Connect with us

Tech

Elon Musk suggests spate of xAI exits have been push, not pull

Elon Musk is addressing a wave of departures from xAI, including two more co-founders who left this week, bringing the total to six out of the original 12.

At an all-hands meeting Tuesday night, Musk suggested the exits were about fit, not performance. “Because we’ve reached a certain scale, we’re organizing the company to be more effective at this scale,” he said, according to The New York Times. “And actually, when this happens, there’s some people who are better suited for the early stages of a company and less suited for the later stages.”

Wednesday afternoon on X, he went further, making clear these departures weren’t voluntary. “xAI was reorganized a few days ago to improve speed of execution,” Musk wrote. “As a company grows, especially as quickly as xAI, the structure must evolve just like any living organism. This unfortunately required parting ways with some people.”

He added that the company is “hiring aggressively” and closed with a quintessentially Musk pitch: “Join xAI if the idea of mass drivers on the Moon appeals to you.”

Losing half your co-founders in a relatively short period raises questions, and Musk’s comments seem designed to control the narrative, reframing the exits as necessary rather than a problem for the outfit.

In total, at least nine engineers, including the two co-founders, have publicly announced their departure from xAI in the past week — though two of those exits appear to have occurred a few weeks ago.

Three of the departing staff members have said they will be starting something new alongside other former xAI engineers, although no details are available about the new venture. Others have hinted at a desire for more autonomy and smaller teams to build frontier tech more rapidly, pointing to the anticipated surge in AI productivity.

Techcrunch event

Boston, MA
|
June 23, 2026

Yuhuai (Tony) Wu, an xAI co-founder and reasoning lead, said in a post announcing his resignation: “It’s time for my next chapter. It is an era with full possibilities: a small team armed with AIs can move mountains and redefine what’s possible.”

Shayan Salehian, who worked on product infrastructure and model behavior post-training at xAI and previously worked at Twitter/X, said last week he was leaving to “start something new.” 

Vahid Kazemi, who had a brief stint working on machine learning, posted Tuesday that he left a few weeks ago, adding: “IMO, all AI labs are building the exact same thing, and it’s boring … So, I’m starting something new.”

Roland Gavrilescu, a former xAI engineer, left in November to start Nuraline, a company building “forward-deployed AI agents,” but posted again on Tuesday that he left the firm to build “something new with others that left xAI.”

The departures come at a moment of significant controversy for xAI. The company is facing regulatory scrutiny after Grok created nonconsensual explicit deepfakes of women and children that were disseminated on X — French authorities last week raided X offices as part of an investigation. The company is also moving toward a planned IPO later this year, after being legally acquired by SpaceX last week.

Musk is also facing personal controversy after files published by the Justice Department show extended conversations with convicted rapist and sex trafficker Jeffrey Epstein. The emails show Musk discussing a visit to Epstein’s island on two separate occasions, in 2012 and 2013. Epstein was first convicted of procuring a child for prostitution in 2008.

xAI maintains a headcount of over 1,000 employees, so the departures are unlikely to affect the company’s short-term capabilities. Still, the rapid pace of the recent departures had taken on a life of its own online, with users jokingly announcing on X that they too are “leaving xAI” despite never having worked there — a sign of how quickly the narrative of a “mass exodus” snowballed on Musk’s social network.

Still, forced co-founder exits are rarely a sign of smooth scaling. While Musk frames the reorganization as calculated, the fact that several engineers followed the co-founders out the door — and that at least three are starting something new together — suggests the departures may also reflect deeper tensions. In frontier AI, where talent is scarce and reputation matters, xAI’s ability to attract and retain top researchers will be tested as it competes with OpenAI, Anthropic, and Google.

TechCrunch has reached out to xAI for more information.

Timeline of departure announcements

The following employees have publicly announced their departures from xAI on X in recent days:

February 6:  Ayush Jaiswal, engineer, wrote: “This was my last week at xAI. Will be taking a few months to spend time with family & tinker with AI.”

