In a hands-on test, Claude showed Resy availability but could not book, and ChatGPT's agent clicked through OpenTable but set the wrong party size.
In a hands-on test, Claude showed Resy availability but could not book, and ChatGPT's agent clicked through OpenTable but set the wrong party size.
The source summarizes a PCWorld hands-on test from April 2026. Claude pulled up a live Resy availability grid for The Smith. ChatGPT's agent mode spent about five minutes clicking through OpenTable's web interface. Neither could finish, and each handed the task back to the person.
Claude declined to complete the booking. ChatGPT's agent selected 2 people instead of 3 and asked the user to take over. The article then covers the August 2026 in-chat booking launches that removed this gap.
Check the party size yourself: the agent's dropdown slip was a real failure, and the article says error rates for these systems are not published.
Check OpenTable and Resy for a table at [restaurant or cuisine and city] for [number] people on [date] around [time]. List the available slots you find. Before you book, pause and show me the date, time and party size you have selected so I can verify them, and only proceed after I approve.
Score every demo request against your ICP, update HubSpot and ping the right rep in Slack within a minute of the form submit.
Every new Zendesk ticket gets a category, priority and sentiment tag plus a draft reply in an internal note before an agent opens it.
Copilot and Claude review each PR for bugs, missing tests and risky changes before a human looks, so reviewers can focus on design.