Episode notes
A Chinese model trained in China… gives American-style answers? Yep. In today’s episode, we break down why DeepSeek still thinks in English, what this means for AI bias, and how even Meta is banning ChatGPT from WhatsApp. Plus, we look at Google’s rumored Gemini 3.0 Pro models, OpenAI’s Atlas browser, and the $30K AI embryo screener that’s now open-source.
We’ll talk about:
- What happens when you ask ChatGPT, Claude, or DeepSeek in Arabic, Chinese, or Hindi
- Why all 3 AIs give shockingly similar answers and why that's not an accident
- The Gemini 3.0 Pro leak and the new AI browser everyone’s testing
- A new AI tool that screens embryo DNA for disease but only if you’ve got $30K
Keywords: ChatGPT, Claude 4.5, DeepSeek, Gemini 3.0 Pro, OpenAI Atlas, AI embryo s ...
Keywords
ChatGPTDeepSeekMeta AIopen-source AIClaude Sonnet 4.5AI biasLLM languagesAI embryo screeningOpenAI AtlasGemini 3.0 Pro