February 7: Shayan Salehian, who worked on product infrastructure and model behavior post-training and was previously at X, wrote: “I left xAI to start something new, closing my 7+ year chapter working at Twitter, X, and xAI with so much gratitude.” He added that working closely with Elon Musk taught him “obsessive attention to detail, maniacal urgency, and to think from first principles.”

February 9: Simon Zhai, MTS (member of technical staff), wrote: “Today is my last day at xAI, feeling very fortunate about the opportunity. It has been an amazing journey.”

February 9: Yuhuai (Tony) Wu, co-founder and reasoning lead, wrote: “I resigned from xAI today. It’s time for my next chapter. It is an era with full possibilities: a small team armed with AIs can move mountains and redefine what’s possible.”

February 10: Jimmy Ba, co-founder and research/safety lead, wrote: “Last day at xAI. We are heading to an age of 100x productivity with the right tools. Recursive self improvement loops likely go live in the next 12 months. It’s time to recalibrate my gradient on the big picture. 2026 is gonna be insane and likely the busiest (and most consequential) year for the future of our species.”

February 10: Vahid Kazemi, an ML PhD, wrote that he had left xAI “a few weeks ago,” adding: “IMO, all AI labs are building the exact same thing, and it’s boring. I think there’s room for more creativity. So, I’m starting something new.”

February 10: Hang Gao, who worked on multimodal efforts, including Grok Imagine, wrote: “I left xAI today.” He described his time there as “truly rewarding,” citing contributions to Grok Imagine’s releases and praising the team’s “humble craftsmanship and ambitious vision.”

February 10: Roland Gavrilescu, the engineer who left in November to start Nuraline, posted: “I left xAI. Building something new with others that left xAI. We’re hiring :)”

February 10: Chace Lee, a member of the Macrohard founding team, wrote: “Taking a brief reset, then back to the frontier.” (Macrohard is an AI-only software venture under xAI designed to fully automate software development, coding, and operations using Grok-powered, multi-agent systems. Its name is a dig at Microsoft.)

Got a sensitive tip or confidential documents? We’re reporting on the inner workings of the AI industry — from the companies shaping its future to the people impacted by their decisions. Reach out to Rebecca Bellan at rebecca.bellan@techcrunch.com or Russell Brandom at russell.brandom@techcrunch.com. For secure communication, you can contact them via Signal at @rebeccabellan.491 and russellbrandom.49.


source

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Tech

Revolut eyes valuation of up to $200B in eventual IPO

British neobank Revolut seems to be eyeing a major valuation bump when it eventually goes public. The company is targeting a market cap between $150 billion and $200 billion in an initial public offering, the Financial Times reported on Tuesday, citing anonymous investor sources.

The fintech giant, which secured a full banking license in the United Kingdom in March after years of waiting, was most recently valued at $75 billion, up from $45 billion in 2024, in a secondary share sale that made it one of Europe’s most valuable private tech companies.

Revolut’s co-founder and CEO, Nik Storonsky, last week said that the company’s IPO was at least “two years away,” according to Bloomberg.

According to PitchBook and the Financial Times, the company is working on another secondary share sale, scheduled for the second half of 2026, that would value it at more than $100 billion.

As of November 2025, the company had raised a total of $5.89 billion, according to PitchBook. Revolut reported revenue of $6 billion in the financial year ended December 31, 2025, up from $4 billion in 2024. The company’s net profit grew to $1.7 billion, up from $1 billion in 2024, and counted 68.3 million retail customers at the end of 2025.

Revolut declined to comment.

Founded in 2015, Revolut offers a range of services spanning multi-currency accounts, payment and transfer services, crypto products, insurance, and more. The neobank has been pouring truckloads of cash into expanding its operations internationally, and recently applied for a banking license in the United States.

