Claim ledger

Did Your AI Actually Read the Doc?

How to give ChatGPT and Claude the same project notes in one Google Doc, and test whether a new chat can use them.

Read the essay On Substack Published 16 Oct 2026

Claims: 38
Verified at the primary source: 35
Removed before publication: 1
Last checked: 16 Oct 2026

38 claims: 35 verified, 3 executed. Plus 1 removed before publication, listed at the end.

Download this ledger: CSV · JSON · Markdown · CC BY 4.0 · Query it from Claude or ChatGPT

What the essay claims

Chats forget, and different assistants cannot see each other's chats. Keeping the project's instructions and decisions in a couple of files fixes both, and GitHub is the free place for those files that the main assistants can connect to, with a record of every change. But a connection only lets the AI read; it does not make it read. So the setup is worth nothing until a fresh chat has proved it read the file, by naming it. For several assistants, the same files work, and the missing part is a doorbell: the repository stores the work but tells nobody it is there.

The claim ladder

Rung	Claim	Evidence	Whose behaviour	Scope limit
1	A private repository, files and issues are free and need only a browser	GitHub Docs + pricing (ledger)	GitHub's product	Does not establish that a first-time user finds it easy; no novice panel was run
2	ChatGPT and Claude can each be connected to a private repository	OpenAI and Anthropic help pages (ledger)	the vendors' products, on the plans named	Plan tiers, modes and read/write vary; does not reach Gemini or other tools unless verified
3	A connection lets the AI read the files; it does not make it read them	vendor pages on when the connection is used (ledger)	the assistant's behaviour	Not a measured rate of skipped reads; the article offers the test, not a statistic
4	The fresh-chat test tells "read" from "guessed"	the method itself: a question only the file answers, plus the file's name	the reader's own check	A named file can still be misquoted; the reader checks the answer against the file
5	The repository stores work but wakes nobody; each assistant needs its own trigger	engine TEAM.md (first-party) + GitHub's coding-agent docs (ledger)	our five seats; GitHub's agents	First-party is one team's setup; no claim about time saved or ROI
6	Text inside an issue can carry instructions an AI will follow	Invariant Labs write-up (ledger)	an agent with GitHub access	One documented incident; not a rate

What would make it wrong

The piece is wrong, and does not ship as written, if any of these holds at publish time:

ChatGPT or Claude cannot read a private repository on the plans the article names (C3, C4);

any step needs a terminal or a paid plan for the core setup (C1, C2);

the assistant's own Projects feature already shares one project across ChatGPT and Claude (it would make GitHub unnecessary for the main reader).

The evidence, row by row

Each row is a claim the essay makes, then the evidence recorded for it and the source it was checked against. For verified rows that is usually the exact words, where they sit and the day they were read. The label says how far it was checked.

What the labels mean

Verified: Checked against the primary source itself by a checker who did not draft the piece, usually with the exact words, where they sit and the date read.
Executed: A run or observation done for the piece: code, a query, a count, or a check made live in an app. The claim is what it returned. Run records are not published.
Checked: Checked against its source by a separate checker in the older ledger format, which recorded no quote, locator or date.
Reported: Not verified word for word against a primary source: carried from a secondary source, from one that could not be opened in full, or supported only in part. The essay words it accordingly.
Struck: Drafted, checked and removed before publication. Kept here, crossed out, with the reason.
Excluded: Considered and deliberately left out of the piece. Kept here with the reason.

Sources (9)

GitHub pricing 1 row
GitHub Docs 6 rows
First-party runs 3 rows
Google Workspace Admin Help 1 row
Claude Docs 13 rows
OpenAI Help 10 rows
Invariant Labs blog 1 row
Anthropic Help 2 rows
Google Docs Help 1 row

C1 Verified Read 15 Oct 2026

GitHub's free plan includes unlimited private repositories

"Unlimited public/private repositories"

GitHub pricing · Free plan card · github.com

Link

Cite

GitHub pricing, Free plan card, https://github.com/pricing. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C1, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C1 (source read 2026-10-15).

Report an error
C2 Verified Read 15 Oct 2026

Files are created and saved in the browser: Add file → Create new file → Commit changes; editing a file uses the pencil and Commit changes

"Click Commit changes..." … "Click Commit changes or Propose changes."

