All Episodes

September 16, 2025 10 mins

Elon Musk’s xAI has abruptly laid off around 500 data annotators responsible for training its chatbot, Grok. These workers were central to shaping the model’s tone, accuracy, and behavior. Their sudden removal signals a major shift toward automation and synthetic data, raising questions about Grok’s future quality and alignment. In this episode, we break down what the layoffs reveal about xAI’s strategy, how they impact Grok’s development, and why removing human feedback now could come at a cost.

Mark as Played
Transcript

Episode Transcript

Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:01):
Hey everybody, welcome back to the Elon Musk podcast.
This is a show where we discuss the critical crossroads, the
shape, SpaceX, Tesla X, the Boring Company and Neurolink.
I'm your host Will Walden. XAI laid off nearly its entire
data annotation team last week, cutting around 500 contractors

(00:24):
who had been responsible for training Grok, which is the
chatbot that is used on X slash Twitter.
And these workers made-up the bulk of the workforce behind the
scenes. They labeled data the curated
prompts for Grok and fine-tuned Grok's responses across a wide

(00:45):
range of tasks. Now, what does XAI plan on
doing? Are they going to replace these
people with some other people, or are they going to allow Grok
to moderate itself now? X AI didn't publicly acknowledge
the layoffs, though workers received notice without advance
warning, and most of them were contractors based outside the US

(01:07):
working through third party staffing firms.
The team had been assembled overthe past year to accelerate
Grok's progress from an early stage model into a commercial
project. Now cutting them so suddenly
signals a shift in strategy thatX AI hasn't explained to anybody
yet. And the timing matters for this

(01:28):
because X AI only launched Grok 2 in March and has already begun
teasing more versions of Grok inthe future.
The company claims that Grok 3 is training on a custom compute
cluster with 100,000 NVIDIA H1 hundreds in.
Musk has promised Grok will surpass opening eyes ChatGPT
four and five, and will also match GPT 6 when it comes out.

(01:52):
And there they've limited 500 trainers during this transition.
And it forces the question of how Grok will be improved
without humans during this feedback at this kind of scale.
And these weren't just generic contractors, though many of them
worked full time on language tasks tailored specifically to

(02:13):
Grok's rebellious tone and X based content integration.
They created prompts, reviewed output, and flagged errors.
Some were building specialized instruction databases to help
Grok, answer coding questions, generate memes, or hold long
form conversations. Several had backgrounds in
machine learning or linguistics,even if they weren't formally

(02:35):
engineers. And the layoffs came abruptly.
One worker described being locked out of the system mid
task, while others found Slack access revoked before receiving
any formal notice. Most had been expecting long
term contracts and had been toldtheir roles were important to
Grok's future intentions. Sounds very corporate to me and
it's happened to a bunch of people in the past.

(02:57):
Different corporations. If this has happened to you,
please leave a comment on your podcast platform right now.
It's happened to me. I worked in e-commerce before.
One day I tried to log into the system so I could check to see
if we had any orders coming in early in the morning and if so,
I wanted to check out there. If I was doing analytics for the

(03:19):
company, I was running the wholee-commerce system there, the
whole the whole site, and I was checking analytics to see what
these people were doing early inthe morning.
I would wake up at 5:30 or 6 andcheck the analytics and I
couldn't log into my e-mail. I couldn't log into the
e-commerce site. I couldn't log into our employer

(03:41):
portal or our employee Porter portal.
So I called my boss at 6:00 in the morning and asked him what's
going on? Like could we get this fixed?
I don't know if it's a glitch orwhatever.
He said yeah, we shut it all down.
So unlike Xai, he shut the wholee-commerce site down but also
didn't tell me what was going onbecause he told me, and this is

(04:04):
pretty scummy, but he told me I had no right to know what was
going on. Even though I had just invested
a year of my life into building this platform from scratch and
we were getting good sales, we were profitable and I don't know
why they shut it down. To be honest with you.
I have no idea what the heck this guy was thinking because it
was going to be a successful business and luckily I landed on

(04:26):
my feet. Here I am doing this podcast and
I'm successful at being a contractor and also helping
people with their businesses andtheir YouTube and podcast
businesses. So it landed on my feet.
All is good on my end. But these people, 500 people
were laid off. I'm not going to use the word
laid off. That's too nice.

(04:50):
Elon fired them, left them without paychecks, left them
without a way to earn money for their families, left them in
limbo. Who knows if these people were
living paycheck to paycheck, we don't know what they were being
paid. Maybe them being laid off now
they're, you know, they have no money for food for their family.

