On the one hand, you have all the promising work in artificial intelligence that will one day form the backbone of its business. Robotics alone could eventually generate north of $10 trillion in revenue for Tesla, the CEO estimated—or 15 times what retail behemoth Walmart takes in annually.

And then you have its car business: Unlike the futuristic visions of android butlers in every home, the company’s current pillar records tangible sales and profits today in the here and now. There, the picture looks significantly more challenging. 

Lift up Tesla’s hood, and even the casual observer will spot the rapid onset of rust in its core automotive division.

“I’m surprised to see the stock up,” admitted Deepwater Asset Management partner Gene Munster following Wednesday’s results. 

Despite what he called “largely a messy and difficult outlook for 2025,” the shares are trading 4% higher at press time, after Musk claimed a new Tesla robotaxi service would launch in June.  

Tesla expects Q1 margin hit from losing several weeks of Model Y production

If you look at the company’s actual car sales, growth is tepid at best. Meanwhile, margins earned on its electric vehicles resumed their steady descent in the fourth quarter owing to the liquidation of excess inventory.

Investors’ hopes for a quick rebound were then dashed when Tesla finance chief Vaibhav Taneja warned profitability would be further hit by February’s production start of the refreshed Model Y crossover, internally dubbed Juniper, simultaneously across factories on three separate continents.

“It is an unprecedented change, and we are not aware of anybody else taking the bestselling car on the planet and updating all factories at the same time,” the CFO told investors. “This changeover will result in several weeks of lost production in the quarter.” 

‘They’re having trouble selling cars’

Typically, the benchmark most commonly applied for gauging the financial health of Tesla is the quarterly gross margin at its automotive business excluding earnings from the sale of regulatory CO2 credits to combustion engine rivals. 

This metric dipped to its lowest level in five years, falling to just 13.6%—more than a full percentage point below expectations. It used to earn anywhere between 20% to 30%, back when Tesla’s car business boomed thanks to the January 2020 launch of the Model Y.

2/3… biz, the company would have reported ~53c of Q4 earnings vs. Cons. est. of 77c. Oh… & $TSLA's auto margins (excl credits) imploded, as $TSLA is being forced to aggressively lower prices to move metal given a lot of people simply don't… pic.twitter.com/R2bQhr8rhI

— Gordon Johnson (@GordonJohnson19) January 29, 2025

Net income would have missed substantially had it not been for a noncash, $600 million windfall gain. Under new accounting rules, Tesla was able to mark the value of its Bitcoin holdings higher to reflect the going market price.

“Even after all the layoffs last quarter when they got the benefit of higher margins because of that, now margins are continuing to decline because they‘re having trouble selling cars for obvious reasons,” Gerber Kawasaki CEO Ross Gerber told Bloomberg TV on Wednesday. “They have to [pull] pretty much every demand lever possible to get rid of the vehicles, and we’re seeing it in the numbers.”

Inventory of unsold vehicles slashed

At least part of that Tesla blamed on clearing out vehicles in stock, which resulted in the average selling price dropping more steeply than the savings it could achieve from lower material costs.

“Coming into the fourth quarter, our focus was to reduce inventory levels in the automotive business, and we accomplished that by ending the quarter with the lowest finished goods inventory in the last two years,” Taneja said.

If Tesla did manage to clear out the remaining stock of unsold Model Y vehicles ahead of Juniper’s March debut, that could boost confidence in Musk’s recent promise to investors. During the last investor call three months ago, he predicted volumes would surge by a minimum 20% this year, if not 30% potentially.

But the furthest the company wanted to venture in its official outlook statement was a prediction that vehicle sales would increase this year after 2024’s shock decline. That’s despite the arrival of the refreshed Model Y in March, responsible for two-thirds of Tesla’s car sales, and a new entry model coming in the first half. 

Musk admits 100 million unit sales target for Optimus sounds ‘insane’

“They didn’t address that 20% to 30%, so I think for all practical purposes for 2025 they backed off of that growth expectation,” Deepwater’s Munster said. 

Musk, meanwhile, spent the majority of the call talking up his hopes for AI and Full Self-Driving, rather than cars, which barely got a mention. Training compute, according to the shareholder release, increased by over 400% last year as Tesla built its own “Cortex” data center in Texas.

First, he claimed a Tesla-owned and -operated robotaxi service would roll out across much of the U.S. this year, expanding to then include the addition of customer cars sometime in 2026.

Then he predicted pilot production of his Optimus robot could hit 10,000 units this year and then grow fivefold in size every year to eventually hit a staggering 100 million units. 

“That sounds absolutely insane, I realize that,” Musk conceded. “But I think they will prove to be accurate.”

Two factions square off as Tesla AI bulls war with EV bears

He once harbored such bold predictions for the car business as well, convincing many of his faithful retail shareholders that Tesla would sell 20 million cars in the year 2030. But rather than arrive at this figure through a bottom-up analysis of the auto market, drilling down into the various segments and body styles, the target stemmed from a goal of replacing 1% of the world’s 2 billion–strong combustion car fleet that year.

Critics pointed out such mathematical equations were unrealistic, since they represent more than the total of the two largest carmakers, Toyota and Volkswagen Group, combined. Moreover, it would assume Tesla could grab by itself over a fifth of total demand for all new light vehicles sold in the world. 

Last year Tesla expunged this target from its annual impact report quietly, discreetly, and without any ado. So there is reason to be skeptical regarding Musk’s grand vision, given Optimus is still a prototype.

Wedbush Securities tech analyst Dan Ives admitted Q4 results, with their contrasting stories, would ultimately do little to shed light on who is ultimately right—Tesla’s AI optimists or its EV pessimists.

“For the bears/skeptics solely focused on numbers and margins with no attention to the autonomous narrative, they will feel more confident in their negative thesis,” he wrote.