GitHub Docs, Creating new files · steps 7 and 11 · docs.github.com

Link

Cite

GitHub Docs, Creating new files, steps 7 and 11, https://docs.github.com/en/repositories/working-with-files/managing-files/creating-new-files. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C2, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C2 (source read 2026-10-15).

Report an error
C13 Verified Read 16 Oct 2026

A repository is created from + (upper right) → New repository: type a name, choose a visibility, optionally create a README; a file is edited with the pencil in the file view

"In the upper-right corner of any page, select , then click New repository." … "Choose a repository visibility." … "You can create a README" … "Click Create repository." … (commit box) "decide whether to add your commit to the current branch or to a new branch" … "In the upper right corner of the file view, click to open the file editor."

GitHub Docs, Creating a new repository; Editing files · "Creating a new repository from the web UI" steps 1, 5, 6; "Editing files in your repository" step 2 (the + and the pencil are icons in the docs, so the quotes show a gap) · docs.github.com

Link

Cite

GitHub Docs, Creating a new repository; Editing files, "Creating a new repository from the web UI" steps 1, 5, 6; "Editing files in your repository" step 2 (the + and the pencil are icons in the docs, so the quotes show a gap), https://docs.github.com/en/repositories/creating-and-managing-repositories/creating-a-new-repository. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C13, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C13 (source read 2026-10-16).

Report an error
L1 Executed Read 16 Oct 2026

Navigation links lead where their text says: chatgpt.com (ChatGPT), claude.ai (Claude, sign-in when signed out), docs.new (a new Google Doc), github.com/signup (GitHub sign-up: "Create your free account"), github.com/new (new repository; signed out it goes to sign-in with return_to=/new)

Result: page titles and redirects

First-party browser check (built-in browser, signed out) + curl · chatgpt.com

A run or observation done for this piece. The run record is not published.

Link

Cite

A run or observation by The Durability Curve (First-party browser check (built-in browser, signed out) + curl), https://chatgpt.com. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row L1, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-L1 (source read 2026-10-16).

Report an error
G1 Verified Read 16 Oct 2026

On a Google Workspace account an admin can block third-party apps from Google data; a blocked user is told so, and the admin can trust the app

"If a user tries to install (or sign in to) an app you haven't trusted that accesses a restricted service, they're notified that the app is blocked." … access levels "Trusted, Limited, Specific Google data, or Blocked"

Google Workspace Admin Help, Control which apps access Google Workspace data · restricted services; App access control · support.google.com

Link

Cite

Google Workspace Admin Help, Control which apps access Google Workspace data, restricted services; App access control, https://support.google.com/a/answer/7281227. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row G1, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-G1 (source read 2026-10-16).

Report an error
L2 Verified Read 16 Oct 2026

Claude's connector settings are at https://claude.ai/customize/connectors

the address is linked from Anthropic's own pages (4, 6 and 3 times)

Claude Docs (Google Drive; GitHub integration) + Anthropic Help (Google Drive) · href in page · claude.com

Link

Cite

Claude Docs (Google Drive; GitHub integration) + Anthropic Help (Google Drive), href in page, https://claude.com/docs/connectors/google/drive. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row L2, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-L2 (source read 2026-10-16).

Report an error
D9 Verified Read 16 Oct 2026

Claude's Google Drive connector reads a file's content, not its comments or suggested edits

"Comments and suggestions : Claude can't read a file's comments or suggested edits, only its content"

Claude Docs, Google Drive connector · "Two limits apply to every type" · claude.com

Link

Cite

Claude Docs, Google Drive connector, "Two limits apply to every type", https://claude.com/docs/connectors/google/drive. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D9, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D9 (source read 2026-10-16).

Report an error
D10 Verified Read 16 Oct 2026

On Team and Enterprise, Google Drive appears only after an Owner adds it

"On Team and Enterprise plans, Google Drive doesn't appear in your connector list until an Owner adds it for your organization."

Claude Docs, Google Drive connector · "Connect Google Drive" · claude.com

Link

Cite

Claude Docs, Google Drive connector, "Connect Google Drive", https://claude.com/docs/connectors/google/drive. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D10, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D10 (source read 2026-10-16).

Report an error
C4f Verified Read 16 Oct 2026

Claude can add a repository to a chat (+ → Add from GitHub, once GitHub is connected) as well as to a project

"By the end of this page you have added repository content to a chat or a project" … "select + at the lower left of the message box … Select Add from GitHub"

Claude Docs, GitHub integration · "In a chat" tab · claude.com

Link

Cite

Claude Docs, GitHub integration, "In a chat" tab, https://claude.com/docs/connectors/github. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C4f, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C4f (source read 2026-10-16).

