How to Turn Scanned Documents into Searchable Files with AI

You've probably got a shoebox of old photos, a stack of handwritten notes, or a folder of printed forms sitting in the cupboard. What if you could search inside them the same way you search your emails? With a free AI tool and your phone camera, you can turn those paper piles into searchable, text-ready files in just a few minutes.

What you'll need

Before you start, gather these:

• A phone with a working camera, or a home scanner (most printer/scanner combos do the job)
• A free ChatGPT account — sign up on the ChatGPT website or through the ChatGPT app on iOS or Android using your email address
• A document, photo, or page of notes you want to digitise
• A notes app on your phone or computer (Google Docs, Microsoft Word, or even the basic Notes app) to paste the extracted text into

The free version of ChatGPT handles image uploads, so you don't need a paid plan for this task.

Step 1 — Snap a clear photo of your document

Open your phone's camera app and hold it steady about 30 cm above the page. Make sure the whole document fits in the frame, with even lighting and no shadows crossing the text. If the page is crumpled, flatten it out first so no words are hidden in the creases.

💬 Tip: Tap the screen on the text area before snapping. This locks the camera's focus on the words instead of the background, which makes the text much sharper.

You'll know it worked when the words on the page look crisp and easy to read when you pinch to zoom in on the photo afterwards.

Step 2 — Open ChatGPT and start a new chat

On your phone, open the ChatGPT app and tap the "New chat" icon at the top of the screen (it looks like a pencil or a small plus sign). On a computer, go to the ChatGPT website and click "New chat" in the left-hand sidebar.

💬 Menu path: ChatGPT app → tap the pencil/plus icon at the top, OR open the side menu and choose "New chat".

You'll know it worked when you see a fresh, empty conversation with the placeholder "Message ChatGPT" sitting at the bottom of the screen.

Step 3 — Attach your photo to the chat

Tap the paperclip icon (📎) or the "+" button to the left of the message box. On phone, choose "Photos" and select the image you just took. On computer, click "Upload from computer" and pick the file.

💬 On mobile: Tap "+" → "Photos" → tap your image → tap "Add". On desktop, drag and drop the image straight into the message box if you prefer.

You'll know it worked when a small thumbnail of your photo appears in the message area, waiting to be sent.

Step 4 — Ask the AI to extract the text

In the message box, type a clear instruction telling ChatGPT to read the text from the image. Then tap the up-arrow send button (or press Enter on a computer).

💬 Prompt you can copy:

"Please extract all the text from this image and return it as plain, editable text. Keep the original line breaks where possible."

You'll know it worked when ChatGPT replies within a few seconds with your document's text formatted as readable paragraphs in the chat.

Step 5 — Copy the text from the reply

Once the AI shows you the extracted text, long-press on the message (on phone) or highlight it with your mouse (on computer) and choose "Copy".

💬 On iPhone: Long-press the text → drag the blue handles to select all → tap "Copy". On Android, long-press → "Select all" → "Copy". On computer, highlight the text and press Ctrl+C (or Cmd+C on Mac).

You'll know it worked when you see a small "Copied" notification pop up at the top of your screen.

Step 6 — Paste and save the text in a new document

Open Google Docs, Microsoft Word, or your phone's Notes app. Create a new blank document, tap inside it, and choose "Paste". Then save the file with a clear, searchable name so you can find it again later.

💬 Naming examples: "Tax Receipts 2024", "Year 10 History Notes", or "Grandma's Recipe Book".

You'll know it worked when you can later use Ctrl+F (or the "Find" tool on your phone) to search for a word inside that document and it finds it straight away.

---

Common mistakes

• Photo is too dark or blurry → The AI will guess the words and you'll get nonsense back. Retake the photo with the page well-lit and the camera held steady.
• Prompt is too vague → If you just say "read this", the AI might describe the image instead of giving you the text. Always ask specifically for "plain text" or "editable text".
• Forgetting to proofread the result → AI regularly misreads handwriting and faded print. Always read through the output and fix any typos before saving the final file.
• Closing the chat before copying → The chat stays in your history, but it's safer to paste the text into your own document straight away, in case you lose the thread later.

---