While more data and compute yield linear improvements, true step-function advances in AI come from unpredictable algorithmic breakthroughs like Transformers. These creative ideas are the most difficult to innovate on and represent the highest-leverage, yet riskiest, area for investment and research focus.

Google's TurboQuant algorithm enables near-lossless context compression, drastically reducing memory usage and inference costs. This breakthrough could democratize powerful AI by making it far cheaper and faster to run, much like the fictional 'middle-out' compression from the show 'Silicon Valley' was a game-changer.

China is gaining an efficiency edge in AI by using "distillation"—training smaller, cheaper models from larger ones. This "train the trainer" approach is much faster and challenges the capital-intensive US strategy, highlighting how inefficient and "bloated" current Western foundational models are.

Models like Gemini 3 Flash show a key trend: making frontier intelligence faster, cheaper, and more efficient. The trajectory is for today's state-of-the-art models to become 10x cheaper within a year, enabling widespread, low-latency, and on-device deployment.

As AI demand outstrips Earth's power supply, the industry is pursuing two strategies. Elon Musk is escaping the constraint by moving data centers to space. Everyone else must innovate on compute efficiency through new chip designs and model architectures to achieve 70-100x gains per token.

Chinese AI models like Kimi achieve dramatic cost reductions through specific architectural choices, not just scale. Using a "mixture of experts" design, they only utilize a fraction of their total parameters for any given task, making them far more efficient to run than the "dense" models common in the West.

When multiple models can solve a task reliably ('benchmark saturation'), the strategic goal is no longer to find the most intelligent model. Instead, it becomes an optimization problem: select the smallest, cheapest, and fastest model that still meets the performance bar, creating a major competitive advantage in inference.

OpenAI is designing its custom chip for flexibility, not just raw performance on current models. The team learned that major 100x efficiency gains come from evolving algorithms (e.g., dense to sparse transformers), so the hardware must be adaptable to these future architectural changes.

Arvind Krishna forecasts a 1000x drop in AI compute costs over five years. This won't just come from better chips (a 10x gain). It will be compounded by new processor architectures (another 10x) and major software optimizations like model compression and quantization (a final 10x).