# Google Has More Data Than Almost Anyone. So Why Is It Bidding $10 Million On Old Emails?

## Executive summary

Google's $10 million bid for Spirit Airlines' work records highlights a critical tension in AI development: the difference between the *performance* of work (the record of messages, meetings, and documents) and the *actual value* of knowledge work. The speaker argues that current data monetization models undervalue complex human judgment, which is often messy, unrecorded, and context-dependent. For AI agents to be truly useful, the focus must shift from merely consuming data archives to structuring data in ways that enable collaboration and define repeatable, high-leverage processes.

## Key takeaways

- The Value Gap in Work Data: AI labs are purchasing massive archives (emails, Slack, transcripts) to train agents to perform knowledge work. However, much of the value humans create—such as finding the right information or getting permission to schedule a meeting—is not captured by these records. The market is currently buying the 'performance' of work, not the 'work' itself.
- Verifiable vs. Knowledge Work: The speaker distinguishes between 'verifiable work' (binary outcomes, like code that runs or fails) and 'knowledge work' (ambiguous, context-dependent tasks, like resolving an invoice discrepancy). The latter is difficult to prove and is what most human value consists of.
- The Problem of Defining Success for Agents: When training agents, the success criteria must be clearly defined. The speaker argues that relying solely on the closure of a ticket or the completion of a document (the performance) fails to capture the valuable intervention (the actual work) that led to the outcome.
- Future Focus: Structuring Data for Collaboration: The most valuable work for agents is not what is sold in data archives, but work that enables agents to collaborate with humans by structuring data (e.g., using good markdown files) and defining a clear 'record of truth' within a system.

## Technical details

- Data Sources and Scope: The data being purchased includes emails, Slack messages, Zoom call transcripts, business documents, and code. The Spirit Airlines deal involved a massive inventory of 100 million emails and half a billion Teams records.
- Agentic Capability and Training: Buyers (like Google) are aiming to build 'agentic capability' that can be resold. Companies like Merkor are addressing the difficulty of proving knowledge work by building training environments (e.g., Deep Tune) that combine software, assignments, and success checks.
- System Integration and Workflow: Real-world messes often involve multiple systems (e.g., System A for invoices, System B for purchase orders) and require human 'tribal knowledge' to resolve discrepancies, which is the core of the value that data archives fail to capture.

## Practical implications

- Workers should be empowered to articulate the true value of their work, focusing on how they drive revenue or cut costs, rather than relying on the volume of messages or documents.
- Companies should prioritize structuring data in a way that enables agents to collaborate with humans, rather than simply selling raw data archives.
- When considering data sales, organizations must understand the scope of the proposed use and ensure that the data is not devalued over time by the agent's limited understanding of the 'mess' of the business.

## Topics

AI Agents, Knowledge Work, Data Monetization, Build Engineering, Data Governance, System Integration, Spirit Airlines, Merkor, Deep Tune, GP.AI

Source: https://www.youtube.com/watch?v=wep2EQ9Y_Tc