The UN set 2020 to bring the least-developed countries universal internet. In the deadline year 24% of their people were online — and across the 37 with 2023 data, the median country is still majority-offline

When the world's governments adopted the Sustainable Development Goals in 2015, one target carried a hard date and named the poorest countries directly. Target 9.c pledged to 'significantly increase access to information and communications technology and strive to provide universal and affordable access to the Internet in least developed countries by 2020.' We pulled the World Bank series that scores the world on internet use — IT.NET.USER.ZS, the share of individuals who used the internet, sourced from the International Telecommunication Union — and read its consistent 'Least developed countries: UN classification' aggregate across 2015-2024. These are ITU estimates, and internet use is a use-based proxy for the target's access-and-affordability language, distinct from the official 9.c.1 indicator, which measures mobile-network coverage. In 2015 the LDC group sat at 10.7%. In 2020, the deadline year, it reached 24.1% — roughly one person in four. By 2024 it was 31.4%. For scale: high-income economies were at 88.4% in the 2020 deadline year and 93.0% in 2024; the world as a whole ran 60.1% in 2020 and 71.2% in 2024. The average person in an LDC in 2024 (31.4%) was still less likely to be online than the average person worldwide had been in 2015 (39.9%). The gap between the LDC and high-income averages was 67.6 points in 2015 and 61.6 points in 2024 — it narrowed about six points across the decade. Country by country, we kept every UN-listed LDC with a 2023 value: 37 of the 44 (Eritrea, Myanmar, Sudan, South Sudan, Timor-Leste, Tuvalu and Yemen have no 2023 figure in the series). The median of the 37 is 28.8%. Thirty-two of the 37 have under half their people online; 13 sit at or below 25%. Use runs from Burundi at 7.2% to Kiribati at 88.0%. None of the 37 has reached universal, four years past the deadline.

The following is analysis, not fact. The single vantage most desks take is the growth story, and it is real: LDC internet use nearly tripled over the decade, from 10.7% in 2015 to 31.4% in 2024. Read against the pledge, that framing buries the point. Target 9.c was the one SDG connectivity commitment with a firm date, and it singled out exactly these 44 countries. On the target-setter's own numbers the deadline came and went with the average LDC resident offline, and four years later the median reporting LDC is still majority-offline. Fast growth from a floor this low does not close the distance. The high-income world was already near saturation when the clock started, so even fast growth in the LDCs against a stalling top end shaved only about six points off a 68-point gap in ten years. Internet use tracks income closely across these groups — the World Bank's low-income aggregate was 22% in 2023 against the LDCs' 30% — but that is an association in the data, not a mechanism these figures isolate; we joined no price, infrastructure or literacy variable, so read it as a pattern the record is consistent with, not a cause it proves. ^sspeculative

The strongest counter is that we are scoring the wrong clause. Target 9.c speaks to 'access,' and it is formally tracked on indicator 9.c.1 — the share of people living under a mobile network — a coverage measure the ITU reports runs well ahead of actual use across the LDCs. On that clause the poorest countries are far closer to universal than these use figures suggest. Measured on use, as we do here, the picture is stricter and, we would argue, closer to what 'access to the Internet' means to a household, but it is not the letter of the tracking indicator. ^mmodeled Two more caveats bound the read. These are ITU estimates, weakest where household surveys are old, and seven LDCs have no 2023 value at all — so the 37-country panel is incomplete, not the full 44 (the missing seven could sit above or below the median). The panel uses the current UN-LDC membership (44, as of 2025, per UN CDP/OHRLLS): Bhutan and São Tomé — LDCs during the 2023 data year but graduated December 2023 and December 2024 — are excluded as current non-LDCs, and both sit above 50% (90% and 59%), so leaving them out does not change the 32-under-50 count. And the direction of travel is up, fast: the ITU's own Facts and Figures 2024 puts LDC use near 35%, a higher figure than the World Bank's 31.4% but one that still puts about a third online, and both say the gap is closing — just not by 2020, and not soon.

**View from a telecom regulator in an LDC.** The tower on the hill is not the target. Coverage maps have looked good for years; what the 2020 pledge actually asked for was universal *and affordable* access, and a signal a household cannot afford to use is not access in any sense a family would recognize. Scored on who is actually online, the honest verdict is the one that stings — the deadline passed, and most of our people are still on the other side of it.