[Will Brown] Will Transformative AI look like an abundance of specialized models, in August 2026?

By 2030, will large language models still be at the peak of AI? [DRAFT]

By the end of 2026, will we have transparency into any useful internal pattern within a Large Language Model whose semantics would have been unfamiliar to AI and cognitive science in 2006?

Will there be a significant advancement in frontier AI model architecture by end of year 2026?

Will we reverse-engineer a language model into an interpretable (python) program by 2027?

Will models be able to do the work of an AI researcher/engineer before 2027?

Will AI models solve at least 2 FrontierMath Open Problems before 2027?

Will AI wipe out Manifold by 2030?

Will AI be Recursively Self Improving by mid 2026?

Benchmark Gap #1: Once we have a language model that achieves expert human performance on all *current* major NLP benchmarks, how many years will it be before we have an AI with human-level language skills?