Sears Home Services Exposed Nearly 3.7 Million Customer Records Through AI Chatbot Databases
Sears Home Services left nearly 3.7 million customer service records unsecured in AI chatbot and scheduling databases, exposing data spanning 2024 to 2026. The exposure included text files, spreadsheets, audio recordings, and chat logs tied to the company's Samantha customer-facing chatbot and KAIros scheduling platforms.
The leaked data contained 2.1 million text files with scheduling conversations, over 200,000 spreadsheet logs, nearly 1.4 million audio recordings of customer calls, and more than 54,000 complete chat logs. Customer information included names, email addresses, phone numbers, physical addresses, and details about products, services, deliveries, and repairs.
Sears owner Transformco restricted access to the databases after the exposure was discovered. The company advised affected customers to monitor for potential misuse of their information.
What This Means for Customer Support Teams
The incident highlights a core risk for customer support operations: AI systems that handle sensitive customer interactions require the same security controls as any database storing personal information. Many support teams rely on chatbots and call-recording platforms without verifying how those systems store and protect data.
If your organization uses AI chatbots or automated scheduling tools, verify with your vendors how customer data is stored, who has access, and what encryption is in place. Request documentation of security audits and incident response procedures.
Audio recordings of support calls deserve particular attention. These files contain not just customer complaints but payment methods, account numbers, and other sensitive details spoken during conversations. Unsecured audio storage creates exposure that extends beyond typical data breaches.
A Recurring Problem
This is Transformco's second major data incident in five years. The company disclosed a cyberattack affecting current and former employee data in 2019. Repeated breaches suggest systemic gaps in security oversight, not isolated incidents.
For customer support professionals, the pattern matters. Vendors with histories of security failures are statistically more likely to have future incidents. When evaluating AI tools or outsourced support platforms, ask about previous breaches and how the vendor addressed them.
Learn more about AI for Customer Support and how to implement these tools responsibly, or explore resources for AI in call center operations.
Your membership also unlocks:
