My AI Buddy Made Me Better Than Your Entire Team

The boundary between AI as a tool and AI as a teammate is dissolving before our eyes. Drawing from ground breaking research with 776 Procter & Gamble professionals working on actual product innovation challenges, we unpack how generative AI is fundamentally reshaping workplace collaboration, expertise sharing, and team performance.

TLDR:

• AI-assisted individuals perform as well as traditional two-person teams without AI
• Teams using AI create higher-quality solutions and work 12.7% faster
• AI breaks down knowledge silos between different expertise domains
• Less experienced employees with AI can perform at expert levels
• People working with AI report more positive emotions and less anxiety
• AI serves as more than a writing tool—it's a true thought partner
• Using AI creates more balanced ideas combining both technical and commercial perspectives
• Teams with AI are 9.2 percentage points more likely to develop top-tier solutions
• Despite creating better work, AI users feel less confident about their solutions
• The "cybernetic teammate" concept captures how AI is becoming a collaborative partner rather than just a tool

The findings are striking. Individuals with AI support performed at the same level as traditional two-person teams without AI. Teams augmented with AI not only produced higher quality solutions but did so 12.7% faster while exploring more dimensions of each problem. Most surprisingly, AI demolished the traditional knowledge silos between technical and commercial domains, creating more balanced and integrated solutions where expertise flowed freely.

Perhaps most transformative was the democratizing effect—less experienced employees using AI performed at expert levels, effectively leveling the playing field across experience levels. And contrary to concerns about cold machine interactions, people working with AI reported higher levels of enthusiasm and lower anxiety than conventional teams.

This isn't just about productivity gains. We're witnessing the emergence of what researchers call the "cybernetic teammate"—a collaborative partnership where AI doesn't replace human judgment but expands our creative and analytical capacities. As one person with AI can now match a team's output, organizations may need to reconsider traditional team structures, expertise valuation, and collaboration frameworks.

Have you experienced working with AI as a teammate rather than just a tool? How might this reshape your approach to solving complex problems? Join the conversation about how "cybernetic teamwork" is redefining what's possible in workplace collaboration.

AI is transforming teamwork from a tool into a genuine teammate.

Reading the full paper to understand how these insights might apply to you and your work.

HBR Working paper, "The Cybernetic Teammate: A Field Experiment on Generative AI Reshaping Teamwork and Expertise,"

Link: https://papers.ssrn.com/

Chapters

• 0:00 Introduction to AI and Teamwork
• 1:57 Research Setup: P&G's Innovation Experiment
• 4:19 Performance Findings: AI Boosts Results
• 6:43 Breaking Knowledge Silos with AI
• 8:38 Emotional Impact and AI Collaboration
• 10:27 Future of Work: The Cybernetic Teammate

Transcript

Speaker 1: 0:00

Teamwork right. It's how we've always gotten things done solved problems.

Speaker 2: 0:03

Yeah.

Speaker 1: 0:04

From small groups to like huge collaborations Totally. But with AI changing so fast, you know, it kind of makes you wonder if this technology can really be a teammate.

Speaker 2: 0:15

Oh, absolutely.

Speaker 1: 0:16

Or is it always just going to be a tool? You know what I mean.

Speaker 2: 0:19

Like yeah, it's a fundamental question.

Speaker 1: 0:21

It is For you, the learner, we're diving into this headfirst today.

Speaker 2: 0:25

Exactly. We know you want to get to the heart of the matter quickly.

Speaker 1: 0:28

Right to the good stuff.

Speaker 2: 0:30

So this research is perfect for that. It tackles this question directly.

Speaker 1: 0:33

Well, sure, we're looking at a paper called get this the Cybernetic Teammate. Ooh.

Speaker 2: 0:40

I like that.

Speaker 1: 0:41

It's about how generative AI reshapes teamwork and expertise.

Speaker 2: 0:46

Catchy title.

Speaker 1: 0:47

It is right, and it's from some serious folks Harvard Business School, wharton, even people from Procter Gamble so it's practical too.

Speaker 2: 0:54

Really grounded in the real world.

Speaker 1: 0:56

Yeah, totally, and generative AI, just so you know.

Speaker 2: 0:59

Right. Got to make sure we're all on the same page. It's AI that creates stuff like text, like some tools you might have heard of Exactly Okay.

Speaker 1: 1:07

So our mission today to give you the key insights from this paper.

Speaker 2: 1:11

That's it.

Speaker 1: 1:12

How does AI actually impact? How well teams perform?

Speaker 2: 1:16

You know, performance is crucial.

Speaker 1: 1:18

How does it affect how they share knowledge expertise?

Speaker 2: 1:21

Yeah, yeah.

Speaker 1: 1:22

And get this even, the social dynamics, the feelings involved, the human side of things. Exactly, and what's so cool is that this wasn't just some lab experiment.