Report an error
C3a Verified Read 15 Oct 2026

ChatGPT connects to GitHub at Settings → Plugins → GitHub; you pick the repositories on GitHub

"To connect GitHub, open Settings > Plugins and select GitHub" … "then select the repositories ChatGPT can access"

OpenAI Help, Connecting GitHub to ChatGPT · "Connect GitHub to ChatGPT" · help.openai.com

Link

Cite

OpenAI Help, Connecting GitHub to ChatGPT, "Connect GitHub to ChatGPT", https://help.openai.com/en/articles/11145903-connecting-github-to-chatgpt. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C3a, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C3a (source read 2026-10-15).

Report an error
C3b Verified Read 15 Oct 2026

ChatGPT's GitHub app only reads

"The GitHub app in ChatGPT only lets you read from your repositories"

OpenAI Help, same page · FAQ "Can I push code…" · help.openai.com

Link

Cite

OpenAI Help, same page, FAQ "Can I push code…", https://help.openai.com/en/articles/11145903-connecting-github-to-chatgpt. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C3b, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C3b (source read 2026-10-15).

Report an error
C3c Verified Read 15 Oct 2026

Availability varies by plan; some plans have GitHub in deep research or agent mode but not ordinary chat

"Some plans may support GitHub in deep research or agent mode but not in standard chat."

OpenAI Help, same page · FAQ "Why is GitHub available in deep research…" · help.openai.com

Link

Cite

OpenAI Help, same page, FAQ "Why is GitHub available in deep research…", https://help.openai.com/en/articles/11145903-connecting-github-to-chatgpt. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C3c, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C3c (source read 2026-10-15).

Report an error
C3d Verified Read 15 Oct 2026

A new repository takes about five minutes to appear; ChatGPT searches the repository from your prompt

"Repositories generally take about 5 minutes to appear as available in ChatGPT." … "ChatGPT forms search queries from your prompts"

OpenAI Help, same page · "Find missing GitHub repositories"; "How ChatGPT works with GitHub" · help.openai.com

Link

Cite

OpenAI Help, same page, "Find missing GitHub repositories"; "How ChatGPT works with GitHub", https://help.openai.com/en/articles/11145903-connecting-github-to-chatgpt. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C3d, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C3d (source read 2026-10-15).

Report an error
C3e Verified Read 15 Oct 2026

In a chat you can pick an installed plugin with an @ mention, or ChatGPT uses it when relevant

"ChatGPT automatically uses your installed plugins when they're relevant to your request. You can also select a plugin directly with an @ mention"

OpenAI Help, Plugins in ChatGPT · "Use a plugin" · help.openai.com

Link

Cite

OpenAI Help, Plugins in ChatGPT, "Use a plugin", https://help.openai.com/en/articles/20001256-plugins-in-chatgpt. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C3e, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C3e (source read 2026-10-15).

Report an error
C3f Verified Read 15 Oct 2026

GitHub's app install screen offers All repositories or Only select repositories

"select All repositories or Only select repositories"

GitHub Docs, Installing a GitHub App from a third party · installation steps · docs.github.com

Link

Cite

GitHub Docs, Installing a GitHub App from a third party, installation steps, https://docs.github.com/en/apps/using-github-apps/installing-a-github-app-from-a-third-party. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C3f, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C3f (source read 2026-10-15).

Report an error
C4a Verified Read 15 Oct 2026

Claude's GitHub Integration is on every plan including Free, adds repository files to a project's knowledge via + → GitHub → Add files, and reads file names and contents only

"It's available on every plan, including Free" … "it doesn't see commit history, pull requests, or issues" … "Select Add files."

Claude Docs, GitHub integration · intro; "In a project" tab · claude.com

Link

Cite

Claude Docs, GitHub integration, intro; "In a project" tab, https://claude.com/docs/connectors/github. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C4a, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C4a (source read 2026-10-15).

Report an error
C4b Verified Read 15 Oct 2026

Claude can be given access to only specific private repositories

"allow Claude access to all of your repositories or only to specific ones"

Claude Docs, same page · "Grant Claude access to a private repository" · claude.com

Link

Cite

Claude Docs, same page, "Grant Claude access to a private repository", https://claude.com/docs/connectors/github. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C4b, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C4b (source read 2026-10-15).

