AGI in 5 Years or 50?

Will we achieve artificial general intelligence in the next five years, or will it take fifty? This question isn't just philosophical anymore. It affects every startup decision, every investment thesis, and whether we're living through the most important technological transition in human history or just another hype cycle.

Leopold Aschenbrenner is one of the most compelling voices arguing we're closer than most people think. He's a former OpenAI researcher who wrote "Situational Awareness," a detailed manifesto predicting AGI by 2027. His argument is that we just need to keep scaling. The progress from GPT-4 to GPT-5 to whatever comes next follows predictable curves, and those curves point to human-level intelligence in the next few years.

Rodney Brooks is an MIT robotics professor and the founder of iRobot (yes, the Roomba company). More importantly, he's been right before about AI hype. He correctly predicted that self-driving cars and general robotics would take way longer than the optimists claimed. His position is that current LLMs are impressive but fundamentally limited. We'll need architectural breakthroughs, not just bigger models and more compute. AGI isn't around the corner, it's decades away, and the people saying otherwise don't understand how far we still have to go.

The implications for this debate are massive. If Leopold is right, the world changes in a few years. If Rodney is right, we're wasting billions on a scaling strategy that won't work.