Besides the U.K., Revolut has a banking license in the European Union, and it operates in Australia, Japan, New Zealand, Singapore, Brazil, and the U.S. Revolut launched operations in India last October, is about to start operating in Colombia this year, and has received a banking license in Mexico.

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

source

Continue Reading

Tech

Amazon taps Sweden’s Einride for its electric big rigs

Einride is adding 75 of its electric heavy duty trucks to Amazon’s Relay freight network as part of a deal that gives the Swedish startup a toehold in the e-commerce giant’s operations. Einride will also provide charging infrastructure across five locations in the United States, under the agreement announced Tuesday.

Amazon isn’t buying or operating the electric trucks. Instead, Einride will own and manage (using its own Saga AI software) the trucks, which can be used by drivers in Amazon’s Relay freight network. Relay, launched in 2017, is an app that truck drivers can use to book hauling gigs with Amazon.

Einride CEO Roozbeh Charli, who took over as chief nearly a year ago, said working with Amazon is a powerful validation of the startup’s technology and strategic vision.

“By deploying our intelligent platform within one of the world’s most sophisticated logistics networks, we are accelerating growth, while continuing to build industry-leading operational expertise,” he said in a statement.

Einride has gained attention and investment for its two-pronged approach to freight. The company has developed and now operates a fleet of about 200 heavy-duty electric trucks for companies like Heineken, PepsiCo, and Carlsberg Sweden in Europe, North America, and the UAE. It has also developed autonomous pod-like trucks, which stand out for their cab-less design.

The agreement with Amazon doesn’t include the autonomous pods.

Einride has landed this agreement at a critical time: The startup is finalizing a merger with blank-check company Legato Merger Corp. and is expected to go public soon.

Techcrunch event

San Francisco, CA
|
October 13-15, 2026

While the agreement might not carry the same weight for Amazon, which has a market cap of $2.7 trillion, it does contribute to its low-carbon goals. Amazon has said it wants to reach net-zero carbon emissions across its operations by 2040.

“This rollout is an important step forward in addressing one of the toughest challenges we face in decarbonizing our transportation network — electrifying heavy-duty trucking,” an Amazon spokesperson said in an emailed statement. “We’re excited to continue to collaborate with Einride and learn from these operations as the trucks hit the road.”

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

source

Continue Reading

Tech

YouTube expands its AI likeness detection technology to celebrities

YouTube is expanding its new “likeness detection” technology, which identifies AI-generated content, such as deepfakes, to people within the entertainment industry, the company announced on Tuesday.

The technology works similarly to YouTube’s existing Content ID system, which detects copyright-protected material in users’ uploaded videos, allowing rights owners to request removal or share in the video’s revenue.

Likeness detection does the same, but for simulated faces. The feature is meant to help protect creators and other public figures from having their identities used without their permission — a common problem for celebrities who find their likenesses have been used in scam advertisements.

The technology was first made available to a subset of YouTube creators in a pilot program last year before expanding more broadly to include politicians, government officials, and journalists this spring.

Image Credits:YouTube

Now YouTube says the technology is being made available to those in the entertainment industry, including talent agencies, management companies, and the celebrities they represent. The company has support from major agencies like CAA, UTA, WME, and Untitled Management, which offered feedback on the new tool.

Use of the likeness detection tool does not require entertainers to have their own YouTube channels.

Instead, the feature scans for AI-generated content to detect visual matches of an enrolled participant’s face. Users can then choose to request removal of the video for privacy policy violations, submit a copyright removal request, or do nothing. YouTube notes that it won’t remove all content, as it permits parody and satire content under its rules.

In the future, the technology will support audio as well, the company says.

Related to this, YouTube has also been advocating for similar protections at a federal level, with its support for the NO FAKES Act in Washington, D.C. This would regulate the use of AI to create unauthorized re-creations of an individual’s voice and visual likeness.

The company hasn’t yet said how many removals of AI deepfakes have been managed by the tool so far, but noted in March that the amount of removals was still “very small.”

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

source

Continue Reading