkeyMeta becoming a global AI company, proving the doubters wrong. Is this just a teenager’s dream, or the beginning of something groundbreaking? What do you think? Could OkeyMeta truly challenge the giants, or is it a long shot? Share your thoughts below—let’s spark a conversation!