(05:11):
Maybe they have 3 or 4 kids and none of those kids can eat now
because Elon Musk fired them. I'm being harsh, but it's the
harsh reality of corporate America and corporate Elon.
He just gets rid of people whenever he feels like it.
And I'm a news guy. I'm going to tell how how it is.
I like the companies that Elon is part of.
If that makes you frustrated andmad that I said that, think

(05:35):
about it from a perspective, a broader perspective, not just in
your head, not just what you feel, but think about it like
this. If that happened to you and you
had four kids, I think you'd be upset.
I think one kid even. It's just you.
What are you going to do now? Hopefully you have some savings.
You're eating into your savings because nobody told you what's

(05:56):
going on. Nobody told you that after X
amount of time, you're going to need a new job.
So please, you know, make the right choices, Save some money
because after this job, you're going to need to find another
one. And with this job market, good
luck, especially in tech right now, good luck.
AI is taking all your jobs and XAI and Grok, well, maybe

(06:20):
they're being trained by a Grok version in the background now.
Musk has also hinted at plans toautomate more of the data
labeling and training process using synthetic data, but XAI
has not explained what it's using to replace human
supervision of this. Maybe they have some higher ups
out of those 500 people. Maybe they have their managers

(06:40):
or something. I'm not sure to keep track of
the AI, and additional addition to that, we don't know what's
going on behind the scenes. In traditional AI development,
human annotators play a criticalrole in model alignment in
response tuning. Scaling grok with minimal
oversight creates real risk, especially for a product
designed to answer questions sourced from a volatile platform

(07:02):
like X. There's a huge amount of insane
people on X, and we all know that.
And if they're training their data on X, good luck, you know,
sourcing that. But you know, X and Grok and AI
models are getting so advanced, they might just have the answer

(07:26):
now. This isn't the first time that
Musk has gutted a support team in pursuit of automation.
After taking over Twitter, he eliminated most of the
platform's content moderation staff and replace them with AI
driven filtering and community based reporting.
That move introduced ongoing accuracy and abuse problems that
continue to affect the platform up until today.

(07:48):
Now, the Grok layoffs look like a similar gamble with the same
potential consequences. We're going to have to keep an
eye on Grok's next release and see how wild it gets.
Musk has said that Grok will eventually be a personal AI
assistant fully integrated into X and Tesla platforms, and that
vision requires tight performance, adaptive
conversations in a contextual awareness across different user

(08:11):
inputs. And removing the people who
trained Grok to handle that complexity calls into question
how far along that vision actually is, or whether XAI is
just skipping critical steps to meet aggressive release targets
and get more investor money. Some of the laid off workers had
only recently been on boarded and we're in the middle of

(08:32):
building new data sets. They just got there.
Others have been asking for clearer task guidelines and
internal documentation to keep up with shifting model behavior.
Instead of getting answers, theyjust got fired.
This suggests that XAI may not have had a cohesive plan for
ongoing model improvement, even as it is prepared for the next
version of Grok. Now, XAI has repeatedly

(08:57):
positioned itself as an engineering 1st operation.
Now, that posture often devalueshumans that that are required to
build useful and aligned AI systems.
But removing the people who are responsible for shaping how Grok
actually talks to people, actually responds to people and

(09:18):
reasons, strips the model of an essential feedback loop, one
that can't be easily replaced with more compute alone.
The data annotation team function as Grok's immune
system, catching and correcting errors before they reached end
users. Their absence will show up in
the product itself. If Grok starts sounding more

(09:39):
erratic, a little crazy, missingcontext, or, as AI gurus like to
say, hallucinating or just straight up lying to you more
confidently, this is where it all started.
The 500 people that were fired without reason because they

(10:03):
think they can do it without them.
Hey, thank you so much for listening today.
I really do appreciate your support.
If you could take a second and hit this subscribe or the follow
button on whatever podcast platform that you're listening
on right now, I greatly appreciate it.
It helps out the show tremendously and you'll never
miss an episode. And each episode is about 10
minutes or less to get you caught up quickly.

(10:25):
And please, if you want to support the show even more, go
to Atreoncom Stage Zero. And please take care of
yourselves and each other. And I'll see you tomorrow.
Advertise With Us

Popular Podcasts

On Purpose with Jay Shetty

On Purpose with Jay Shetty

I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!

Stuff You Should Know

Stuff You Should Know

If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.

The Joe Rogan Experience

The Joe Rogan Experience

The official podcast of comedian Joe Rogan.

Music, radio and podcasts, all free. Listen online or download the iHeart App.

Connect

© 2025 iHeartMedia, Inc.