What's Actually New in Unicode 18.0

The Unicode Consortium published version 18.0 of the Unicode Standard on Sept. 16, 2026, adding 13,007 characters and bringing the total encoded across the standard to 172,808. Most of the new total comes from a single addition: the Seal script, the pictographic ancestor of the characters cut into Qin dynasty bronze edicts, which alone contributes 11,328 characters.

What the 13,007 new characters actually are

Unicode 18.0 adds three new scripts. Seal, also called Small Seal, covers the archaic Chinese script used on stone and bronze seals for roughly two thousand years before it fossilized into a decorative form. Jurchen is a medieval script used by the Jurchen people, ancestors of the Manchu, and had no dedicated Unicode block before this release. Proto-Cuneiform is encoded for numerals only, the counting marks that predate full cuneiform writing in ancient Mesopotamia. The release also adds seven new blocks, including a Bengali Supplement, Archaic Cuneiform Numerals, Musical Symbols Supplement, and Miscellaneous Symbols and Arrows Extended, plus 27 new variation sequences for CJK strokes and math operators.

New in Unicode 18.0	Detail
Total new characters	13,007
Total characters encoded	172,808
New scripts	Seal (11,328 characters), Jurchen, Proto-Cuneiform (numerals)
New emoji	9 base, 19 counting thumb-sign skin tones
New currency symbols	Rufiyaa (U+20C2), UAE Dirham (U+20C3), Omani Rial (U+20C4)

The nine new emoji, and when they will actually show up

According to Emojipedia's rollout tracking, the base set is Cracking Face, Leftward Thumb Sign, Rightward Thumb Sign, Monarch Butterfly, Pickle, Lighthouse, Meteor, Eraser and Net With Handle. Counting skin-tone variants for the two thumb signs brings the total to 19 additions. Getting a character into the standard is not the same as getting it onto a keyboard: Google is previewing draft designs through its Noto Emoji font pages this month, but WhatsApp and Samsung are not expected to add support until the first quarter of 2027, Apple is targeting its iOS 28 update in March or April 2027, and Microsoft is not expected to ship support in Windows 11 until late summer or early fall of 2027.

Why encoding a character doesn't make it render

A Unicode code point is an entry in a table, not a drawing. Type U+1FADD, the Pickle emoji's code point, into a text field today and most devices will show a blank box or a placeholder glyph, because no font installed on the system yet maps that number to a picture. The same gap applies to Seal and Jurchen: a developer can store the text, search it and sort it correctly the moment a library adopts Unicode 18.0's data tables, but nothing will actually draw the glyphs until a font ships that covers those blocks, which for two scripts this obscure could take considerably longer than an emoji update that every platform has a commercial reason to prioritize.

The release also synchronizes five Unicode Technical Standards to version 18.0: the Collation Algorithm, which governs sort order; Security Mechanisms; IDNA Compatibility Processing; Emoji; and Link Detection. The Security Mechanisms and IDNA updates matter specifically for anyone running a domain registrar, a browser or a spam filter, because every new script adds characters that can look confusingly similar to existing ones in a URL, the same class of homograph problem that domain-confusables lists exist to catch. Until registries and browsers pull in the updated confusables data for Seal, Jurchen and Proto-Cuneiform, those new code points sit in a gap where old anti-spoofing tables don't yet know they exist.

What ships alongside it, and what doesn't

Unicode 18.0 arrived the same week as two other developer-platform releases with their own version-specific catches: Swift 6.4 shipped Sept. 15 with faster WebAssembly bridging, and .NET 11 reached release candidate on its way to a Nov. 10 launch. None of the three projects has published a date for when its own string libraries will report Unicode 18.0's script and confusables data rather than the 17.0 tables most runtimes still ship. The Unicode Consortium's own code charts for the new release run to 179 megabytes, which is itself a rough measure of how much rendering, sorting and security software downstream now has to catch up on before any of version 18.0 is more than an entry in a table.