Skip to content
TrackPodcasts
technologySep 15, 202623:11

Oracle Cuts Staff to Pay AI Debt

Get every episode summarized

Each time Elon Musk Podcast publishes, we email you a written briefing from the transcript — the topics, who appeared, and any specific claims, with the ad reads skipped.

Email me new episodes

Free for 3 shows. No card needed.

About this episode

“You think you know a browser, but Gemini and Chrome? It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50-page restoration block, or finally break down that long article you've had open for weeks.”From the transcript

Oracle has initiated a fresh wave of layoffs as part of a significant organizational restructuring aimed at funding its massive AI infrastructure investments. Reports indicate that the company’s total debt has climbed to $125 billion, prompting a multi-billion dollar budget adjustment to manage these financial shifts. This move follows a prior reduction of approximately 13% of its global staff earlier in the year, as the firm prioritizes artificial intelligence integration over traditional headcounts. Impacted employees are currently utilizing social media platforms to share resources, offer networking support, and navigate the sudden loss of access to corporate systems. Despite the hardship, many former staff members are highlighting the valuable skills gained at the tech giant while seeking new opportunities in a shifting tech industry.

Know when Elon Musk turns up

Follow Elon Musk and once a week we email you every new episode they appeared on — including guest spots the show notes never mention, because we read the transcript.

Follow Elon Musk

Free. Pick your own day and time.

Hosts & guests

Transcript ready

769 searchable segments. Every word is indexed and playable.

Oracle Cuts Staff to Pay AI Debt

Elon Musk Podcast

0:00
23:11

Full transcript

Elon Musk Podcast — Oracle Cuts Staff to Pay AI Debt. Machine-transcribed; use the interactive transcript above to jump the player to any line.

This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome? That's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50-page restoration block, or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it. Ready to make anything online makes sense? There's no place like Chrome. Check responses set up require compatibility and availability varies 18 plus. College football is back. So, Hilton called to me the superstition concierge to make your fan rituals a reality. Need a room to match your lucky number? We got you. Want to make sure our team doesn't wash your lucky jersey? Oh, that smells lucky. Hilton's unmatched hospitality can keep up with any superstition. Even a marching bandwicker call at 555 and 55 seconds. Hit it! When you need a team that will do whatever it takes on game day, it matters where you stay. Hilton, for this day. Good things come in force. Like ukulele strings, boy band members.

And PayPal payin' for. No fees, no interest, no impact on your credit score. Just the flexibility to pay the way that works for you. Strum it, hum it, pay it in for. Download the PayPal app to get started. Subject to approval. Learn more at paypal.com-payin-for. Paypalling NMLS 910457. Oracle is cutting some internal teams by double digit percentages. And it's all to help finance $125 billion debt load build entirely on acquiring artificial intelligence infrastructure. Yeah, it's, I mean, it's a pretty brutal trade-off sitting right there on the balance sheet. You have this very abstract pursuit of computing power, you know, endless server racks and specialized chips. And it is being paid for through the direct immediate subtraction of human capital. So we are looking at how a legacy technology giant basically reorganizes its entire workforce just to feed the financial demands of AI. Which really brings up the question of how a company even maintains its operational core

while, you know, liquidating the very people who built it just to pay for servers. Exactly. And the financial mechanics behind these cuts, they give you a pretty clear sense of the friction involved here. They actually expanded their internal restructuring budget by $700 million. Wow. Yeah. So that brings the total amount dedicated specifically to managing these layoffs to about $2.8 billion. See, when you look at a term like restructuring budget in a regulatory filing, it, well, it really obscures what's actually happening. It sanitizes it. Right. It completely sanitizes it. That is $2.8 billion in cash set aside entirely for the logistical friction of just removing people. 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. Because you have to think about what goes into spending nearly $3 billion just to stop employing people. It is not just, you know, handing someone a check and asking for their badge. No, not at all. You've got severance packages that scale based on tenure.

