It's 2025 and most content is still written for humans instead of LLMs. 99.9% of attention is about to be LLM attention, not human attention.
E.g. 99% of libraries still have docs that basically render to some pretty .html static pages assuming a human will click through them. In 2025 the docs should be a single your_project.md text file that is intended to go into the context window of an LLM.
Repeat for everything.
Sriram Krishnan will face backlash in USA because American public don’t want talented Indians to fix things for their country
Sriram Krishnan will struggle in India because Indian politicians don’t want talented Indians to fix things for their country
If you haven’t figured out the joke yet, *any* fixed benchmark will fall rapidly the instant it becomes an optimization target for the top labs.
Correspondingly, no fixed benchmark is AGI. AGI is the ability to generalize to an adversarially chosen new benchmark.
the difference between o1 and sonnet-3.5 is like the difference between a super car and a sports car. you can’t notice the difference on city roads (in fact the sports car is better in many cases), but on tracks you most definitely do notice it