diff --git a/NEWS.md b/NEWS.md
index 1c415854d3..4e108fae47 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -43,7 +43,7 @@
       Item 2 has 0 rows but longest item has 3; filled with NA
     ```
 
-5. `%like%` on factors with a large number of levels is now faster, [#4748](https://github.com/Rdatatable/data.table/issues/4748). The example in the PR shows 2.37s reduced to 0.86s on a factor lengh 100 million containing 1 million unique 10-character strings. Thanks to @statquant for reporting, and @shrektan for implementing.
+5. `%like%` on factors with a large number of levels is now faster, [#4748](https://github.com/Rdatatable/data.table/issues/4748). The example in the PR shows 2.37s reduced to 0.86s on a factor length 100 million containing 1 million unique 10-character strings. Thanks to @statquant for reporting, and @shrektan for implementing.
 
 6. `keyby=` now accepts `TRUE`/`FALSE` together with `by=`, [#4307](https://github.com/Rdatatable/data.table/issues/4307). The primary motivation is benchmarking where `by=` vs `keyby=` is varied across a set of queries. Thanks to Jan Gorecki for the request and the PR.
 
@@ -131,6 +131,8 @@
 
 18. `as.data.table()` on `xts` objects containing a column named `x` would return an `index` of type plain `integer` rather than `POSIXct`, [#4897](https://github.com/Rdatatable/data.table/issues/4897). Thanks to Emil Sjørup for reporting, and Jan Gorecki for the PR.
 
+19. A fix to `as.Date(c("", ...))` in R 4.0.3, [17909](https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=17909), has been backported to `data.table::as.IDate()` so that it too now returns `NA` for the first item when it is blank, even in older versions of R back to 3.1.0, rather than the incorrect error `character string is not in a standard unambiguous format`, [#4676](https://github.com/Rdatatable/data.table/issues/4676). Thanks to Arun Srinivasan for reporting, and Michael Chirico both for the `data.table` PR and for submitting the patch to R that was accepted and included in R 4.0.3.
+
 ## NOTES
 
 1. New feature 29 in v1.12.4 (Oct 2019) introduced zero-copy coercion. Our thinking is that requiring you to get the type right in the case of `0` (type double) vs `0L` (type integer) is too inconvenient for you the user. So such coercions happen in `data.table` automatically without warning. Thanks to zero-copy coercion there is no speed penalty, even when calling `set()` many times in a loop, so there's no speed penalty to warn you about either. However, we believe that assigning a character value such as `"2"` into an integer column is more likely to be a user mistake that you would like to be warned about. The type difference (character vs integer) may be the only clue that you have selected the wrong column, or typed the wrong variable to be assigned to that column. For this reason we view character to numeric-like coercion differently and will warn about it. If it is correct, then the warning is intended to nudge you to wrap the RHS with `as.<type>()` so that it is clear to readers of your code that a coercion from character to that type is intended. For example :
diff --git a/R/IDateTime.R b/R/IDateTime.R
index 0bf7bf0fac..cfacfacbb2 100644
--- a/R/IDateTime.R
+++ b/R/IDateTime.R
@@ -7,6 +7,10 @@ as.IDate = function(x, ...) UseMethod("as.IDate")
 
 as.IDate.default = function(x, ..., tz = attr(x, "tzone", exact=TRUE)) {
   if (is.null(tz)) tz = "UTC"
+  if (is.character(x)) {
+    # 4676 mimics for back-compatibility a similar patch applied to as.Date.character in r79119
+    is.na(x) = !nzchar(x)
+  }
   as.IDate(as.Date(x, tz = tz, ...))
 }
 
diff --git a/inst/tests/tests.Rraw b/inst/tests/tests.Rraw
index 12a38ffae4..218de465da 100644
--- a/inst/tests/tests.Rraw
+++ b/inst/tests/tests.Rraw
@@ -17705,4 +17705,7 @@ test(2190.9, DT[1:2, a:=call('sum', 1)], error="type 'language' cannot be coerce
 DT = data.table(i1 = c(234L, 250L, 169L, 234L, 147L, 96L, 96L, 369L, 147L, 96L), i4 = c(79L, 113L, 270L, -121L, 113L, 113L, -121L, 179L, -228L, 113L), v = 0)
 test(2191, DT[1:5, sum(v), by=.(i5 = 1:5 %% 2L), verbose=TRUE], data.table(i5=1:0, V1=c(0,0)), output="gforce")
 
+# base::as.Date was error when first item blank, affecting as.IDate, #4676
+test(2192.1, as.IDate(c('', '2020-01-01')), structure(c(NA_integer_, 18262L), class=c("IDate","Date")))
+test(2192.2, as.IDate(c('2020-01-01', '')), structure(c(18262L, NA_integer_), class=c("IDate","Date")))