Speaker 2: 1:31

Oh no.

Speaker 1: 1:31

They had 776 professionals at P&G working on real product innovation stuff.

Speaker 2: 1:38

Wow, that's a lot of people.

Speaker 1: 1:39

It is right. Real challenges they face every day.

Speaker 2: 1:42

That's great data.

Speaker 1: 1:43

It is. So today we'll see how they set up the experiment, what they learned about performance, expertise and you know, all those social things, and then we'll zoom out and think about what it all means for the future of work. Sound good.

Speaker 2: 1:54

Sounds like a plan.

Speaker 1: 1:55

So how did they even run this experiment?

Speaker 2: 1:57

OK, so they created four different groups to compare different ways of working. First, individuals working alone. No AI help like completely solo.

Speaker 1: 2:07

Got it the lone wolves.

Speaker 2: 2:10

Yeah, exactly Then. Classic human teams, Someone from research and development paired up with someone from the commercial side.

Speaker 1: 2:17

Okay, R&D and commercial.

Speaker 2: 2:19

Yep, and they also didn't have any AI.

Speaker 1: 2:21

No AI for them either.

Speaker 2: 2:23

Nope, Then to see what AI does, they added two more groups.

Speaker 1: 2:27

All right, bring on the AI.

Speaker 2: 2:29

Individuals working on the same tasks, but this time they got an AI tool.

Speaker 1: 2:33

Ah, so they're not alone anymore, they have AI.

Speaker 2: 2:35

Exactly. And the final group, a team, again R&D and commercial, but with AI. They all got training on how to use it.

Speaker 1: 2:43

So everyone using AI was trained up.

Speaker 2: 2:45

Yeah, they made sure everyone was comfortable with the tool.

Speaker 1: 2:48

Makes sense.

Speaker 2: 2:49

For you, the learner. It's important to note these weren't made-up tasks.

Speaker 1: 2:52

Oh, this was real work.

Speaker 2: 2:53

These were experienced P&G people working on actual new product development stuff.

Speaker 1: 2:58

Wow, like what kind of stuff?

Speaker 2: 3:00

Think new ideas for baby care products, feminine care, grooming, oral care you know everyday things.

Speaker 1: 3:06

Wow so like real products people use.

Speaker 2: 3:08

Yep In their actual markets, Europe and the Americas. That's why it's so relevant.

Speaker 1: 3:12

Totally, and they didn't just use any AI, they had a specific tool built on GPT-4 and later GPT-4.

Speaker 2: 3:19

You powerful stuff.

Speaker 1: 3:20

And it was accessed through Microsoft Azure. And that training they got it was a whole hour on using this tool for consumer product innovation.

Speaker 2: 3:28

So not just general AI knowledge, very specific training.

Speaker 1: 3:32

They even got a guide with like suggested prompts to get them started.

Speaker 2: 3:36

Cool like training wheels.

Speaker 1: 3:38

Kind of right. And here's what's really interesting they focused on the very beginning of innovation, what they call the seed stage.

Speaker 2: 3:45

Ah, where it all starts.

Speaker 1: 3:46

And the P&G leadership. They said this phase was crucial for the whole process.

Speaker 2: 3:50

So better starting ideas lead to better products in the end.

Speaker 1: 3:54

That's the thinking. All the teams, with or without AI, collaborated remotely through Microsoft Teams.

Speaker 2: 4:00

Oh, so like how many companies work now.

Speaker 1: 4:01

Exactly, totally normal for a global company like P&G.

Speaker 2: 4:04

Makes sense. So we have the setup. Now let's see what happened when these different groups actually started working. First up performance. How good were their solutions? No-transcript. Well, teamwork does have its perks. Teams without AI. They improve the quality of their ideas. It was a 0.24 standard deviation increase compared to individuals working alone, also without AI.

Speaker 1: 4:30

So two heads are better than one, even without AI.

Speaker 2: 4:33

Yeah, basically confirms that collaboration helps.

Speaker 1: 4:35

Yeah.

Speaker 2: 4:35

But here's where it gets really interesting what happened when they added AI.

Speaker 1: 4:39

Yeah, that's the big question.

Speaker 2: 4:41

The effect was honestly remarkable. Individuals with AI they performed, as well as those human teams without AI.

Speaker 1: 4:48

Wow, so the AI really boosted them.

Speaker 2: 4:50

Big time A 0.37 standard deviation increase compared to those solo individuals. And it gets better the teams using AI they showed even more improvement 0.39 standard deviation increase.

Speaker 1: 5:02

So AI helped both individuals and teams, but teams still did a bit better.

Speaker 2: 5:07

That's the takeaway. It's like AI gave individuals superpowers, letting them play at the team level.

Speaker 1: 5:11

And a team plus AI was like supercharged.