You've got out placement services and then the legal compliance costs, right? You're terminating employees across dozens of different countries. And every single one of those countries has its own labor laws. Exactly. Like the European Union alone requires these incredibly long mandatory consultation periods before you can even finalize a layoff of a certain size. So they're literally paying lawyers in France and Germany and the UK just to figure out how to legally stop paying their own engineers. Yeah. And then you have the physical assets too. You have to break commercial leases on office space that you just don't need anymore. Writing down the depreciation of the hardware. Right. The hardware that all those employees were using it all adds up to that $2.8 billion. But I mean, isn't that just the harsh reality of the tech sector right now? Like if they don't shift to AI infrastructure, they risk total obsolescence. Sure. So these cuts as brutal as they are, management clearly views them as entirely necessary for survival. Oh, they're definitely making a calculated bet. They look at the carrying cost of the current workforce and they just decide that capital is better spent

servicing the debt on physical infrastructure. Right. The $125 billion debt. Yeah. But the paradox here is that they have to burn $3 billion of capital just to execute the pivot. That is $3 billion that isn't buying GPUs. Right. It's not buying cooling systems. It's not buying chips. It's just paying for the friction of the pivot. And to kind of contextualize the size of what we're dealing with here, you just have to look at their previous headcount reduction in that round. The global workforce dropped by 21,000 employees, which is, I mean, that's a lot of people. It represented a 13% reduction in their total personnel. You remove 13% of a workforce that size and you are essentially removing a small city from your organizational chart. Yeah. The connective tissue of daily operations just takes a severe hit when that volume of people exits the system all at once. And we can link those numbers directly to the reasoning they provided in the regulatory filings. Management explicitly stated that these ongoing workforce reductions are driven by the integration of AI

across various operational functions. So they're essentially stating that they are automating their internal mechanisms. Yes. The jobs aren't just being eliminated to save cash. Right. They are being actively. This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome, that's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration block, or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it. Ready to make anything online make sense? There's no place like Chrome. Check responses set up require compatibility and availability, very 16+. College football is back. So, Hilton called to me the superstition concierge to make your fan rituals a reality. Need a room to match your lucky number? We got you. Want to make sure our team doesn't wash your lucky jersey? Oh, that smells lucky. Hilton's unmatched hospitality can keep up with any superstition. Even a marching bandwicker call at 555 and 55 seconds. Hit it! When you need a team that will do whatever it takes on game day,

it matters where you stay. Hilton, for this day. Good things come in force, like ukulele strings, boy band members, and PayPal paying for. No fees, no interest, no impact on your credit score, just the flexibility to pay the way that works for you. Strument, hum it, pay it in four. Download the PayPal app to get started. Subject to approval. Learn more at PayPal.com-payin-four. Paypalling NMLS 910457. We replaced by the infrastructure that the company is taking on all this debt to build. They are literally swapping salaries for server maintenance. It's a very blunt exchange. It is. And I want to look at the actual execution of these cuts, because the human experience of this corporate strategy is just jarring. Yeah. We had the experience of Yuri Rich. He was a vice president of Sovereign Cloud over there. Okay. And he discovered his termination solely because his corporate account access was just abruptly revoked. Which you have to imagine that scenario, right?

You wake up for your coffee, you sit down at your kitchen island to start your work day. As you do every day. Exactly. You open your laptop, try to authenticate into your email, and the screen just flashes an error. Yeah. You haven't been fired by your boss. You have been evicted by a script. And the psychological impact of that method is severe. There was no email sent to his personal account beforehand. Right. There was no phone call from HR. No meeting with a supervisor. Just a blank screen. The only notification was the mechanical failure of his credentials. You could just no longer access any corporate resources. It basically functions as a digital eviction. It's like the locks are changed before the tenant is even notified. Yes. You try to log into your daily workflow. And the system just rejects your identity. It treats you as an intruder in the digital home you've occupied for years. Which creates this really stark contrast with the tools the company is actually building. I mean, they're investing $125 billion into AI

