Will flagship LLMs begin integrating advertisements into their responses before 2026?

Will an LLM that someone is trying to shut down stop or avoid that in some way before 2026?

What will Manifolders mostly use LLMs for, by EOY 2025?

Will an LLM do a task that the user hadn't requested in a notable way before 2026?

Will RL work for LLMs "spill over" to the rest of RL by 2026?

Will one of the major LLMs be capable of continual lifelong learning (learning from inference runs) by EOY 2025?

What will be true of Anthropic's best LLM by EOY 2025?

Will there be major breakthrough in LLM Continual Learning before 2026?

Will there be a major bioattack by EOY 2025 where an LLM provided relevant information to the attacker(s)?

Will we get a new LLM paradigm by EOY?