Speaker 2: 5:14

Exactly. It shows how much AI can enhance what a single person can do.

Speaker 1: 5:18

And it wasn't just about quality either.

Speaker 2: 5:19

Right. They also looked at efficiency how long it took to come up with those solutions.

Speaker 1: 5:24

Because AI is all about productivity right.

Speaker 2: 5:26

Totally. People with AI spent 16.4% less time on their solutions.

Speaker 1: 5:30

Wow, almost a fifth of their time saved.

Speaker 2: 5:33

And AI-powered teams. They were 12.7% faster, so less time overall.

Speaker 1: 5:39

That's a lot of time saved. But were they rushing? Were the AI solutions less detailed?

Speaker 2: 5:43

Actually the opposite. The AI-assisted solutions were way longer. Ah, so more detailed it seems people explored more aspects of the problem, wrote more, but still finished faster.

Speaker 1: 5:53

It's kind of amazing. That is pretty cool. So what about those truly groundbreaking ideas, the top-tier ones?

Speaker 2: 5:59

This is where it gets really exciting.

Speaker 1: 6:03

They looked at the top 10% solutions, the very best of the best, like the cream of the crop Exactly.

Speaker 2: 6:07

Teams with AI were 9.2 percentage points more likely to have those top solutions.

Speaker 1: 6:11

So AI not only helps you do better, it helps you reach excellence.

Speaker 2: 6:15

You got it. It's like combining human ingenuity with AI power, but there's a twist.

Speaker 1: 6:19

Oh, what's that?

Speaker 2: 6:20

Even though their solutions were better, the people using AI were less confident about them.

Speaker 1: 6:25

Really.

Speaker 2: 6:26

They were 9.2 percentage points less likely to think they'd be in the top 10%. Like they underestimated themselves.

Speaker 1: 6:32

Maybe they thought the AI did all the work.

Speaker 2: 6:34

Yeah, maybe. It's like they discounted their own contribution.

Speaker 1: 6:37

Interesting. Okay, so AI affects performance, but what about how people share knowledge?

Speaker 2: 6:43

This is where AI really starts breaking down walls. Remember those teams without AI.

Speaker 1: 6:49

Yeah, the R&D and commercial folks.

Speaker 2: 6:50

Well, the commercial people. They liked ideas with a strong commercial focus.

Speaker 1: 6:54

Makes sense. They know that world.

Speaker 2: 6:56

Right and the R&D folks. They lean towards technical solutions.

Speaker 1: 7:00

Also makes sense.

Speaker 2: 7:01

Like their backgrounds, created these invisible silos in their thinking.

Speaker 1: 7:05

We all have our comfort zones, right.

Speaker 2: 7:07

We do, but with AI this difference vanished.

Speaker 1: 7:10

There are more silos.

Speaker 2: 7:12

Both groups, With AI proposed more balanced ideas, both commercial and technical.

Speaker 1: 7:17

So the AI helped them thing outside their boxes.

Speaker 2: 7:20

Exactly, it's like AI became this bridge between different areas of expertise.

Speaker 1: 7:24

That's amazing. So we're seeing these performance gains, but who's contributing? What Is AI changing the team dynamics?

Speaker 2: 7:31

It is definitely. It's like AI, is this facilitator helping people see beyond their own roles.

Speaker 1: 7:35

Hmm, interesting Like a neutral third party.

Speaker 2: 7:38

Exactly, and it had a huge impact on less experienced employees. They called them non-core job individuals.

Speaker 1: 7:45

So people who weren't product development experts.

Speaker 2: 7:47

Right Without AI. These folks struggled when working alone Makes sense. They lacked experience.

Speaker 1: 7:53

Yeah, totally.

Speaker 2: 7:54

Even in teams without AI they didn't prove much. But here's the kicker With AI, these less experienced folks performed as well as the experienced teams.

Speaker 1: 8:03

Seriously, that's incredible.

Speaker 2: 8:06

It really is. Ai leveled the playing field big time. It's like a democratized expertise.

Speaker 1: 8:10

So anyone with AI can contribute like an expert.

Speaker 2: 8:15

That's the potential, and when they looked at how teams collaborated, they saw the same thing. Without AI, the team solutions were either very technical or very commercial what they call a bimodal distribution.

Speaker 1: 8:24

Two extremes.

Speaker 2: 8:25

Yep. But with AI the solutions became more balanced, more uniform, like everyone contributed equally.

Speaker 1: 8:32

So AI made the teamwork more even.

Speaker 2: 8:34

And more inclusive. Now let's talk about the social side, the feelings. This is where things get surprising.

Speaker 1: 8:41

Yeah, how did using AI affect people's moods? I'd imagine it could feel a bit isolating.

Speaker 2: 8:46

You'd think so. Right, but they found the opposite. People who used AI, they felt more positive emotions.

Speaker 1: 8:53

Really Like what.

Speaker 2: 8:54

Excitement, energy, enthusiasm.

Speaker 1: 8:56

Yeah.

Speaker 2: 8:57

And less negative stuff like anxiety and frustration.

Speaker 1: 9:01

So working with a machine actually made them feel better.

Speaker 2: 9:04

That's what the data shows. Their positive feelings were as good as, or even better than, human teams without AI.

Speaker 1: 9:11

Maybe because AI takes some of the pressure off.

Speaker 2: 9:13

Maybe, or maybe the way we interact with AI through language makes it feel more engaging.

Speaker 1: 9:19

It's not just clicking buttons. You're having a conversation.

Speaker 2: 9:21

Exactly and they found a correlation the more people expected to use AI in the future, the more positive they felt during the experiment.

Speaker 1: 9:28

So liking AI makes you want to use it more, which probably makes you like it even more.

Speaker 2: 9:33

Right, it's a positive feedback loop.

Speaker 1: 9:34

Interesting. So AI affects not just the work but also the workers.

Speaker 2: 9:38

For sure. Now, how were people actually using AI? Were they just taking what is spit out?

Speaker 1: 9:43

Yeah, was it just copy and paste?

Speaker 2: 9:45

Well, they looked at how much of the AI generated content ended up in the final solutions.

Speaker 1: 9:50

Okay, so how much did they actually use?

Speaker 2: 9:52

It was a lot Over 75% in many cases. So people were definitely using the AI's output.

Speaker 1: 9:57

But not always right.

Speaker 2: 9:58

Right, some people had zero AI content in their final work.

Speaker 1: 10:02

So they were using AI for brainstorming, not just writing the final product.

Speaker 2: 10:10

Exactly, it was like a thought partner. And here's the key thing Even in those solutions with a lot of AI content, the meaning was closer to human-made solutions.

Speaker 1: 10:16

So they were still adding their own touch, their own understanding.

Speaker 2: 10:18

Totally. Ai wasn't just writing for them, it was collaborating with them.

Speaker 1: 10:22

Humans were still in control.

Speaker 2: 10:23

Absolutely so. What does this all mean for how we work in the future?

Speaker 1: 10:27

Yeah, big picture time.

Speaker 2: 10:29

Well for you, the learner. Think about this. Maybe our idea of team size needs to change.

Speaker 1: 10:36

If one person plus AI, can do the work of two.

Speaker 2: 10:39

Exactly, we could have more flexible, more efficient teams, maybe even individuals, leading projects with AI support.

Speaker 1: 10:46

Wow, that's a big shift.

Speaker 2: 10:47

It is. We could work faster, aim higher, but we need to train people to work with AI.

Speaker 1: 10:53

AI skills are essential now.

Speaker 2: 10:54

Definitely. It's not just about your expertise anymore. It's about how you combine it with AI.

Speaker 1: 10:59

And since AI helps break down those knowledge silos, we need to teach people to think more broadly, like be more well-rounded.

Speaker 2: 11:07

Right, those old departmental boundaries are fading away.

Speaker 1: 11:10

AI is changing the whole game.

Speaker 2: 11:12

It is. It's not just automating tasks, it's changing how we think, how we collaborate.

Speaker 1: 11:17

It's a new way of working.

Speaker 2: 11:18

The researchers call it a cybernetic teammate, and that's spot on.

Speaker 1: 11:22

Yeah, it's not just a tool, it's a partner.

Speaker 2: 11:24

It helps you, it gives you feedback, it even affects your mood.

Speaker 1: 11:27

That term really captures it. So, the learner, what are the big takeaways for you?

Speaker 2: 11:32

AI makes you perform better, sometimes as well as a whole team. It helps you share knowledge, break down those silos and it actually makes you feel better about your work.

Speaker 1: 11:42

It's like that friendly coworker who's always got your back.

Speaker 2: 11:44

And for me, the most amazing thing is that individuals with AI can reach the same level as teams without it and AI-powered teams. They have a way better shot at creating those groundbreaking innovations.

Speaker 1: 11:57

Yeah, that's powerful. So, as you think about your own work, your own teams, ask yourself what doors does this cybernetic teammate open?

Speaker 2: 12:06

What new possibilities are there?

Speaker 1: 12:07

And what does it mean for how humans and AI will work together in the future?

Speaker 2: 12:12

We highly recommend you read the whole paper, Dig deep, think critically and see how these insights apply to you.

Speaker 1: 12:18

Yeah, this research is a peek into the future of teamwork. We're just starting to grasp what it all means.

Speaker 2: 12:23

It's an exciting time.

Speaker 1: 12:24

It is Thanks for joining us on this deep dive, the Learner, until next time.

Speaker 2: 12:27

Keep learning.