macOS 26.2
Juli Clover (release notes, security, enterprise, developer, full installer, IPSW):
macOS Tahoe 26.2 includes Edge Light, a feature that illuminates your face with soft light when you’re on a video call in a room with poor lighting. The update also adds alarms for the Reminders app, new podcast features, updated AirDrop settings, and more.
When was the last time Apple released new OS versions on a Friday?
I think macOS 26.2 once again erased my Local Network permissions.
Previously:
Update (2025-12-15): Nick Heer:
I have found the version of Safari in this build of MacOS 26.2 is noticeably buggy. It sometimes stops letting me scroll a webpage and, in rare cases, I have found the browser wholly crashes when closing tabs.
Finder in tahoe 26.2
Bizare choice to show blurred through hidden left column. Corner radius of left menu doesn’t match blur container - so there is a tiny area where it isn’t blurred!
I have also confirmed, as I suspected from the lack of change in the RichText.mdimporter, that the ‘LG bug’ in Spotlight remains, and still hasn’t been fixed.
Apple (Hacker News):
Enables low-latency communication between Thunderbolt 5 hosts for use cases including distributed AI inference using MLX.
macOS 26.2 is showing a gray background for a very noticeable second on login before displaying my desktop background image (which is the default).
As part of macOS 10.14 Mojave, Apple introduced a number of privacy controls for user data. At the same time, Apple also introduced device management options to allow authorized applications to access data protected by those privacy controls. These permissions are referred to collectively as Privacy Preferences Policy Control (PPPC) and are deployed via management profiles from an MDM server. However, up until macOS Tahoe 26.2, there was no way to see in the Privacy & Security section of System Settings which applications had which permissions granted via PPPC management profiles.
I’d been holding back on switching to Tahoe 26-26.1 to let the bugs get worked out, but it seems this major release I’ve seen a lot more complaints, including here. Just curious if you’ve seen improvements with 26.2.
Previously:
Update (2025-12-16): Howard Oakley:
Several of those who have already updated to macOS Tahoe 26.2 have remarked how much larger their download was than the 3.78 GB expected for Apple silicon Macs, with some reporting over 10 GB. Here I ponder how that could happen.
[…]
What is puzzling about the 26.2 update is that it wasn’t preceded by a Background Security Improvement (BSI) or Rapid Security Response (RSR). Two of the top security vulnerabilities fixed in 26.2 (and in the Safari updates for 15.7.3 and 14.8.3) are both in WebKit, which is supplied in the Safari cryptex.
Update (2025-12-17): David Deller:
Apple responded to one of my MusicKit feedbacks (slow playlist loading, FB18157502) and based on my testing, it appears to have been fixed on macOS 26.2. Still broken on 15.7.3 (current latest Sequoia). Even so, seems like progress! Happy to see it.
Update (2025-12-26): Rob Jonson:
We’re on Tahoe 26.2 and I can’t resize finder columns because the scrollbar completely covers them.
I guess someone at Apple must enjoy looking at toolbars like this from time to time. I don’t.
The Gigabit Ethernet port on my Thunderbolt hub for my M1 MacBook Pro stopped working. Thought it was a hardware problem because I restarted, but the hub’s connection light stayed dead. Tried swapping ports, then cables, then even the hub.
Hadn’t thought to delete the system setting altogether. Restarted to flush the NVRAM one more time. Once it restarted, the Mac recognized the Ethernet port and the hub showed a connection light.
Worth noting that this is the second odd software setting problem masquerading as a hardware issue I’ve had since upgrading to macOS 26 “Tahoe.” Can’t remember the details of the first at the moment but it took me a while to isolate and correct, regardless.
Update (2025-12-30): Atom:
After the update, my Mac no longer maintains sleep properly. When it goes into standby, it wakes up briefly every ~15–20 seconds, then goes back to sleep, repeating this cycle indefinitely. This makes proper standby essentially impossible.
Update (2026-01-07): Juli Clover:
The M4 iPad Pro models, M3 iPad Air models, A17 Pro iPad mini, M2 to M5 MacBook Pro models, M2, M3, and M4 MacBook Air models, and other Wi-Fi 6E Macs and iPads now support 160MHz maximum channel bandwidth when connected to 5GHz Wi-Fi networks, the same theoretical maximum throughput supported by 6GHz networks. Previously, these devices were limited to 80MHz.
7 Comments RSS · Twitter · Mastodon
When was the last time Apple released new OS versions on a Friday?
Seems like tempting fate and/or Murphy, but how many more weeks until they're all on a skeleton crew for Christmas? One week? One and a half, tops?
Realistically it wasn't likely to get better in a week. I get the impression they are running on fumes from this whole year and just want to get to holiday break. Clearly this was a late release and they wanted to get these fixes and features out the door.
I sincerely feel for what must be going on for the rank and file developers right now. I hope they do get a much deserved holiday and come back to a much improved new year, with a fresh opportunity to start moving in the right direction.
As a result of having all the OSes in lockstep, they were sort of forced because some iPhone App Store features need to be available in some places (Japan?) by the end of the year.
Anyone can verify this claim "They broke drag&drop of JPEGs into Messages. Results in a blank white icon and can't be send. CMD+C and CMD+V still works." ?
There’s an update on Atom’s sleep issue. He writes: «I’ve now connected the mouse using the Logitech Bolt receiver instead […], and the problem is completely gone.»
>Restarted to flush the NVRAM one more time. Once it restarted, the Mac recognized the Ethernet port and the hub showed a connection light.
I've never _quite_ understood this, dating back to the 1990s (back then, PRAM). It always seemed voodoo-y.
Is there some device tree cache in the NVRAM, with subtle cache invalidation bugs? Does "flush the NVRAM" even actually flush an actual NVRAM chip any more?
