Control AI Inbox Response Tone and Scope to Prevent Undeliverable Commitments

Problem Statement:

• AI Inbox occasionally offers actions that hosts cannot actually deliver on (e.g., "I'll contact the neighbors right away" about barking dogs)
• When hosts are unavailable, AI makes commitments on their behalf that they cannot fulfill
• This creates awkward situations and potential guest experience issues

Current Workaround:

• Hosts must manually intervene to correct or clarify AI responses
• Hosts need to monitor AI replies closely to prevent overpromising
• Time-consuming to catch and fix inappropriate commitments

Impact:

• Wasted time correcting AI mistakes
• Risk of guest dissatisfaction from unfulfilled promises
• Undermines trust in the AI system
• Creates additional workload for hosts

Proposed Solution:

• Add settings to control the tone and scope of AI responses
• Allow hosts to restrict AI from offering actions they cannot perform
• Enable hosts to configure AI behavior based on their availability
• Provide granular control over what types of commitments AI can make

Unlock:

• AI handles situations autonomously without making commitments on your behalf
• Hosts have confidence that AI won't overpromise or offer undeliverable solutions
• Reduced need for manual intervention and correction
• Smoother guest interactions even when hosts are unavailable