Survey: 29% of Hospitality Brands Let Staff Use Public AI With No Guardrails
Hospitality Technology's forthcoming 2027 AI Impact Study finds that 29% of restaurant brands let team members use AI tools with no guardrails, standards, or training. Restaurant technology expert Chris Munz said on a recent HT webinar that stores run 20 to 25 systems with hundreds of integrations, so an ungoverned tool can halt a brand, with an outage potentially costing millions of dollars in lost revenue. The risk often starts with a manager improvising during a dinner rush, with staff feeding store operational data, schedules, or unvetted code into public AI models. Munz recommended drafting a clear AI policy, fitting security into operations, auditing vendor access, cleaning up old per…
Impact and considerations
Hospitality and restaurant companies are adopting AI tools quickly, yet nearly a third lack basic controls. For hotel operators and business-travel stakeholders, data exposure and system outages directly threaten revenue continuity and guest trust, making AI governance part of operational risk management.
Key points
- Hospitality Technology's forthcoming 2027 AI Impact Study shows 29% of restaurant brands let team members use AI tools with no guardrails, standards, or training.
- Expert Chris Munz said stores run 20 to 25 systems with hundreds of integrations, so an ungoverned tool can halt a brand, with an outage potentially costing millions of dollars in lost revenue.
- The risk usually starts with a manager improvising during a dinner rush, with staff feeding store operational data, schedules, or unvetted code into public AI models.
- Munz recommended drafting a clear AI policy defining what data employees can input into public tools, focusing on guest details, passwords, and store financials.
- Build security protocols around actual store realities so cumbersome measures do not push staff to unsecure workarounds.
- Audit third-party vendor user permissions, multi-factor authentication, and read-and-write access to core systems, and remove access for departed staff.
- Build an incident plan identifying who to call and what steps to take when a system goes down.
Sources and time
- Primary source
- Hospitality Technology
- Other sources
- 0
- First source publication
- 15 Sept 2026, 00:40
- Page published
- 15 Sept 2026, 01:17
- Last updated
- 15 Sept 2026, 00:40
- Original links
- Hospitality Technology：29% of Hospitality Brands Let Staff Use Public AI With No Guardrails. Here’s How to Address It. (opens in a new tab)Primary source · en · Published 15 Sept 2026, 00:40