The data handling blindspot that won't self-correct

The gap that hides from self-selection

Organizations frequently let employees choose their own GenAI training topics. The assumption is reasonable: people know where they struggle and will select accordingly.

The data says otherwise — at least for data handling.

What the data shows

Among respondents whose weakest SJT dimension is Data Handling (n=46), the overwhelming majority prefer training on something else entirely.

93% of data-handling-weak respondents do not request Data Handling training. Instead, they gravitate toward Prompting (37%) and Workflow (30%). Data Handling ranks dead last at 7% of selections — despite being their most critical gap.

Why it matters

This is a self-correcting problem in theory but not in practice. If the people who most need data handling skills never select that training, the gap persists indefinitely. Meanwhile, they continue using GenAI tools daily — pasting sensitive information, sharing internal documents, and operating without the mental model for what is and is not safe to input.

The blindspot exists because data handling errors do not feel like errors to the people making them. Pasting a spreadsheet into a chatbot feels like using a tool, not leaking data. Without objective identification, this gap will never surface through voluntary training programmes. The GenAI Capability Pulse surfaces exactly this kind of hidden gap through scenario-based assessment.

What to do about it

• Make Data Handling training mandatory for those who need it: Assign it based on diagnostic results, not self-selection. The people who need it most will never choose it.
• Use objective targeting: A short scenario-based assessment can identify who is weak on data handling and route them to the right module automatically.
• Add guardrails alongside training: Clear do/don't guidance, redaction checklists, and input review steps reduce reliance on awareness alone.

If you let people choose their own training, the biggest blindspot stays invisible. Assign Data Handling modules based on diagnostics, not preferences.

These findings are drawn from the GenAI Capability Pulse — a scenario-based assessment that measures what non-technical teams actually do with GenAI, not what they think they can do. If your organization is scaling GenAI adoption, start with a baseline.

Source: AGASI GenAI Capability Pulse. Data Handling-weak subgroup n=46 (respondents whose lowest SJT dimension was Data Handling). Training preferences base: N=147.