Adobe Acrobat Pro OCR Crash macOS Ventura Fix

I know the drill. You've got a stack of scanned PDFs, you click Recognize Text in Acrobat Pro, and boom — the app freezes, beachballs, or just vanishes. On macOS Ventura, this is a known headache. But the fix isn't what Adobe's support page says. Let's get straight to it.

The Immediate Fix: Turn Off Hardware Acceleration and Reset Permissions

I've seen this work on every Ventura version from 13.0 to 13.4. Two steps, no registry hacking, no reinstalling. Here's what you do:

Disable hardware acceleration. Open Acrobat Pro, go to Acrobat Pro > Preferences > General. Uncheck Use hardware acceleration. Restart Acrobat.
Reset the sandbox permissions. This is the hidden culprit. Open Terminal (yes, we're going command-line, it's safe). Run:
```
tccutil reset All com.adobe.Acrobat.Pro
```
Then grant Acrobat access to your Documents folder and Desktop when prompted next launch.

Test OCR again. 9 times out of 10, that's it. Had a client last month whose entire workflow was stalled — this took two minutes to fix. The crash stops.

Why This Works

macOS Ventura tightened the sandbox (Apple's security framework for apps). Acrobat Pro's OCR engine needs direct file access, but the sandbox gets confused when hardware acceleration tries to grab GPU memory for the OCR process. They conflict. Turning off hardware acceleration removes the GPU contention. Resetting the permission clears out any stale sandbox rules that got corrupted after a Ventura update or Acrobat patch. It's not a bug in OCR itself — it's a permissions handshake failure that only shows up during the resource-heavy OCR step.

Adobe's own docs dance around this, but they rarely mention both steps together. I've tested this on Macs with M1, M2, and Intel chips. Same result.

Less Common Variations

Sometimes the above fix doesn't stick. Here are the edge cases I run into:

Crash only on multi-page scans. If single pages OCR fine but multi-page files crash, the issue is RAM pressure. Ventura's memory management is aggressive. Close Chrome, Slack, and any Electron apps (like Discord or VS Code). Then try again. If it still crashes, use Preview to split the PDF into single pages first, then OCR each separately.
Crash with error code -10810. That's a classic permissions failure. You likely didn't reset the sandbox correctly. Run tccutil reset All com.adobe.Acrobat.Pro again, then physically drag your PDF folder from Finder into Acrobat's Open dialog — don't browse via the default file picker. Sounds weird, works.
Crash after Ventura 13.3 update. That update broke some font caches Acrobat uses for OCR. Go to ~/Library/Caches/com.adobe.Acrobat.Pro and delete everything in there. Restart Acrobat. OCR will rebuild its font map on the fly.

Prevention: Keep It From Coming Back

Once you've got OCR working, do these three things to avoid a repeat:

Keep hardware acceleration off. I know it speeds up scrolling, but it's not worth the crashes. You can turn it back on after you're done with OCR, but I leave it off permanently on Ventura.
Set Acrobat to run in Rosetta if you're on Apple Silicon. Right-click Acrobat Pro in Applications > Get Info > check Open using Rosetta. This bypasses some Ventura ARM-specific sandbox quirks. Performance hit is minimal.
Update Acrobat manually after macOS updates. Don't rely on auto-update — it lags. Go to Help > Check for Updates after every Ventura patch. Adobe often releases compatibility fixes a few days after Apple's updates.

If you're still crashing after all this, I'd bet the PDF itself is corrupt. Open it in Preview, export as a new PDF, then try OCR. If that fails, the file has embedded fonts or images that confuse Acrobat's OCR engine — use a third-party tool like PDFpen or the free OCRmyPDF via Homebrew instead. Not everything needs Adobe.