to create seamless, intelligent, highly responsive, automated systems. Right. Tools that understand context. Exactly. But the human relations process is handled with the blunt force of an automated script that just severs access overnight. And that contrast is exactly what makes this so difficult for the workforce. Yeah. The company sells the promise of sophisticated technology. But internally, they deploy technology as a blunt instrument. And we should probably look at what the role of Vice President of Sovereign Cloud actually entails because that adds another layer to this. Sovereign Cloud operations deal with highly sensitive, often air-gapped data centers. These are for government clients or industries with really strict compliance requirements. Right. And for anyone listening who hasn't worked with government contracts an air-gapped server basically means it has absolutely no physical connection to the outside internet. He's completely isolated. Isolated. The data sovereignty requirements mean that only specific, highly cleared individuals in very specific geographic locations

can even be in the same room as the hardware. So this is not some peripheral administrative role we are talking about. This is core infrastructure leadership. Yeah. We're talking that the person responsible for ensuring that highly classified regulated data remains secure. And revoking access for an executive at that level without any formal off-boarding conversation, it points to a security protocol that just supersedes all other operational considerations. The system essentially treats the newly laid off employee as an immediate critical threat vector. It's just a mechanical ruthlessness. Yeah. The identity and access management system receives a flag from the HR database and then it just cascades through the entire network. Active directory revokes the token, the VPN connection drops, and the laptop is effectively turned into a brick. Wait, just for those who haven't worked in Enterprise IT, explain what that token actually is. Is it like a digital master key that just evaporates? Basically, yeah. In modern enterprise architecture, you don't just log in once and have free reign for the day.

You are issued an authentication token. Okay. And that token constantly verifies their identity in the background as you move from your email to your database to your HR portal. So it's always checking. Constantly. So an HR flags your file for termination. An automated workflow tells the central directory to instantly invalidate that token. Uh, so the second that token is destroyed, every single application you have open simultaneously realizes you no longer exist in the corporate hierarchy. Exactly. Your email client stops syncing. Your Slack messages just won't send. Access to cloud storage is blocked. Yeah. The zero-trust architecture is fully deployed against the company's own leadership. And the efficiency of a lockout is actually the only part of the process that functions flawlessly. That is a great point. Right. The company manages the digital execution perfectly, but they entirely abandon the interpersonal communication that, you know, traditionally, a company is the end of an employment contract. In this environment, naturally leads to some pretty intense practical survival tactics emerging among the staff.

Of course. When the company operates like a machine, the employees have to figure out how to outsmart the machine. Which brings us to Percy Chan. He actually posted advice for his colleagues. Oh, yeah, I saw this. Yeah, he was telling them to immediately disconnect their laptops from the network and turn off their Wi-Fi to save personal files before the system lockouts occur. It's like these defensive maneuvers resemble a fire drill for digital survival. It really does. The abrupt termination process forces employees to basically treat their own IT department as an adversary. Yeah, he is advised to pull the plug on the router or shut off the Wi-Fi adapter. It's a physical intervention against a digital process. Right, you have to sever the connection before the termination command can actually reach your local machine. You have workers actively trying to outpace the automated security scripts just to secure their own data. Because they're trying to preserve, you know, performance reviews, personal contacts. If you port folio materials that they will absolutely need to secure their next job.

Yeah, you have thousands of employees sitting at their desks, just monitoring their connections, waiting for the moment and application goes down or their email client throws an authentication error. The baseline anxiety of that working environment must be staggering. It reveals a complete breakdown of the social contract between the employer and the employee. Totally. When the standard operating procedure for a layoff involves zero notice and immediate credential revocation, well, the rational response from the workforce is preemptive data hoarding and network evasion. And the irony is that the security protocols designed to protect the company's intellectual property from external hackers are now being utilized to isolate the employees. Right, the threat model has shifted internally. The workforce is managed as a liability rather than an asset. So if you are locked out overnight, you have absolutely no way to even contact your team to ask what happened. Yeah. You can't check the internal directory to see who else disappeared. Right.

