33%. Not yet.

How far software has moved toward solved since ChatGPT launched in November 2022, based on 4 questions and the signals behind them.

The line is recalculated from today's data rather than recorded at the time, and it can jump when a new signal joins.

The goals behind each bar are our judgment. Moving them all easier or harder gives 29% to 39%. Why each goal.

Most bars use a log scale, so they track multiples rather than the share of a goal. Growing 32 times fills half the bar for a goal of 1,000 times.

How it works

\begin{aligned} \text{Score} &= \frac{1}{n}\sum_{i=1}^{n} \text{Question}_i \\[6pt] \text{Question}_i &= \frac{\sum_{j=1}^{m} w_j\,\text{Signal}_j}{\sum_{j=1}^{m} w_j} \end{aligned}

Overall score (out of 100) · 2023-01 – 2026-11

Inspect a point 2026-11 — Overall score (out of 100): 33

View data table

Period / observation	Overall score (out of 100)
2023-01	1
2023-02	2
2023-03	5
2023-04	6
2023-05	6
2023-06	7
2023-07	7
2023-08	7
2023-09	7
2023-10	6
2023-11	6
2023-12	8
2024-01	11
2024-02	8
2024-03	11
2024-04	11
2024-05	12
2024-06	14
2024-07	14
2024-08	15
2024-09	15
2024-10	15
2024-11	15
2024-12	17
2025-01	16
2025-02	20
2025-03	21
2025-04	22
2025-05	23
2025-06	26
2025-07	25
2025-08	26
2025-09	26
2025-10	26
2025-11	27
2025-12	27
2026-01	27
2026-02	29
2026-03	32
2026-04	33
2026-05	32
2026-06	32
2026-07	33
2026-08	33
2026-09	34
2026-10	33
2026-11	33

Can AI do the work?

56%

Is it being used to build?

41%

Are fewer people needed?

28%

Is software cheaper and more reliable?

Money is compared with the economy and inflation, and failures with how much code is pushed. A rise scores 0%, which means no sign of change yet. AI agents may be inflating the number of pushes, which would make failures look rarer than they are.