The "No Text Detected" error almost always means the file you uploaded is technically an image rather than a text-based document. This happens often with scanned intake forms, faxed referrals, and PDFs generated directly from a copier or imaging device. To your eyes the page looks like readable text, but under the hood there are no selectable characters for an AI assistant to parse. Without a text layer, even the best medical GPT has nothing to work with.
The fix is to run the document through optical character recognition (OCR) before uploading. OCR analyzes the image, identifies each character, and embeds a searchable text layer into the file. Several tools handle this well: Adobe Acrobat includes a built-in OCR function, the free NAPS2 utility offers OCR during scanning, and many modern office scanners and multifunction printers can generate searchable PDFs automatically when you enable the setting. After processing, try selecting text on the page with your cursor. If highlighting works, your document is ready for upload.
One important caution for health professionals handling protected health information: if you use Adobe Acrobat's OCR feature, make sure you are not relying on its cloud-based components. Any PHI processed through a third-party cloud service without a signed BAA introduces HIPAA risk. Stick with local, on-device OCR options, or use a scanner that performs the conversion before the file ever leaves your network. This keeps your workflow aligned with the same standards that make BastionGPT a trusted healthcare AI platform for medical documentation and clinical note-taking.
To streamline this for busy practices, we've built automatic OCR directly into our early release environment at https://ultra.bastiongpt.com. Sign in with your existing BastionGPT credentials and upload the same document there. The system will detect the image-based file, apply OCR server-side within our secure infrastructure, and return readable results without any manual preprocessing on your end. This feature is actively being refined and will roll out to all accounts by default in the coming weeks.
If you've tried OCR processing and the error persists, or if you run into trouble using the early release environment, reach out to our support team at [email protected]. We'll investigate the specific file and make sure nothing else is getting in the way of your documentation workflow.