Editorial illustration for Cars.com and Amtrak Block AI Agents as Websites Quietly Close Doors
Cars.com and Amtrak Block AI Agents Access
Try booking a train ticket through an AI agent on Amtrak's site this month and you'll likely hit a wall that wasn't there a few weeks ago. Same story on Cars.com. Users of Meta's Muse, xAI's Grok, OpenAI's Dots, and a tool called Instinct have started running into blocks that didn't exist before, and nobody's posting an announcement about it. The agents just stop working on certain sites, with no explanation offered to the person who was relying on them to get something done.
This is the gap between what AI companies are promising and what the rest of the internet is willing to tolerate. OpenAI, Meta, and Google have all pitched a future where an agent handles the tedious parts of your life: booking dinner, clearing your inbox, buying shoes. Yelp, Delta, JPMorgan Chase, and Charles Schwab have their own ideas about who gets to touch their systems and how. Some of that is about money, some of it is about security, and some of it is just companies deciding they'd rather not find out what happens when a bot starts clicking through a banking portal unsupervised.
Amazon already went to court over this exact tension.
Yelp told TechCrunch that it only allows AI agents that pay for its data licensing program, while Delta Air Lines is preventing them from booking tickets on behalf of users. JPMorgan Chase and Charles Schwab have also confirmed that they’re blocking AI agents from accessing accounts due to security concerns.
Why this matters
For founders building agentic products, this is the gap between demo and deployment. An agent that can't reliably reach Cars.com or book an Amtrak ticket isn't autonomous, it's a liability wrapped in a pitch deck. Amazon's lawsuit against Perplexity already showed that the biggest sites have lawyers ready to defend their traffic, and that calculus won't change because OpenAI or Meta ships a flashier agent.
If anything, the companies with the most to lose from bot traffic, airlines, retailers, ticketing platforms, have every incentive to lock the door quietly rather than announce a policy. That makes the current wave of agent demos look less like a preview of the future and more like a best-case scenario running on borrowed access. Developers should treat site compatibility as a moving target, not a solved problem, and build in fallbacks for when access disappears overnight.
Researchers tracking agent capability benchmarks should ask how many of those benchmarks assume an open web that's already closing. The real contest here isn't model intelligence, it's who controls the gate, and right now the gatekeepers are winning.
Related Reading
Industry Applications Westinghouse teams with Google Cloud to build AI platform for nuclear power
Industry Applications NVIDIA NeMo powers telco reasoning model for autonomous network workflows
Industry Applications Month-1 Agent Adds Holistic Observability with Trace IDs and Token Tracking
From Archives Microsoft AI Tool to Help Resource-Strapped Weather Agencies Forecast Better
From Archives Alibaba Qwen Unveils QwenGyre: Elastic RL for Long-Horizon Agents
Common Questions Answered
Why are AI agents like Meta's Muse and OpenAI's Dots being blocked from Cars.com and Amtrak?
Companies like Cars.com and Amtrak have quietly implemented blocks to prevent AI agents from accessing their services, though they haven't publicly announced their reasons. This appears to be part of a broader trend where major websites are restricting AI agent access due to concerns about unauthorized traffic, data usage, and liability issues.
What security concerns are JPMorgan Chase and Charles Schwab citing for blocking AI agents?
JPMorgan Chase and Charles Schwab have confirmed they're blocking AI agents from accessing user accounts specifically due to security concerns. These financial institutions are preventing agents from booking transactions or accessing sensitive account information on behalf of users to protect customer data and prevent unauthorized activities.
How is Yelp's approach to AI agents different from other websites?
Yelp has taken a more structured approach by allowing AI agents to access its platform only if they pay for its data licensing program. This creates a monetization pathway rather than an outright ban, distinguishing Yelp's strategy from companies like Delta Air Lines that prevent agents from booking tickets entirely.
What does the blocking of AI agents reveal about the gap between demo and deployment for agentic products?
The widespread blocking of AI agents demonstrates that an agent unable to reliably access major websites like Cars.com or Amtrak becomes unreliable and potentially a liability rather than a functional product. This gap shows that founders building agentic products face real-world obstacles that don't exist in controlled demo environments, as major companies with significant traffic and legal resources are actively defending against bot access.
How does Amazon's lawsuit against Perplexity relate to the blocking of AI agents?
Amazon's lawsuit against Perplexity demonstrated that major websites have legal teams ready to defend their traffic and data from unauthorized AI access. This legal precedent suggests that the calculus for companies blocking AI agents won't change even as companies like OpenAI and Meta release more advanced agent products.