Where do these people even go to figure out if they are alone? So that leads directly to these grassroots support networks that are forming completely outside the company's purview. Mm-hmm. We see the creation of invite-only Slack groups. Which is fascinating. And these groups currently have over 3,000 former employees in them. It's like a shadow organization forms almost immediately. When the formal corporate structure discards these individuals, they just reconstruct a parallel network to manage the fallout. Yeah. And 3,000 people, I mean, that is a significant sample size of institutional memory suddenly operating outside the corporate firewall. And there is a strict vetting process for these groups. Yeah. Managed by people like Terry Hale. Right. The vetting creates this necessary firewall where x employees can actually speak freely without the presence of current management. The need for that firewall is really telling, though. It shows there is a lingering fear of corporate surveillance. Oh, absolutely. You have people worried that speaking openly about severance packages or organizational failures

could actually jeopardize their exit agreements. So they require a completely sterilized environment to process the layoff? Yeah. They use these channels to share job leads, referrals, recruiter intel. And just to answer basic questions about severance. Right. And the groups operate under names like sweet escape, exorical riefies, and exorical. It basically replaces the function of an outplacement agency. Exactly. The former employees are leveraging their collective network to absorb the shock of the layoff. Because the company provided no transitional support beyond the financial package. Right. They're just crowdsourcing the recovery process. And Terry Hale noted that the only requirement for entry into these groups is proof that you were laid off from the company. The invite link is never posted publicly. It is a closed loop system of displaced talent. Yes. But think about the logistics of that for a second. How do you prove your layoff when you don't have access to your corporate email to show a termination letter? That is a great question. You can't just forward an email.

Right. They have to rely on personal networks, LinkedIn connections, and secondary verification just to let people into a support group. You actually see a really high level of organizational skill and how these groups are managed. Terry Hale's background in operations and building inclusive teams translates directly into how this recovery network is structured. They are applying their professional competencies to the problem of their own unemployment. Exactly. Which brings us to the tools these former employees are using to bounce back. Oh, man. And this is where the situation becomes incredibly cyclical. Yeah. This is the part that really gets me. So MacBul Khan, one of the former employees, he built an AI-driven app called Intentful. Works specifically to help his former colleagues prepare for interviews. This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome, that's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50-page restoration block or finally break down that long article

you've had open for weeks. Gemini and Chrome is here for it. Ready to make anything online makes sense? There's no place like Chrome. Check responses set up require compatibility and availability varies 18 plus. In Game Day, it matters where you stay. Hilton, for this day.

So let me just get this straight. Yeah. The company lays off a massive portion of its workforce to pay for AI infrastructure. Correct. And the immediate reaction of the displaced workers is to build an AI application to get a new job. That is exactly what happened. That feels incredibly dystopian, doesn't it? It really does. Khan's application focuses on helping users build their personal brand and prepare for star-driven interviews. Okay, and just to clarify, the star method that's situation, task, action, and result. Right. It is the standard behavioral interview framework in the tech industry right now. Using an AI to simulate a star interview is basically an escalation of the hiring arms race. Yeah, also. Well, you have companies using AI to filter resumes, analyze cover letters, and assess candidates, right? Yeah. Now the candidates are using AI to optimize their responses, anticipate those algorithmic filters and generate the perfect behavioral examples. The technology just sits on both sides of the interview table. Exactly. It's machines talking to machines to decide if a human gets a paycheck.

