diff --git a/CLAUDE.md b/CLAUDE.md
index 653a60014..eec8c2f86 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -146,7 +146,7 @@ Version format: `major.minor.patch.dev` (e.g., `3.0.1.9021`)
 - Stable CRAN releases drop the dev suffix (e.g., `3.0.1`)
 - `NEWS.md` uses sections **New**, **Fixes**, **Updates** with GitHub issue references (`#NNN`)
 
-### Version bump required for every PR
+### Version and date bump required for every PR
 
 Before opening a pull request, always increment the four-digit dev counter by 1 in **both** of these files:
 
@@ -162,6 +162,12 @@ Before opening a pull request, always increment the four-digit dev counter by 1
 
 Read the current version from `DESCRIPTION`, add 1 to the last numeric component, and write the new version to both files in the same commit as the rest of the PR changes.
 
+Also bump the date to the current date in **`DESCRIPTION`**, where it's in the `Date:` field in ISO format:
+
+```
+Date: 2025-12-31
+```
+
 ## Internal State
 
 The package uses a private `AMR_env` environment (created in `aa_globals.R`) for caching expensive lookups (e.g., microorganism matching scores, breakpoint tables). This avoids re-computation within a session.
diff --git a/DESCRIPTION b/DESCRIPTION
index d96888093..79d28bcfe 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,5 +1,5 @@
 Package: AMR
-Version: 3.0.1.9022
+Version: 3.0.1.9023
 Date: 2026-03-03
 Title: Antimicrobial Resistance Data Analysis
 Description: Functions to simplify and standardise antimicrobial resistance (AMR)
diff --git a/NEWS.md b/NEWS.md
index 26195aa1b..fbbdc9da7 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,4 +1,4 @@
-# AMR 3.0.1.9022
+# AMR 3.0.1.9023
 
 ### New
 * Integration with the **tidymodels** framework to allow seamless use of SIR, MIC and disk data in modelling pipelines via `recipes`
@@ -18,6 +18,7 @@
   - `eucast_rules()` has become a wrapper around that function.
 
 ### Fixes
+* Fixed a bug in `as.ab()` where certain AB codes containing "PH" or "TH" (such as `ETH`, `MTH`, `PHE`, `PHN`, `STH`, `THA`, `THI1`) would incorrectly return `NA` when combined in a vector with any untranslatable value (#245)
 * Fixed a bug in `antibiogram()` for when no antimicrobials are set
 * Fixed a bug in `as.sir()` where for numeric input the arguments `S`,  `I`,  and `R` would not be considered (#244)
 * Fixed some foreign translations of antimicrobial drugs
diff --git a/R/ab.R b/R/ab.R
index b72705f44..58b9abaef 100755
--- a/R/ab.R
+++ b/R/ab.R
@@ -119,7 +119,14 @@ as.ab <- function(x, flag_multiple_results = TRUE, language = get_AMR_locale(),
   x[x %like_case% "^PENICILLIN" & x %unlike_case% "[ /+-]"] <- "benzylpenicillin"
   x_bak_clean <- x
   if (already_regex == FALSE) {
+    x_bak_clean_before_gen <- x_bak_clean
     x_bak_clean <- generalise_antibiotic_name(x_bak_clean)
+    # generalise_antibiotic_name() rewrites "PH"->"F" and "TH"->"T", which
+    # mangles short valid AB codes (e.g. "ETH"->"ET", "PHN"->"FN", "STH"->"ST")
+    # making them unrecognisable in the lookup. Restore any values that were
+    # already valid AB codes before generalisation (#245).
+    is_valid_ab_code <- x_bak_clean_before_gen %in% AMR_env$AB_lookup$ab
+    x_bak_clean[is_valid_ab_code] <- x_bak_clean_before_gen[is_valid_ab_code]
   }
 
   x <- unique(x_bak_clean) # this means that every x is in fact generalise_antibiotic_name(x)