Copy article

OpenAI Report Reveals AI Power Users Leave Colleagues Six Times Behind

ended 12. December 2025

OpenAI's latest enterprise data, covering over a million business customers, reveals workers operating at 6x productivity differences despite identical AI access. The gap shows up everywhere: customer service teams, HR departments, finance analysts, marketing writers.

Part of the report looked at data scientists. The heaviest users engaged AI tools 16 times more than their median colleagues: people doing the exact same job, in the same building, with the same software. Among all monthly active users, 19% have never even tried the data analysis features their bosses are paying for.

Workers who apply AI to seven different types of tasks report saving five times more time than those using it for just four. The people saving over 10 hours weekly consume 8x more computing power than colleagues reporting no time savings at all.

We'd like your views:

  • Is this the new workplace divide between AI experimenters and permission-waiters?
  • What happens to team morale and performance when some members quietly become 6x more productive while others wait for official guidance?
  • Should managers track who's adapting versus who's stuck? How could that tracking work?
  • Are we watching permanent career gaps open in real-time between frontier workers and their reluctant peers?
  • Are the workers who rely on AI outputs a productivity poster-child or a risk given accuracy concerns?
  • Should companies be monitoring the uptake of new AI tools within their staff and encouraging sharing wins and cautions?

5 responses from the Newspage community

Copy all

Star Quote
Copy

Colette Mason
AI Ethics Consultant at Clever Clogs AI
The frontier workers aren't just working faster, they're building a critical new skill their colleagues haven't clocked yet: quality assurance alongside AI augmentation.

AI hovers around 70% accuracy, meaning their real jobs become curators, not consumers of AI output. The winners aren't the ones generating the most output. They're the ones who've learned to spot the 30% that's wrong before it's too late.

But perhaps it's not a skills gap, but more of a conscience gap. Maybe they're watching others in the department wave through AI slop without checking, and they've decided that's not a productivity win, it's a reputation risk.

Real AI maturity isn't about message volume. It's about building review loops that don't bottleneck your humans. It's about knowing when to trust the machine and when to bin its suggestions. That's the skill gap that actually matters, and that's seldom in anyone's usage statistics.
Copy

Editor
Anita Wright
Chartered Financial Planner at Ribble Wealth Management
If a handful of people quietly become six times more productive, you create a strange mix of envy, anxiety and unrealistic expectations. Targets start to move, but no one explains that the underlying process has changed. You also risk quality drift if the heavy users are not being asked how they check outputs, and the team simply assumes that anything neatly written by AI must be correct.
On careers, it is hard to avoid the conclusion that gaps are opening in real time. Those who learn to weave AI into many different tasks are building a meta-skill that will compound; those who ignore it may drift towards the “I never really got the hang of Excel” camp. Treated as a junior assistant whose work is checked, AI is a genuine productivity boost. Treated as an oracle, it is a risk. That is exactly why firms should monitor uptake, capture examples of wins and near misses, and make sharing both a normal part of team culture.
Copy

Editor
Kate Underwood
Founder & Chief People Strategist at Kate Underwood HR and Training
This is 100% the new workplace divide: the AI tinkerers and the “I’ll wait for a policy” crew. Same tools, same logins, completely different futures. If some of your team are quietly 6x more productive while others are still staring at the login screen, morale will crack. The grafters get resentful, the avoiders feel exposed, and suddenly you’ve got a two-tier team without ever changing a job title. Managers absolutely should track who is adapting, but not with a big stick. This is about curiosity and willingness to play, not age or pay grade. Ignore it and you’ll hard-bake permanent career gaps into your organisation. Heavy AI users are both your secret weapon and a walking risk if no one checks the output. So yes, monitor uptake, talk about it openly, and get people sharing wins and horror stories. Stop treating AI like magic and start treating it like any other tool humans need training and boundaries to use.
Copy

Justin Brandon
Director at Scape Interiors West
One of our team members is really enthusiastic about using AI and constantly learning new tools. I’ve noticed her productivity shoot up. She can complete tasks such as image edits, quick videos, and market research in a fraction of the time. She’s always happy to show the rest of us what she’s learned, and it’s been great for keeping the team moving forward.

That said, not everyone feels the same way. Another colleague spent years building up his graphic design skills in the college, and AI has taken away a lot of the freelance work he used to rely on. For him, it’s not exciting, it feels like the ground is shifting under his feet.

Because of this mix of reactions, we’ve had to share AI learning with a bit more care and understanding. It’s important to recognise the positives, but also to acknowledge the real worries some people have, and make sure everyone feels respected as we adapt together.
Copy

Rohit Parmar-Mistry
Founder at Pattrn Data
The stat that data scientists ignore analytics features is baffling. It’s like a Starfleet engineer ignoring the warp drive to shovel coal. If experts ignore the tools, it suggests the software fails against the messy reality of real-world data.

I’m sceptical of this '6x productivity' victory lap. We obsess over speed but forget direction. Blindly trusting the ship's computer usually invites disaster. Giving a bad driver a faster car doesn't make them a racer; they just crash at higher speeds.

The report ignores the vital question: were the outputs actually better? Or did we just generate six times more average work? Saving 10 hours is meaningless if the result is buggy code or hallucinated data that requires human cleanup anyway.

We risk a dangerous divide between those who master their craft and those who let the 'black box' do the thinking. Managers: Don't mistake burning computing power for creating genuine business value.