Ideas
ProductiviTree: The Human Side of Productivity by Santiago Tacoronte
AI at Work: The Hidden Work Your Rollout Will Break First, with David Dean
Hosted by Santiago Tacoronte
Your process documents describe one company. The workarounds, the chasing and the four-day approvals describe another. Point AI at the second one and it copies the mess at speed.
Original episode
Listen or watch
Conversation summary
Process documents describe the official version of a company. Workarounds, chasing, and four-day approvals reveal another version: the unwritten system people rely on to keep work moving. David Dean draws on 20 years in enterprise technology, including 10 years leading AI, automation, and data programs across healthcare, insurance, hospitality, and regulated industries, to explain why AI pointed at that hidden system can copy its problems at speed.
The conversation explores the gap between the official contract of a job and the unwritten contract that makes it work. David describes the inbox as a record of how an organization behaves and calls 79 percent of the workforce AI survivalists: people who are neither for AI nor against it, but are trying to do their jobs while figuring out what AI means for them.
Listeners will learn to name undocumented behaviors their roles depend on, recognize an at-risk AI rollout by asking who was consulted, and start a work diary with nothing more than the notes app on a phone. The practical starting point is to understand the work before choosing a tool—and keep consequential decisions with people who understand lived experience and lived consequence.
Listen before a tool is rolled out across your team, especially if nobody has asked the people who do the work.
Key insights
- Every job description ends with “other duties apply,” and that line is where much of the real work lives.
- Workarounds can begin as a need for control or a way to survive the week. They often persist after the conditions that created them have changed.
- David calls 79 percent of workers AI survivalists: people who are neither for nor against AI, but are focused on doing their jobs.
- AI behaves less like deterministic software with fixed inputs and outputs and more like people: its open-ended inputs and outputs bring opportunity along with uncertainty.
- David describes the frontier of AI as self-discovery. People need to document what they actually do and where work challenges them before deciding which behavior a tool should augment or replace.
- AI copies and industrializes human behavior, including dysfunction, which helps explain why some projects stall.
- Adoption takes time and effort. Employees need room to decide what is useful for them and to build trust with AI instead of having a solution imposed on them.
- Lived experience and lived consequence remain human, so leaders retain responsibility for risk decisions.
- Everyday acceptance of AI output is itself a decision with risk and return. The higher the downstream consequence, the more people need to stay involved.
- An early warning sign for an AI project is that nobody spoke with the people who will use the tool.
- A dictated work diary on a phone can make invisible behavior visible enough for a team to examine. AI can help find signals in that account, while the person interprets them and chooses what to do.
Questions answered
What is the unwritten contract of a job?
The undocumented behaviors, judgment, workarounds, and informal coordination people rely on to get the work done beyond the job description and formal process.
How can you spot an AI rollout at risk?
Ask who was consulted. If the people who will use the tool were not asked how the work actually happens, the project may be based on an incomplete picture.
How can someone start documenting hidden work?
Use a phone’s notes app and dictate a work diary: what you do, how you do it, where work gets stuck, and what you rely on to keep it moving.
Conversation guide
The official job and the work behind it
Why “other duties apply” points to the undocumented behaviors that keep a role functioning.
Workarounds and survival
How people adapt to get through the week, and why those adaptations can outlive their original purpose.
The 79% AI survivalists
A view of workers who are neither pro-AI nor anti-AI and are trying to understand what the technology means for their jobs.
AI mirrors organizational behavior
Why automating a process without understanding it can scale the dysfunction already present.
Who was asked before rollout?
The question that can expose a missing view of the people who will use and live with a new tool.
A work diary anyone can start
How voice notes can capture everyday workarounds and make hidden behavior available for discussion.
Timecodes are omitted where the supplied transcript did not contain verifiable timing.
Edited transcript excerpts
Speaker labels and punctuation were reconstructed from automated transcription and edited for readability.
“There’s always a line item at the very bottom of those job descriptions, which is “other duties apply.” And really, to me, that’s more of the job description.”
“AI is just as chaotic as humans. So you have an infinite number of inputs and an infinite number of outputs that potentially can happen.”
“The frontier of AI is not tooling. It is self-discovery.”
“They are not against AI. They are not for AI. They are just people doing their jobs.”
“The first sign an AI project is going to fail: did you have any conversation with the frontline people that are going to be directly affected by the outcome of that AI project?”
“AI is never going to understand lived experience and lived consequence. That is a human trait.”
“They should put time for self-discovery. They should write a work diary.”
Keep exploring
Follow the idea without losing your place
Start with the closest connections. Open the additional links only when you want to go further.
Reading
Related articles
Source
Original episode
Continue the conversation
Explore the book behind these ideas, bring the conversation to an audience, or learn more about David’s work.