AI and the Chief of Staff

How is AI changing the Chief of Staff role?

AI takes over the role's drafting and admin, such as meeting notes, status decks and first drafts, and leaves the Chief of Staff with a week of decisions and judgment calls. In the Chief of Staff Network's AI Readiness Diagnostic (1,200+ senior leaders), 86% use AI every day, but only 7.3% scored as AI-native, with AI running in their workflow on a schedule. One in four Chief of Staff job postings now ask for AI skills, according to our Pathways Report (1,800+ postings).

In a typical week for a Chief of Staff to a CEO, close to a day of drafting can move to AI, and none of the decisions do. The work gets harder. Alex Craig, Chief of Staff at PortSwigger, cut his one-to-one prep from about 15 minutes to 2, and found his days filled with back-to-back decisions. Your principal will notice when a polished draft hides thin thinking.

At a startup under 50 people, the Chief of Staff often carries operations admin, so more of the week moves at once and the role can look thin within a quarter. Plan the replacement work before you automate. Our AI guide covers what to keep.

Where should a Chief of Staff start with AI?

Start with the AI assistant your company has already approved, and give it your context before you give it access. The order that works: company approval and data rules, then a one-page context file, then connections, then one recurring job at a time. Most Chiefs of Staff can get a useful first version running in about a week, without code, once IT has approved it. Approval itself often takes weeks, so send the request early and practice on material with nothing sensitive in it while you wait.

Connect in order of risk: calendar first, then meeting notes, then email, then shared documents. Make a daily brief your first recurring job, and run it by hand for a week before you schedule it. Each correction you make becomes a line in its instructions.

For a one-month plan: in week one, map one process end to end before you touch a tool. In week two, build the context file. In week three, automate one output, such as the morning brief. In week four, measure it and show your principal the before and after. Our AI setup guide has the checklist and templates.

Should a founder hire a Chief of Staff or use an AI chief of staff tool?

It depends on whether your problem is volume or alignment. Volume is too many emails, meetings that need prep and follow-ups that slip. An approved AI assistant plus a good executive assistant gets you most of the way, and an app sold as an "AI chief of staff" can be part of that. Alignment is leaders who pull in different directions, decisions that don't stick and a strategy nobody can repeat back. That needs a person with authority and context.

Both feel the same from the founder's chair: a full inbox and a full calendar. Run a cheap test. Give an approved assistant or app your calendar and inbox for a month, and list what still lands on you. If it is scheduling, chasing and prep, the tool works. If it is disputes between leaders and projects nobody owns, no software will clear it.

Our view: under about 50 people the problem is usually volume. Once a leadership team of five or six runs functions that depend on each other, alignment problems arrive, and an app only makes the calendar feel managed while decisions slip. See AI tools for Chiefs of Staff and hiring a Chief of Staff.

Should a Chief of Staff use Claude or ChatGPT?

Use the one your company licenses and has approved for company data. Whichever it is, the plan matters: the rules for your data, the admin controls and the contract all change with the plan, so keep company work off personal accounts. If your company runs on Microsoft 365 or Google Workspace, its built-in assistant may be the only one allowed with company data.

In the Chief of Staff Network's Chief of AI Report (early 2025), 34% of companies named ChatGPT as their AI tool of choice, against 8% for Microsoft Copilot, 7% for Claude and 6% for Gemini. At companies with 5,000+ employees, Copilot rose to 33% and Gemini to 17%. That mix has moved since, and Claude and ChatGPT now offer much the same building blocks: projects, skills, connectors and scheduled tasks.

If you have a free choice, run the same three real tasks in each for a week and keep the one whose drafts you edit least. Features and menus change often, so check each vendor's help center before you rely on one. Our Claude guide sets up a project and a chief of staff skill, with the ChatGPT equivalent of each step.

When does a company need a Chief AI Officer instead of a Chief of Staff leading AI?

A company needs a Chief AI Officer when AI becomes a product line, a large budget or a regulated risk. Before that, our view is that a company under a few hundred people needs someone with cross-functional context to pick the first real use cases, write the rules and measure what changed. That is a Chief of AI: usually a Chief of Staff or another senior operator who holds AI adoption as a portfolio alongside the rest of the job.

The two hires cost different amounts. Among US postings that list pay, AI leadership roles post a $213k median and Chief of Staff roles a $180k median, according to the Chief of Staff Network's Pathways Report (1,800+ postings).

To decide, write down the three AI decisions the CEO expects the new hire to make in six months. If a Chief of Staff could make them with a clear mandate and budget, start there. Budget is often the gap: in our Chief of AI Report, only 20% of companies had a dedicated AI budget. Our guide to AI and the Chief of Staff covers the mandate to agree with your principal first.

What are the most common mistakes Chiefs of Staff make with AI?

The most expensive mistake is handing AI the thinking. A memo with three balanced options and no view looks finished, and it is useless to a leader who needed a recommendation. Write your recommendation in one sentence before you ask for any draft that carries judgment. The other common mistakes:

• Company work on a personal account, outside company controls.
• Trusting numbers. A wrong total looks as polished as a right one, so check every name, number, date and commitment against the source.
• One long chat for everything, so answers drift. Use a new chat per task inside one project per principal.
• Scheduling a job too early. If you changed more than a line in any of the last three briefs, keep running it by hand.
• A first AI project too safe to matter. Leadership sees no difference and decides AI doesn't work.
• Trialing every tool. Each one costs an approval, a login and a new place your data lives.

Errors about people cost the most, because people remember that the Chief of Staff got it wrong. Our Claude guide lists the first-month fixes, and our prompt library has a prompt to check any draft.

How do you measure whether AI is paying off for a Chief of Staff?

Measure the return before the cost: hours back on a named job, decisions made sooner and commitments that no longer slip, against a baseline you write down before the tool arrives. Time your morning catch-up and one-to-one prep on three normal days, count last week's commitments you had to chase, and list how long each open decision has waited. Check after a month, and cut any tool that can't show a return within a quarter.

Many companies skip this step. In the Chief of Staff Network's Chief of AI Report, 32% of companies hadn't measured AI's impact at all. Of those that had, the most common measure was time saved (40%), well ahead of financial impact (16%) and quality gains (13%).

Count your checking time honestly, and track where the saved hours go. They count when they move into work only you can do, such as preparing the decisions that matter. If they went to clearing more email, the setup made you faster without making you more useful. A result such as "board prep went from five days to two" is one your CEO can repeat to the board. The scorecard is in our AI setup guide.