Fed Economists Ask Companies If AI Is Killing Jobs, Get Predictable Bullshit Answers

Here's what actually happened: The New York Fed called up a bunch of companies and asked "Hey, are you using AI to fire people?" And companies said "Oh no, we'd never! We're just using it to help our wonderful employees be more productive!"

Translation: We're not firing people right now because that would be terrible PR during a government survey. But ask us again after the next recession.

What "40% of service firms use AI" actually means:

• Marketing teams use ChatGPT to write blog posts instead of hiring copywriters
• Customer service routes calls through AI chatbots before human agents
• HR uses AI screening tools to reject 80% of resumes without human review
• Managers use AI to write performance reviews (yes, this is happening)

The "retraining" lie everyone's telling:

Every company says they're "retraining" workers to work alongside AI. Bullshit. Here's what's really happening:

1. Phase 1 (now): "AI is just helping Sarah be more productive!"
2. Phase 2 (next downturn): "We discovered we can do Sarah's job with AI and an intern."
3. Phase 3 (inevitable): "Sarah has been restructured to pursue other opportunities."

Real talk from someone who's been through this:

I watched this exact playbook during the last wave of automation. First it was "workflow optimization." Then "process improvement." Then "right-sizing for efficiency." Same corporate speak, same result: fewer jobs.

Why companies aren't firing people yet:

• AI still fucks up constantly and needs human oversight
• They haven't figured out the legal liability when AI screws up customer accounts
• Management is scared of the productivity cliff when they fire everyone who knows how things actually work
• They're waiting for competitors to go first so they don't take the PR hit

The brutal truth from the survey:

Even these diplomatically-phrased Fed questions got companies to admit they're planning "more significant layoffs and scaled-back hiring" as AI gets better. They just said it in economist speak so it sounds less dystopian.

What workers are actually experiencing:

• Being asked to "train the AI" that will replace them (without being told that's what they're doing)
• Having their performance measured against AI metrics they can never match
• Getting layoff warnings disguised as "skill development opportunities"
• Watching entry-level positions disappear because "AI can handle the simple stuff"
• Dealing with increased surveillance and productivity monitoring justified as "AI optimization"
• Fighting algorithmic bias in performance reviews that they can't appeal or understand

The manufacturing vs service split makes perfect sense:

Service jobs are easier to automate because they're mostly information work. You can replace a customer service rep with a chatbot. You can't replace a plumber with GPT-4.

What this study actually proves: Companies know exactly how to give non-answers to government surveys. Wait until the economy softens and suddenly all this "retraining" becomes "optimization" real fast.

The Fed economists got played, and anyone who's worked in corporate America for five minutes could have told them this would happen. Similar studies in Europe and Asia show the same pattern: companies say one thing to researchers and do another when quarterly earnings calls come around.

Here's the real story: Government economists asked companies "Are you firing people because of AI?" and got the same bullshit answers every corporate communications team has been trained to give since the first downsizing in the 1980s.

The three-phase corporate playbook (that I've personally witnessed at three different companies):

1. Phase 1 - "We're investing in our people" - Announce retraining programs, skills development initiatives, "upskilling partnerships" with local colleges
2. Phase 2 - "Difficult market conditions" - Six months later, announce layoffs due to "economic headwinds" and "strategic realignment" (never mention AI)
3. Phase 3 - "Operational efficiency" - Replace laid-off workers with AI tools, contractors, or offshore teams

The Fed economists got played, and anyone who's worked in corporate America for five minutes could have told them this would happen. Companies have been perfecting the art of saying one thing to regulators while doing the exact opposite for decades.

Real talk from someone who's been through this: The "retraining" programs are usually garbage. When IBM announced their "SkillsBuild" initiative in 2019, they laid off 20,000+ people over the next two years while claiming they were "transforming their workforce." The retraining was real - they retrained people right out the door.

What's actually happening in the service sector: Companies are using AI to automate the easy parts of jobs, then expecting remaining workers to handle everything else for the same pay. Customer service reps now manage AI chatbots that handle 80% of inquiries, but when the bot fucks up (and it always does), humans get blamed for the failures.

The manufacturing reality check: Service companies can hide layoffs behind "digital transformation" and "customer experience optimization." Manufacturing companies can't hide empty factory floors. That's why service sector adoption looks higher - it's easier to bullshit about.

Here's what policymakers should actually be asking:

• How many full-time positions were eliminated vs. how many people got "retrained" into lower-paying gig work?
• What percentage of "AI augmentation" programs resulted in headcount reduction within 12 months?
• How many companies moved customer service offshore after implementing AI chat systems?
• What's the real retention rate for workers who went through corporate "upskilling" programs?

The job displacement is already happening, it's just spread across multiple budget line items so no single department looks like the bad guy. Goldman Sachs estimates 300 million jobs will be affected globally, but companies will never admit it in surveys.

Policy advice that might actually work: Stop asking companies what they're planning to do and start measuring what they're actually doing. Track headcount changes by job function, not by company PR statements. Follow the money - when companies claim they're "investing in workers" but their operational expenses are declining, someone's getting fired.

The timing matters: This survey happened during a labor shortage. Companies were desperate to keep workers and avoid bad PR. Wait until the next recession when they don't need to pretend to care about employee retention - then ask these questions again.