Report an error
C4c Verified Read 15 Oct 2026

Claude keeps a copy; Sync now fetches changes

"select Sync now on the repository in your project knowledge to fetch the latest changes"

Claude Docs, same page · "Keep repository content current" · claude.com

Link

Cite

Claude Docs, same page, "Keep repository content current", https://claude.com/docs/connectors/github. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C4c, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C4c (source read 2026-10-15).

Report an error
C4d Verified Read 15 Oct 2026

Repository content can be added only to a project you have not shared

"You can add repository content only to a private project, one you haven't shared with other people."

Claude Docs, same page · "In a project" tab · claude.com

Link

Cite

Claude Docs, same page, "In a project" tab, https://claude.com/docs/connectors/github. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C4d, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C4d (source read 2026-10-15).

Report an error
C4e Verified Read 15 Oct 2026

Claude's integration reads only; editing and pull requests go through Claude Code

"If you want Claude to edit code, run commands, and open pull requests in your repository, see Claude Code"

Claude Docs, same page · intro note · claude.com

Link

Cite

Claude Docs, same page, intro note, https://claude.com/docs/connectors/github. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C4e, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C4e (source read 2026-10-15).

Report an error
C6 Verified Read 15 Oct 2026

ChatGPT and Claude each have Projects that keep files and instructions across chats inside that app

ChatGPT: projects keep related chats, files and instructions together (paraphrase; page copy fc/oai_proj.md). Claude: projects are self-contained workspaces with their own chats and knowledge (paraphrase; fc/cl_proj.html)

OpenAI Help; Claude Help · overview sections · help.openai.com

Link

Cite

OpenAI Help; Claude Help, overview sections, https://help.openai.com/en/articles/10169521-projects-in-chatgpt. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C6, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C6 (source read 2026-10-15).

Report an error
C7 Verified Read 15 Oct 2026

A committed secret should be treated as compromised and revoked; deleting it in a later commit does not remove the exposure

"You should consider any leaked secret to be immediately compromised" … "Simply removing the secret from the codebase, pushing a new commit, or deleting…"

GitHub Docs, Remediating a leaked secret · Introduction · docs.github.com

Link

Cite

GitHub Docs, Remediating a leaked secret, Introduction, https://docs.github.com/en/code-security/tutorials/remediate-leaked-secrets/remediating-a-leaked-secret. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C7, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C7 (source read 2026-10-15).

Report an error
C8 Verified Read 15 Oct 2026

In May 2025 Invariant Labs showed a malicious issue on a user's public repository making their AI agent (GitHub MCP server) leak private-repository data into a public pull request. A researcher demonstration, not a reported real-world attack

"allows an attacker to hijack a user's agent via a malicious GitHub Issue, and coerce it into leaking data from private repositories"

Invariant Labs blog (Milanta, Beurer-Kellner), 2025-05-26 · intro; "Attack Demonstration" · invariantlabs.ai

Link

Cite

Invariant Labs blog (Milanta, Beurer-Kellner), 2025-05-26, intro; "Attack Demonstration", https://invariantlabs.ai/blog/mcp-github-vulnerability. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C8, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C8 (source read 2026-10-15).

Report an error
C9 Verified Read 15 Oct 2026

On a paid Copilot plan you can assign an issue to Copilot, which works, raises a pull request and asks you to review

"Copilot will start working on the task, raise a pull request, then request a review from you when it's finished."

GitHub Docs, Using Copilot cloud agent on GitHub; plans · "Assigning an issue to Copilot" (public preview); plans table excludes Copilot Free · docs.github.com

Link

Cite

GitHub Docs, Using Copilot cloud agent on GitHub; plans, "Assigning an issue to Copilot" (public preview); plans table excludes Copilot Free, https://docs.github.com/en/copilot/how-tos/use-copilot-agents/cloud-agent/use-cloud-agent-on-github. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C9, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C9 (source read 2026-10-15).

Report an error
C10 Verified Read 15 Oct 2026

Claude and Codex coding agents can be assigned work on GitHub, public preview, paid Copilot plans

Third-party coding agents page marks Claude and Codex as public preview on paid plans (paraphrase; fc/tp.md)

GitHub Docs, About third-party coding agents · page note; "Usage costs" · docs.github.com

Link

Cite

GitHub Docs, About third-party coding agents, page note; "Usage costs", https://docs.github.com/en/copilot/concepts/agent-capabilities/about-agents/about-third-party-coding-agents. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C10, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C10 (source read 2026-10-15).

