diff --git a/docs/changelogs/v0.36.md b/docs/changelogs/v0.36.md
index 0bf39522cb4..2a5234477ac 100644
--- a/docs/changelogs/v0.36.md
+++ b/docs/changelogs/v0.36.md
@@ -40,7 +40,7 @@ See [`HTTPRetrieval`](https://github.com/ipfs/kubo/blob/master/docs/config.md#ht
 
 The Bitswap client now supports broadcast reduction logic, which is enabled by default. This feature significantly reduces the number of broadcast messages sent to peers, resulting in lower bandwidth usage during load spikes.
 
-The overall logic works by sending to non-local peers only if those peers have previously replied that they want data blocks. To minimize impact on existing workloads, by default, broadcasts are still always sent to peers on the local network, or the ones defined in `Peering.Peers`.
+The overall logic works by sending to non-local peers only if those peers have previously replied that they have data blocks. To minimize impact on existing workloads, by default, broadcasts are still always sent to peers on the local network, or the ones defined in `Peering.Peers`.
 
 At Shipyard, we conducted A/B testing on our internal Kubo staging gateway with organic CID requests to `ipfs.io`. While these results may not exactly match your specific workload, the benefits proved significant enough to make this feature default. Here are the key findings: