Companies rethink incentives for employees' AI usage

**SPEAKER_1** (0:00)
Are all batteries the same?
That's like asking if all soccer players are the same. Take Messi, the most decorated player ever. Is there any other player who has achieved that? No, just him. Now take Duracell. Is there any other battery with power boost ingredients inside? No, just Duracell. Remember, goats only trust goats, because they're built different. And Messi only trusts Duracell.

**SPEAKER_2** (0:31)
This episode is brought to you by Google Health. Stop chasing someone else's definition of health. What matters is what's healthy for you. Google Health offers a new kind of coach, built with Gemini for effortless tracking, sleep insights and holistic coaching tailored to you. Visit googlestore.com to learn more and start a new relationship with your health. Requires Google account, Google Health app, internet and Google Health premium subscription. Features subject to change. Availability and results vary. Not intended for medical purposes. Works independently of Gemini apps. Check responses for accuracy.

**SPEAKER_4** (1:06)
Take your flexibility beyond the mat with PayPal. Pay nothing at checkout, then enjoy a flexible monthly payment plan that works for you with no sign up or late fees. Find yours in and an easy way to pay with PayPal. Download the PayPal app to get started. Subject to approval, pay monthly consumer loans made by WebBank. Available through PayPaling, NMLS910457.
Learn more at paypal.com/paymonthly.

**SPEAKER_5** (1:30)
A law firm offered a one million pound bonus to staff just for logging a million AI prompts, explicitly admitting they did not care what the staff actually did with the tool.

**SPEAKER_3** (1:41)
Yeah, I mean, humans will always optimize for the exact metric you hand them. It really doesn't matter how absurd the resulting behavior gets.

**SPEAKER_5** (1:49)
Right, you set a target and people just figure out the shortest path to hit it, even if that path completely undermines whatever you were originally trying to accomplish.

**SPEAKER_3** (1:57)
You see the exact same behavior over at Amazon, actually.

**SPEAKER_5** (2:00)
Oh, yeah, they had to pull their internal leaderboards down, right? Because employees were simply automating completely pointless tasks, just to climb the internal rankings.

**SPEAKER_3** (2:08)
Just generating volume to have volume.

**SPEAKER_5** (2:10)
Yeah, padding their stats with garbage data, just so their name would appear at the top of a companywide dashboard.

**SPEAKER_3** (2:16)
So how did the corporate mandate to adopt this technology become completely detached from producing actual value?

**SPEAKER_5** (2:23)
Well, think about the initial panic across the corporate world to force tech adoption over the last year. Every executive wanted to show they were ahead of the curve.

**SPEAKER_3** (2:33)
Right. Companies pushed incredibly hard for tool uptake using basically whatever levers they had available.

**SPEAKER_5** (2:40)
They threw financial incentives at people. They threw veiled threats about falling behind.

**SPEAKER_3** (2:44)
And they relied entirely on raw usage metrics, like basic token counts, prompt leaderboards, just these very crude measurements of activity.

**SPEAKER_5** (2:54)
If you're not subscribed yet, take a second and hit follow on whatever podcast app you're using. It helps us keep making this. We appreciate you being here. And for a different social media experience, no bots, no billionaires, check out pushdup.com, dshdup.com.
There's a link in the show notes.

**SPEAKER_3** (3:12)
Because measuring visible metrics is easy. You can just put them on a dashboard, show that dashboard to a board of directors, and claim progress.

**SPEAKER_5** (3:19)
But it creates a complete illusion of actual adoption.

**SPEAKER_3** (3:22)
Yeah, there is a psychological aspect to this that leadership often misses. When you optimize for visible metrics rather than genuine behavioral change, the adoption looks highly successful on paper.

**SPEAKER_5** (3:33)
Right, the graphs go up and to the right.

**SPEAKER_3** (3:35)
Exactly. But it totally falls apart in practical application.
Employees are rational actors within the system you build for them.

**SPEAKER_5** (3:43)
So if the system rewards raw output without qualitative assessment, they will just give you raw output. Right.

**SPEAKER_3** (3:50)
They will not go through the cognitive friction of genuinely integrating a new tool into their workflow.

**SPEAKER_5** (3:56)
They just find the easiest way to make the graph go up. It is exactly like giving a drummer an award for the total number of beats they play during a song, entirely ignoring whether those beats actually serve the music.

**SPEAKER_3** (4:10)
That is a great analogy. Imagine being a recording studio and telling a session drummer they get paid 10 cents for every single time they hit a drum or a cymbal during a three-minute pop song.

**SPEAKER_5** (4:19)
What happens then? They are not going to play a rhythm.

**SPEAKER_3** (4:21)
No, they are going to play a continuous deafening drum roll for three minutes straight.

**SPEAKER_5** (4:26)