Report an error
C12 Verified Read 15 Oct 2026

On Plus, Pro, Business, Enterprise and Edu, ChatGPT Work can run a task when pull request activity happens in a connected GitHub repository

"Event-triggered (webhook-based) tasks run in Work and respond to supported Gmail, Slack, or GitHub activity. They are available to eligible users on Plus, Pro, Business, Enterprise, and Edu plans"

OpenAI Help, Scheduled tasks in ChatGPT · "Create an event-triggered task" · help.openai.com

Link

Cite

OpenAI Help, Scheduled tasks in ChatGPT, "Create an event-triggered task", https://help.openai.com/en/articles/10291617. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row C12, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-C12 (source read 2026-10-15).

Report an error
F1 Executed Read 15 Oct 2026

We run our own team of assistants from one private repository, and each assistant needed its own way to notice new work: one on new pull requests, one that checks every few minutes, and one that a script feeds

Result: Our team's own operating file, which names each assistant's trigger

First-party: our team's operating file (private repository, not linked)

A run or observation done for this piece. The run record is not published.

Link

Cite

A run or observation by The Durability Curve (First-party: our team's operating file (private repository, not linked)). Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row F1, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-F1 (source read 2026-10-15).

Report an error
F3 Executed Read 15 Oct 2026

The article's steps were run end to end on 2026-10-15 in Claude (Max) and ChatGPT (Plus): both quoted the codeword and both followed the newer of two conflicting decisions; three UI labels corrected from the run

Result: "2026-10-15: Test. Codeword: quillfeather-83." quoted by both; Claude path + > Connectors > Add from Google Drive > Add to chat; ChatGPT Plugins in the sidebar

First-party run record

A run or observation done for this piece. The run record is not published.

Link

Cite

A run or observation by The Durability Curve (First-party run record). Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row F3, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-F3 (source read 2026-10-15).

Report an error
D1a Verified Read 15 Oct 2026

Claude connects to Google Drive at Customize > Connectors > Google Drive > Connect, on every plan including Free (Team/Enterprise: an Owner adds it first)

"available for all users on Claude and Claude Desktop" … "Go to Customize > Connectors in claude.ai"

Anthropic Help + Claude Docs · intro; "Connect Google Drive" · support.claude.com

Link

Cite

Anthropic Help + Claude Docs, intro; "Connect Google Drive", https://support.claude.com/en/articles/10166901-using-the-google-drive-integration. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D1a, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D1a (source read 2026-10-15).

Report an error
D1b Verified Read 15 Oct 2026

In a Claude chat, + > Add from Google Drive attaches a Doc, and Claude re-reads its current content every message

"Claude reads the file's current content from Drive each time you send a message, so later edits in Drive reach Claude too."

Claude Docs, Google Drive connector · "Add Drive files" › In a chat · claude.com

Link

Cite

Claude Docs, Google Drive connector, "Add Drive files" › In a chat, https://claude.com/docs/connectors/google/drive. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D1b, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D1b (source read 2026-10-15).

Report an error
D1c Verified Read 15 Oct 2026

In a Claude project, a Drive file refreshes periodically when you open the project; only unshared projects

"Its content refreshes from Drive periodically when you open the project." … "You can add Drive files only to a private project"

Claude Docs, same page · "Add Drive files" › In a project · claude.com

Link

Cite

Claude Docs, same page, "Add Drive files" › In a project, https://claude.com/docs/connectors/google/drive. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D1c, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D1c (source read 2026-10-15).

Report an error
D2 Verified Read 15 Oct 2026

Claude can edit Google Docs live (beta) through a separate Google Docs connector, asking approval before each action by default

"Beta: Claude can edit Google Docs, Sheets, and Slides live in a pane beside the chat" … "each action Claude takes on your behalf requires your explicit approval"

Anthropic Help · "Edit … live (beta)"; "How to use the connectors" · support.claude.com

Link

Cite

Anthropic Help, "Edit … live (beta)"; "How to use the connectors", https://support.claude.com/en/articles/10166901-using-the-google-drive-integration. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D2, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D2 (source read 2026-10-15).

