Canary in the Cubicle: What Stanford’s AI Study Tells Us About the Real Future of Work

In partnership with

Today the canaries in the coal mine aren’t songbirds, they’re low-wage customer support reps. A new and first-of-its-kind Stanford study reveals how AI is rewriting the rules of productivity, inequality, and who gets left behind.

Meanwhile, the internet’s filling up with AI-generated junk, but that may offer a lifeline for premium publishers (and a warning for enterprise brands). Let’s break it all down.

Training cutting edge AI? Unlock the data advantage today.

If you’re building or fine-tuning generative AI models, this guide is your shortcut to smarter AI model training. Learn how Shutterstock’s multimodal datasets—grounded in measurable user behavior—can help you reduce legal risk, boost creative diversity, and improve model reliability.

Inside, you’ll uncover why scraped data and aesthetic proxies often fall short—and how to use clustering methods and semantic evaluation to refine your dataset and your outputs. Designed for AI leaders, product teams, and ML engineers, this guide walks through how to identify refinement-worthy data, align with generative preferences, and validate progress with confidence.

Whether you're optimizing alignment, output quality, or time-to-value, this playbook gives you a data advantage. Download the guide and train your models with data built for performance.

Download the guide

The Stanford Study Every Exec Should Read Twice

A joint study by Stanford’s Digital Economy Lab and the National Bureau of Economic Research is the first real-world, longitudinal analysis of how generative AI affects actual workers over time. It focused on a year of data from a Fortune 500 software firm’s 5,000+ customer support agents, half of whom were given access to a genAI tool built from prior successful tickets.

What they found:

• Productivity jumped by 14% for customer support reps using AI.

• Inexperienced agents saw the biggest gains, up to 35%.

• Veteran agents did not see much change.

• Customer sentiment improved when agents used AI.

• But top performers didn’t get better. AI helped the laggards catch up.

This shifts the current narrative. AI might not be a “10x engineer” factory, but it could be the great equalizer for average performers. And that’s a very different workforce dynamic than what many enterprise execs are betting on.

Why this matters for enterprise strategy:

• Upskilling does not equal Reskilling. Helping mid-tier performers reach “good enough” with AI may be a higher ROI than investing in elite performance upgrades.

• Workforce planning needs a remix. If AI benefits your B-players more than your A-team, what does that mean for hiring, incentives, and org structure?

• AI is about redistribution. Leaders need to think carefully about who benefits first, and how to scale that across teams.

5 fast implications:

1. Frontline use cases > shiny pilots. Forget moonshots; support, HR, IT, and sales reps are your low-hanging fruit.

2. Invest in your B-team. The highest ROI may come from lifting average performers, not chasing elite optimization.

3. AI = knowledge transfer. It embeds institutional know-how into every reply, onboarding, or service interaction.

4. Org design needs rethinking. What happens when your C-players become indistinguishable from B+ players?

5. Copilot beats autopilot. The best use of genAI today is assistive, not autonomous. Build tools that guide, not replace.

While Stanford called this study a “canary in the coal mine,” the canary didn’t die. It filed 14% more tickets and got 35% more love from customers.

Let’s not ignore the implications, and instead proactively design for them.

News Roundup

1. Slop is clogging your feeds, and your brand’s reputation is on the line
YouTube and TikTok are being flooded with AI-generated “slop videos”: low-effort, often misleading junk churned out for views and monetization. Some look like real podcasts. Others are fake celebrity interviews. None are good for trust.
Read more →

2. Trusted news brands may finally have an edge again
A new study from the Nieman Lab found that as AI-generated fakes proliferate, readers are turning toward verified sources, even paywalled ones. This may signal a renaissance for reputation.
Read more →

3. NSF expands U.S. AI infrastructure with new data investments
The National Science Foundation is doubling down on building America’s AI backbone with upgraded data systems and open-access tools to help public and private sectors alike. Think of it as the highway system for AI development.
Read more →

TL;DR:

• AI may boost your average performers more than your stars, so plan accordingly.

• Stanford’s long-term study on customer support shows measurable, sustained gains from gen AI tools.

• AI-generated junk content (“slop”) is everywhere, raising new risks for brand safety.

• Trust is back in style. Premium content and verified sources may win in a sea of fakes.

• NSF’s AI infrastructure investment signals serious momentum for public-private partnerships.

AI will change who your top performers are. And if the people in your org getting the biggest boost from AI are the ones you least expected, you better believe that calls for a rethink of everything from org charts to OKRs.

Stay sharp,

Cat Valverde
Founder, Enterprise AI Solutions
Navigating Tomorrow’s Tech Landscape Together