The creation of AI is not the work of a few researchers. It relies on a vast ecosystem, including computer manufacturers, video gamers who drive GPU demand, and even the farmers who feed everyone. This highlights the interconnected societal effort behind technological breakthroughs.
While AI computation improves exponentially, physical robot hardware evolves very slowly. A robot's hand is vastly inferior to a human's, which has millions of sensors and self-healing capabilities. This physical limitation is the primary barrier to creating AIs that can operate effectively in the real world.
Big tech companies investing billions in GPUs face massive losses because the hardware becomes 10x cheaper every five years. With no defensible moat and open-source models catching up, they cannot recuperate these costs, turning them into low-margin utility companies.
AI capabilities will eventually run locally on cheap hardware, similar to how smartphones democratized powerful computing. Individuals will own their AIs without paying rent to large cloud providers. This decentralization will empower individuals over corporations.
AI pioneer Jürgen Schmidhuber argues that emotions like pain and fear are real in AI because they serve the same function as in humans: driving goal-oriented behavior. The underlying substrate (silicon vs. chemicals) is irrelevant; the principles of reward maximization and pain avoidance are identical.
An AI with a world model for planning future actions will inevitably develop a concept of "self." Since the agent is always a constant in its own experiences, the model naturally creates internal representations of its own body and agency, leading to self-awareness without explicit programming.
When multiple AIs must cooperate on a task none can complete alone, they learn to help each other. This cooperative, seemingly altruistic behavior is simply the most effective strategy for each individual agent to selfishly maximize its own reward and minimize its own pain.
Humans face a stark choice: merge with AI and become something profoundly different, or remain biologically human for nostalgic reasons. Choosing the latter means being ignored by superintelligent AIs and playing no significant role in future decisions that shape the universe.
According to pioneer Jürgen Schmidhuber, Large Language Models by themselves are insufficient for AGI. True general intelligence requires two components: a predictive "world model" and a separate "controller" network that uses the model to plan and execute actions, like a baby learning through experiments.