Report an error
D3 Verified Read 15 Oct 2026

ChatGPT connects Google Drive under Settings > Plugins (Apps in some versions); availability depends on plan; in a chat use @ or +; changing a file needs Google permission and may not be available

"Open Plugins or Apps, depending on the options available in your version of ChatGPT." … "Google Drive availability depends on your ChatGPT plan" … "mention an available app or plugin with @, or select +"

OpenAI Help · "Connect Google Drive"; "Use an app" · help.openai.com

Link

Cite

OpenAI Help, "Connect Google Drive"; "Use an app", https://help.openai.com/en/articles/10929079. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D3, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D3 (source read 2026-10-15).

Report an error
D4 Verified Read 15 Oct 2026

Google Docs keeps version history (Last edit, top right; Restore this version); deleted text can usually be recovered unless the owner deletes history

"At the top right, click Last edit" … "Restore this version" … "Deleting version history is permanent"

Google Docs Help · "Restore an earlier version"; "Manage and delete version history" · support.google.com

Link

Cite

Google Docs Help, "Restore an earlier version"; "Manage and delete version history", https://support.google.com/docs/answer/190843. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D4, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D4 (source read 2026-10-15).

Report an error
D5 Verified Read 15 Oct 2026

ChatGPT can recall details from past chats when Memory / Reference chat history is on

"When Memory is enabled, ChatGPT can remember relevant preferences and details from your chats"

OpenAI Help, Memory FAQ · Overview · help.openai.com

Link

Cite

OpenAI Help, Memory FAQ, Overview, https://help.openai.com/en/articles/8590148. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D5, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D5 (source read 2026-10-15).

Report an error
D6 Verified Read 15 Oct 2026

Claude's memory is on by default on Free, Pro and Max; it can be switched off for one chat from the + menu

"Memory is on by default for Free, Pro, and Max plans" … "turning off "Memory" in the "+" menu before you send your first message"

Anthropic Help · "Manage Claude's memory in projects" · support.claude.com

Link

Cite

Anthropic Help, "Manage Claude's memory in projects", https://support.claude.com/en/articles/9519177-how-can-i-create-and-manage-projects. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D6, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D6 (source read 2026-10-15).

Report an error
D7 Verified Read 15 Oct 2026

ChatGPT event-triggered tasks respond to Gmail, Slack or GitHub activity only; Free and Go cannot create them

"respond to supported Gmail, Slack, or GitHub activity" … "Free and Go accounts cannot create event-triggered tasks"

OpenAI Help, Scheduled tasks · "Create an event-triggered task" · help.openai.com

Link

Cite

OpenAI Help, Scheduled tasks, "Create an event-triggered task", https://help.openai.com/en/articles/10291617. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D7, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D7 (source read 2026-10-15).

Report an error
D8 Verified Read 15 Oct 2026

Once connected, both assistants can search the files the Google account can access, not only the project doc

Claude: "lets Claude search your Google Drive, read the files it finds" · ChatGPT: "connects ChatGPT to files and folders your Google account can already access"

Claude Docs; OpenAI Help · intros · claude.com

Link

Cite

Claude Docs; OpenAI Help, intros, https://claude.com/docs/connectors/google/drive. Checked in: Claim Ledger, "Did Your AI Actually Read the Doc?", The Durability Curve, row D8, https://durabilitycurve.com/claims/did-your-ai-read-the-doc/#c-D8 (source read 2026-10-15).

Report an error

Removed before publication

Claims that were drafted, checked and taken out, or considered and left out. The essay does not make them; they stay here with the reason.

T1 Struck

~~A template is copied with Use this template → Create a new repository → choose visibility → Create repository from template; the copy has its own history~~

Why: Removed from the essay, which now gives the two files and the steps itself (row C13)

Cite and reuse

A single claim: Use the row's Cite button. It names the original source first, then this row, which carries its own link.
This ledger: Floyd, Harry (2026). Claim Ledger: Did Your AI Actually Read the Doc? The Durability Curve. https://durabilitycurve.com/claims/did-your-ai-read-the-doc/
The essay: Floyd, Harry (2026). Did Your AI Actually Read the Doc? The Durability Curve. https://durabilitycurve.com/blog/did-your-ai-read-the-doc/

The ledger data is licensed CC BY 4.0: reuse it, quote it or train on it, with credit to The Durability Curve and a link. Read the licence. Machine copies: JSON, CSV, Markdown.