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

Governance / 4 public sources

How do you control built-in AI features in tools you already pay for?

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 4 public sources. Not first-hand experience.

Most AI at work no longer arrives as a new app. It shows up as a summarize button in a PDF reader, an assistant in the office suite, or a browser extension that came with a tool you already license. A public sysadmin thread describes the pattern: a shared PDF opened in a cloud reader with AI features turned on, and a contract pasted into a summarizer that sends the text off site. Blocking AI domains one at a time was called whack a mole, because new features appear inside tools that are already allowed.

The approaches people described were layered. Classify data first, then keep an approved list that names both the tool and the account tier, since a feature covered by a business agreement with no training on your data is different from the same feature on a personal plan. Others pair firewall AI category blocking with DLP, or align with ISO 42001, and several said the policy needs an owner in leadership rather than in IT alone. Vendor admin controls help but have gaps. Adobe's docs show AI in Acrobat can be switched off per product profile, yet a user in two profiles gets access if either allows it, a personal subscription keeps its own features, and changes may wait until the app restarts. The browser extension has a separate policy setting.

Still open: no one had a full inventory of which licensed tools have AI switched on by default, and turning features off one admin console at a time does not scale past a few vendors.

Sources

  • Reddit: How are folks handling AI sprawl?
  • Vendor docs: Manage services in the Admin Console | Adobe Enterprise
  • Vendor docs: Administer Adobe Acrobat AI Assistant for enterprises
  • Vendor docs: Configure Acrobat extension in Chrome and Edge for enterprises