And we also have Joshua Bartels. He created a custom tool to help tailor resumes more efficiently for these former employees. Right. It automates the process of aligning a workers' historical experience with the specific keywords required by new job descriptions. So the efficiency games that made their roles redundant are the exact same efficiency gains they are relying on to secure new roles? Yeah. They're literally utilizing the mechanism of their own displacement to survive the aftermath. It's wild. But I want to shift a bit because we really have to focus on the human capital and the institutional knowledge that is being discarded in this whole process. Definitely. We have the story of a really as figurado who started as an intern and spent 33 years at the company before being let go in this round. Over three decades. I mean, this guy started decades ago. Yeah. He survived the .com bubble bursting, the financial crash, the shift from local servers to cloud computing and what finally takes him out is an algorithm clearing cap space for servers.

A 10-year of that length in a single technology company is incredibly rare these days. Extremely rare. You have to consider the long-term cost of losing that amount of operational memory. Right. A company functions on documented processes, sure, but it survives on the undocumented knowledge held by people who have seen every iteration of the system. His mental map of the company's architecture and client history is just irreplaceable. Yeah. When you sever that connective tissue, you create these massive operational blind spots. Because a newly hired engineer or even an AI agent might be able to write faster code. Sure. But they do not know why a specific legacy system was configured a certain way for a legacy client who still generates huge recurring revenue. Exactly. And that loss of context leads to cascading failures. Well, think about it. An automated system might identify redundancy in the network architecture, right? Right. And it eliminates it to save computing resources. Makes sense on paper. But it's completely unaware

that the redundancy was a deliberate failsafe engineered after a catastrophic outage years ago. The person who remembered the screaming clients during that outage is gone. The calculation the company is making is that the cost of retaining that historical context is higher than the theoretical efficiency gains of the new AI infrastructure. Right. They are perfectly willing to absorb the operational friction of losing their veterans just to free up capital. But how do you quantify the loss of someone who knows where all the digital bodies are buried? You really can't. You can't put a line item on a balance sheet for knowing how to fix the bespoke database for a major European bank. No, you can't. And we can add the cultural perspective shared by Amy P here. She described the corporate environment as highly rigorous. OK. She actually called it a training ground that prepares workers for complex demanding stakeholders at scale. And that rigor is a double-edged sword for the company, really. How so? Well, they have engineered an environment that forces employees to operate

at a very high level of complexity, right? Right. They train their workforce to navigate difficult stakeholders, manage massive databases and deliver under extreme pressure. And then they lay off that highly trained workforce. Exactly. The skills those employees built the ability to operate at scale are just immediately released into the open market, which explains why competitors are aggressively circling these layoff announcements. Oh, they're swarming. The company has essentially subsidized the training and development of a premium workforce for their rivals. They spent decades forging these engineers and then simply handed them over. That brings us to GenoSero. He immediately used the situation to recruit his former colleagues to disall systems. Right. I saw his posts. Yeah. He is weaponizing the layoffs against his original employer by scooping up the displaced talent. Sero's approach is highly systematic, too. It is. He publicly mentions the casualties resulting from the battle to build AI centers. Right.

And then he positions his new employer as this refreshing change of culture with positive leadership. He is directly attacking the organizational chaos of the layoffs. To attract high-level engineers. Yeah. He even provided a link in his comments showing people how to export their LinkedIn contacts to a spreadsheet. Wow. He is literally teaching them how to systematically map their former colleagues and leverage those connections for external job openings. The immediate transfer of human capital there is just staggering. It is. You have intact teams, people who know how to work together. People who already have established trust and communication protocols. Right. Migrating wholesale to a competitor. The acquiring company gets the benefit of a functioning unit without any of the onboarding friction. Exactly. They don't have to figure out who the natural leader is or how the team resolves conflicts because that dynamic was already forged at the previous employer. They just plug them in and go. Yeah. Zooming out a bit to connect this internal corporate strategy to the current macroeconomic environment.

