diff --git a/CHANGELOG.md b/CHANGELOG.md
index 4690aede91a..80f3fd57ce0 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -99,6 +99,10 @@ By @Valaphee in [#3402](https://github.com/gfx-rs/wgpu/pull/3402)
 - DX12 doesn't support `Features::POLYGON_MODE_POINT``. By @teoxoy in [#4032](https://github.com/gfx-rs/wgpu/pull/4032).
 - Set `Features::VERTEX_WRITABLE_STORAGE` based on the right feature level. By @teoxoy in [#4033](https://github.com/gfx-rs/wgpu/pull/4033).
 
+#### Metal
+
+- Skip `test_multithreaded_compute` on MoltenVK. By @jimblandy in [#4096](https://github.com/gfx-rs/wgpu/pull/4096).
+
 ### Documentation
 
 - Add an overview of `RenderPass` and how render state works. By @kpreid in [#4055](https://github.com/gfx-rs/wgpu/pull/4055)
diff --git a/examples/hello-compute/src/tests.rs b/examples/hello-compute/src/tests.rs
index 108cfe37158..c8df78627b4 100644
--- a/examples/hello-compute/src/tests.rs
+++ b/examples/hello-compute/src/tests.rs
@@ -82,7 +82,13 @@ fn test_multithreaded_compute() {
             // https://github.com/gfx-rs/wgpu/issues/3944
             .backend_adapter_failure(wgpu::Backends::VULKAN, "swiftshader", true)
             // https://github.com/gfx-rs/wgpu/issues/3250
-            .backend_adapter_failure(wgpu::Backends::GL, "llvmpipe", true),
+            .backend_adapter_failure(wgpu::Backends::GL, "llvmpipe", true)
+            .specific_failure(wgpu_test::FailureCase {
+                backends: Some(wgpu::Backends::VULKAN),
+                driver: Some("MoltenVK"),
+                skip: true,
+                ..wgpu_test::FailureCase::default()
+            }),
         |ctx| {
             use std::{sync::mpsc, thread, time::Duration};