Token Thrift
  1. Ask
  2. Agents and governance
  3. Governance
  4. Your account
Started by Token Thrift

Governance / 3 public sources

Who owns the apps non-engineers build with AI tools?

Short answer

Short answer: The team that built and uses the app should own it, with a named owner and a backup person, and IT should own the keys and data access it runs on. Find existing apps by auditing service keys and API keys on company data, since nobody keeps a list of the apps themselves. Then give builders a sanctioned place to build so new apps stop landing on personal accounts.

How do you find apps non-engineers built with AI?

Start from access, not apps. List every service key, API key and OAuth grant on your databases and SaaS tools, ask who uses each one, and rotate unclaimed keys after a warning. The apps that depended on them surface quickly.

Should IT block AI app builders?

Replies in the public threads mostly say no. Blocking pushes the apps onto personal accounts where nobody sees them. A sanctioned workspace, such as an isolated cloud account with company sign in, keeps them visible.

What rule stops this from happening again?

No app touches company data without a named team owner and a backup person, and keys are issued per app, never shared between apps or people.

Written by Token Thrift from the public sources below. Vendor facts checked on November 2, 2026.

See also

Add your answer

Loading replies.

Add your answer

Don't paste anything your employer treats as confidential. Before anything is posted, we take out names of people, your company, its clients and internal projects, plus places and exact numbers, and show you the result to check. Your raw text is never stored. To redact it, we send it to Anthropic's API, which doesn't train on API data by default.

How Ask protects you

  • You post under a random handle. No name or employer is shown, and your email is never shown to anyone.
  • Before a reply is saved, a model removes names, employers, clients, places and exact figures, a fixed leak check reads it again, and you approve the result. Your raw text is never stored. It is sent to Anthropic's API for this step. Redaction lowers the risk; it is not a guarantee.
  • Follow-up questions are passed on by email, and neither person sees the other's address. You can answer privately.
  • A handle's first reply is reviewed before it is published.
  • Replies that are never published are deleted after 30 days. Deleting your handle removes your email; published replies stay up as anonymous text.

More in the privacy policy and the FAQ.

What you get for replying

  • A digest of everyone else's answers on that topic, at most one email a day, one click to stop.
  • A count of replies readers marked helpful, shown next to your handle.
  • Credit by handle in the monthly roundups when your answer is used.

Tips to stay hard to recognize

  • Describe your company by size and industry, not by name, product or city.
  • Leave out unusual tool combinations or internal project details only your team would know.
  • Round numbers and timelines yourself, even though redaction also does it.
  • Wait a while before writing about something that was just announced inside your company.
  • Redaction doesn't change your writing style, so a coworker with context could still recognize you.

Summary of public sourcesWritten by our research agent from 3 public sources. Not first-hand experience.

Marketing, ops and finance teams now ship working internal apps with AI app builders and coding assistants. In the public threads, the same failure shows up again and again: an app built on someone's personal hosting account, talking to a company database with a shared service key, and a builder who has since left. Nobody knew the app existed until a key was rotated and it broke.

The practical advice in the replies starts with the keys, not the apps. List every service key and API key on company databases and SaaS tools and ask who uses each one. Unclaimed keys get rotated after a warning, which surfaces the hidden apps quickly. Teams that went further gave builders a sanctioned place to build, such as an isolated cloud account, and made the building team the named owner responsible for security findings. Blocking the tools tended to push the apps further out of sight.

Still open: how much review a small internal app needs before it touches customer data, and whether IT or the business team should carry the pager when it breaks.

Sources

  • Reddit: How many apps have non-engineers at your company built with AI tools, and who owns them when they break?
  • Reddit: Using Claude to write most of a company's internal tools
  • Reddit: How to handle the sprawl of AI applications?

Newer related questions