≈

The passage around this formula

The clearest empirical trend in this area comes from METR, which measured how the length of software tasks that leading agents can complete at even odds has changed over time, expressed as a time horizon $T_{50}$ - the length of task, in expert-human time, that an agent completes successfully half the time. Kwa and colleagues report that this time horizon has been doubling roughly every seven months since 2019, which can be written as an empirical fit $T_{50}(t) \approx T_{50}(t_0)\cdot 2^{(t-t_0)/\tau}, \qquad \tau \approx 7\text{ months}$. and they further find that this trend is driven substantially by improved reliability and error recovery over long horizons, not only by raw single-step capability [ 4 ] . This is a genuinely informative fact and it is also,…

Read this part in the article →