We see severe external pressures driving this behavior. Yeah. The macro picture is huge here. The 10-year treasury yield is hitting 5%. And oil prices are fueling inflation fears across the broader market. Which really explains the timing. Right. The reason they are moving this fast goes back to the debt. The 125 billion. Exactly. And the debt is suddenly a massive problem because of what's happening with the 10-year yield. It sets the risk-free rate, which fundamentally alters the discount rate for all future cash flows. Let's explain why that matters to a company carrying this much debt just to make it concrete. Sure. If you are a homeowner with an adjustable rate mortgage and you decide to build this massive expensive swimming pool while interest rates are near zero, well, the monthly payments are manageable. Right. The capital is practically free. Exactly. But when the 10-year yield jumps to 5%, the cost of borrowing just explodes. So that 125 billion dollar debt load transforms from a strategic advantage

into a severe operational constraint. Right. The interest payments alone require a massive diversion of cash flow. And that cash flow has to come from somewhere. It cannot come from the AI infrastructure yet because that infrastructure is still being built. It's not being monetized yet. Right. Therefore, it has to come from the existing operational budget, which primarily consists of payroll. Exactly. The macroeconomic reality forces the internal corporate decision. They are liquidating human capital to service the debt on physical capital. Yeah. And the inflation fears driven by oil prices only compound this. Because operating costs for data centers require massive amounts of energy, right? Exactly. Those costs will also rise. So we connect this financial pressure to the broader tech sector's hurdles with AI development in general. There are reports of AI leaders facing major obstacles in efforts to slow down development speeds. The arms races accelerating regardless of the economic friction. Right. Microsoft is releasing a new code of conduct amid these calls for a slowdown.

But the entire sector is rushing to build the infrastructure anyway. And that rush is violently colliding with the economic reality of high-capable costs. Because the structural demands of AI require real tangible things. Data centers, advanced cooling systems, specialized silicon. You need physical materials, land, and energy. Exactly. You cannot code a data center into existence. Right. You have to actually build it and building it in a high interest rate inflationary environment is just exceptionally expensive. So the company looks at its balance sheet. Sees the $125 billion debt. Sees the rising cost of servicing that debt. And they decide the only flexible variable left is the workforce. The assumption is that the AI will eventually generate enough revenue to justify the debt. Right. And that it will automate enough internal processes to justify the layoffs. But in the interim, they are operating in this highly-leveraged, highly volatile transitional phase. It's a very precarious spot. The transition to automated infrastructure

just requires immense physical and financial capital, which is currently being subsidized by liquidating human talent. It really leaves you wondering, you know, when the next major technological shift happens in 10 years, will the AI agents they are building today be the ones getting abruptly disconnected to make room for whatever comes next? If you're not subscribed yet, take a second and hit follow on whatever app you're using. It helps us keep making this. We appreciate you being here. Also, check out our YouTube channel for more business and tech updates. There's a link in the description. Push your limits, train with precision, see the results.

At Equinox, that's high performance loving. Iconic spaces that inspire. Personal training backed by real data, unlimited group fitness classes, from yoga and Pilates to strength and conditioning. Elevate your post-performance ritual with sauna's, steam rooms, cold plunges, and more. Everything you need to lock in and unlock your potential at Equinox. Start today at equinox.com. Here shop Labor Day Savings at the Home Depot today. Offer valid august 27th through September 16th to assembly a C-store online for details. Booking.com is the easiest way

from a day surrounded by noise. To a state surrounded by nature. That's nice. Go on, book it. It's easy. Booking.com. Booking. Yeah. NFL football season is here, which means when you switch to Verizon, it feels like... Just go! ...because you get NFL Sunday ticket from YouTube on us. When you buy an eligible 5G phone on select unlimited plans, which means you can watch every out of market game every Sunday afternoon, and it's all on us. Now that's the Verizon way to kick off the NFL football season right. Switch and get NFL Sunday ticket from YouTube on us, only with Verizon.

More episodes

More from Elon Musk Podcast

View all episodes →