What to do with the hours AI saves you at work

A lot of people now finish in an afternoon what used to take them two days. Ask where the rest of the week went and most of them can’t tell you. The hours didn’t disappear. They got absorbed by more email, more meetings and a slightly bigger version of the same job, and nobody chose that on purpose.

Most people save time and then lose track of it

BCG’s AI at Work 2026 survey, published in June and based on nearly 12,000 employees, managers and leaders across more than a dozen markets, found that 74% of frontline staff now use AI every day or a few times a week. That’s up 23 points from the year before. Among those regular users, 42% say they save about eight hours a week, which is a full day.

BCG also reports that 66% get limited or no guidance on what to do with the freed-up hours, and more than half say they aren’t putting that time toward more strategic tasks. So the typical outcome is someone who gains an extra day each week and spends it doing more of what they were already doing.

That’s a strange result for a technology sold on the promise of freeing people up for better things. It makes sense once you notice that the tool handles the saving and leaves the spending to you.

Unplanned hours get absorbed

Time without a plan fills up fast. A separate study published in HBR by Aruna Ranganathan and Xingqi Maggie Ye, who spent eight months inside a US tech company, found that AI tended to intensify the job for the people using it. They moved faster, took on a wider scope of tasks and let the workday stretch into more hours, often without anyone asking them to. The researchers warned that this pace can slide into fatigue, burnout and weaker decisions.

So the default has a direction, and it points toward more volume. If you don’t decide what the hours are for, your inbox, your team’s backlog and whoever messages you most often will choose for you. The first thing to go is usually the slow learning that would have made you better at the job a year from now.

Spend part of it getting better at the hard parts

The best use of saved time for most early-career people is building the skills AI is worst at covering for. Look at the tasks where you most often have to fix the model’s output, or where you accept its answer because you can’t check it, since that’s where your gap is. (Our piece on where AI is strong and where it fails goes deeper on how to spot it.)

Make it concrete. An analyst who uses AI to draft SQL could spend an hour a week reading the queries line by line and working out why some of the joins go wrong. A marketer who gets AI summaries of customer calls could read three full transcripts a week and compare them against what the summary kept. The point is to build the understanding that lets you catch errors, which is the part employers pay for.

Put a fixed block on your calendar with a name on it, say two hours on Tuesday morning, and protect it like a meeting. Don’t try to claim all eight hours for learning. Reclaiming two or three consistently is a big win.

Move toward tasks that need your judgment

AI mostly speeds up drafting, summarizing and assembling. What stays slow is deciding what’s worth doing, talking to the person who has the problem and checking whether the finished thing solves it. Those are the tasks that build a career, and saved time is your chance to get closer to them.

In practice that could mean offering to review a teammate’s AI-assisted draft before it goes out, asking to sit in on the call where a tradeoff gets settled, or writing the recommendation at the end of an analysis rather than stopping at the summary. Each one puts you in the room where someone has to make a call and defend it.

A small project with your name on it helps too. Pick something that fixes a nagging problem, like an automated weekly report, a cleaned-up onboarding guide or a prototype of a tool your team keeps asking for. Choose something with a clear before and after you can point to. Saved time is invisible by default, and if you finish faster and nobody knows, the usual reward is a longer to-do list.

How to raise it with your manager

A lot of people hide their AI time savings because they worry about getting more tasks piled on or looking replaceable. That’s understandable, and it usually backfires, because the hours get absorbed anyway and you get no credit for them. Given the BCG numbers, your manager probably has no plan for this either, which means you can bring one.

Keep the conversation short and specific. Something like, “AI is saving me about five hours a week on the weekly reporting. I’d like to put two of those toward learning our pricing model properly and building a dashboard the sales team has asked for. Can we check in on how it’s going in a month?” Bring a rough before and after (how long the task took then and how long it takes now), tie the proposal to something the team already cares about, and offer to show others the method so the gain belongs to the whole group.

If your manager says the hours should go to more of the same, that’s useful information too. It tells you how this team values growth, and that’s worth knowing when you think about your next move.

How you measured the time you saved and what you did with it also makes a strong answer when a hiring manager asks about your AI use. Openskill runs the most realistic AI voice interviews, with follow-up questions based on your answers, so you can practice turning your before-and-after numbers into a clear story before that question comes up. Its feedback report breaks the session down question by question, so you can see which answers were specific and which drifted.

Start with one block this week

You don’t need a grand plan. For one week, jot down each time AI saves you a meaningful chunk of time and roughly how much. At the end of the week, pick one place to put some of that time and book it. Then bring the numbers to your next one-on-one.

The hours AI frees up are some of the most flexible time you’ll get early in a career. Treat them as yours to direct, and a year from now you’ll have something to show for them beyond a busier calendar.

Use some of the hours on your next move

If some of your saved time goes to your career, Jobs on openskill searches real, current openings from employers’ own career pages and job boards. Open a role and click “Prepare for this interview” to practice turning your before-and-after numbers into a story for that exact job. You apply on the employer’s own site. openskill is a complete career platform: search millions of jobs with an AI-powered match score, practice any interview in one click, get scored feedback and track your applications.