United Airlines' new AI agent policy highlights the growing difference between traditional search engine indexing, AI-powered search and autonomous online transactions.
United Airlines AI agent restrictions Artificial intelligence is transforming how users search for information, compare products and make online purchases. However, as AI-powered agents become more advanced, businesses are reconsidering how these technologies interact with their digital platforms.
United Airlines has introduced updated website terms, effective November 11, 2026, that restrict unauthorized AI agents while allowing traditional search engines to continue indexing publicly accessible pages.
The airline also prohibits using data obtained from its website or mobile application for AI training, grounding, fine-tuning and model development.
This development offers important insights for SEO professionals, ecommerce businesses and digital marketers preparing for an increasingly AI-driven search ecosystem.
AI-powered assistants are no longer limited to answering questions. Modern autonomous agents can browse websites, compare prices, manage customer accounts and complete transactions.
While these capabilities improve convenience, they also introduce considerable security and operational challenges.
According to United Airlines' official announcement, allowing unauthorized third-party AI assistants to interact with its platforms could expose customers to risks such as:
For example, airline check-in processes require passengers to acknowledge important information regarding prohibited baggage items. Allowing automated agents to complete these activities may create regulatory compliance concerns.
United's updated policy aims to maintain greater control over customer interactions, commercial data and transactional activities.
One of the most important aspects of United Airlines' updated policy is the distinction between search engine crawlers and autonomous AI agents.
Traditional search engines primarily discover, crawl and index website content. AI agents can go beyond information discovery by performing actions on behalf of users.
Activity
Traditional search indexing
Unauthorized AI agents
Index public website pages
Allowed
Not exempt
Display traditional search results
Allowed
Not applicable
Retrieve data for AI grounding
Not expressly permitted
Restricted
Book flight tickets
Not part of indexing
Restricted
Modify reservations
Not part of indexing
Restricted
Extract data for AI training
Not covered by exemption
Restricted
Written permission
Not required for public indexing
Required
This distinction demonstrates that businesses can maintain organic search visibility while introducing separate controls for autonomous AI activities.
United Airlines' revised website terms and conditions broadly define automated agents as software, services or digital features that operate with partial or complete autonomy.
These may include AI chatbots, automated booking assistants, browser extensions, web scraping tools and AI-enabled browser features.
Even if a human supervises or approves an automated agent's actions, the agent may still be subject to these restrictions.
However, United provides exceptions for traditional search engines indexing public pages, assistive technologies and standard web browsers operated by individuals.
This means United is not blocking traditional search engine visibility. Instead, it is establishing boundaries around how automated technologies can interact with its platforms.
AI grounding is one of the most significant aspects of United's revised policy.
Grounding refers to the process of connecting an AI model with external information to improve the relevance and accuracy of its responses.
For example, an AI travel assistant may retrieve flight schedules, ticket prices and seat availability from an airline's website before generating personalized travel recommendations.
United's updated terms restrict the use of its website and application data for activities including:
However, the policy does not explicitly explain how AI-generated answers based on existing search engine indexes will be treated.
This creates an important distinction for businesses investing in AI SEO, Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO).
Being indexed by a search engine does not automatically mean that website data is authorized for every type of AI application.
United Airlines AI agent restrictions
United Airlines has also revised its contract of carriage, extending its AI restrictions beyond its own website and mobile application.
Travel agencies and third-party booking platforms must obtain permission before using AI agents to search, purchase, modify or manage United Airlines travel services.
Under the updated passenger agreement, certain unauthorized AI-assisted bookings may be canceled without prior notice.
According to United's FAQ, bookings made by AI agents before November 11, 2026, will not be canceled solely for that reason.
This development could influence how travel companies integrate AI assistants into their booking platforms in the future.
For years, robots.txt has been an important component of technical SEO and website crawl management.
However, autonomous AI agents present new challenges.
Some AI-powered browsing assistants may not provide identifiable user-agent information. Others may interact with websites through conventional browser environments, making them more difficult to distinguish from human visitors.
According to Search Engine Journal, United's robots.txt file did not specifically name AI crawlers or agents as of November 14, 2026.
Instead, the company introduced restrictions through its website terms, passenger agreement and access policies.
Businesses should understand that robots.txt communicates crawling preferences but is not a comprehensive security or access-control mechanism.
Organizations may need to combine crawler directives with server-side bot detection, API restrictions, authentication systems and updated website policies.
United's decision provides several valuable lessons for businesses preparing for the future of AI search.
A website can remain accessible to traditional search engines while restricting certain AI-powered activities.
SEO teams should evaluate different types of automated access instead of treating all bots equally.
Website owners should regularly analyze server logs and crawler activity to understand which automated systems access their content.
This can help identify legitimate search crawlers and potentially unauthorized automated traffic.
Modern technical SEO strategies should consider website security, AI crawler accessibility, structured data, server performance and content governance.
Website terms and conditions should be reviewed to address automated transactions, commercial data extraction and AI-related content usage.
Legal and technical teams should collaborate when developing these policies.
Blocking all AI crawlers may limit visibility across certain AI-powered discovery platforms.
Businesses should balance data protection with their SEO, AEO and GEO objectives before implementing restrictions.
Source: United Digital Trust Center