Will any AI be shown to accurately simulate an amoeba or heavier on the molecular level before 2025-07-01?

Will @jsuarez pufferfish RL environment be used to calculate friston free energy OR complexity/accuracy/surprise by 2028

By 2026 will any RL agent with learned causal models of its environment achieve superhuman performance on >=10 Atari environments?

Will LLMs such as GPT4 be considered a solution to Moravec’s paradox by 2030?

Will a lab train a >=1e26 FLOP state space model before the end of 2025?

Will any model pass an "undergrad proofs exam" Turing test by 2027?

Will researchers extract a novel program from the weights of an LLM into a Procedural/OO programming language by 2026?

Will "Genetic Algorithm" come back before 2031?

Will certain contemporary publicly available GPT models be generally accepted as conscious at inference time by 2100?

Will Wolfram Research create a general AI based on replacement rules logic by 2030?