Okay, I am not a lawyer so I only understood about half of what just happened. But I am fairly sure, given the context, that Elon Musk’s lawyers may have just fucked up big.
Technology
The craziest part of Musk v. Altman happened while the jury was out of the room
Jared “James Brickhouse” Birchall, Musk’s finance guy and all-around fixer, took the stand after Musk today. Most of his testimony was dull and seemed to exist primarily to get some documents read into the record, which sucks but is a normal part of sitting through trials. But at the very end of his boring testimony something interesting happened. I believe we all got a surprise, something that rarely happens in courtrooms.
The lawyer conducting his direct examination was passed a note by another member of the team, and asked Birchall what was apparently contained on the note: was he familiar with the xAI bid for OpenAI’s assets?
“Sam Altman was on both sides of the table.”
“As I recall, a lawyer we were working with had asked the attorney general of California to ensure that in their fiduciary duty, proper value was being given to the assets of the nonprofit of OpenAI,” Birchall said. In his understanding, there was a negotiation “between Sam Altman and himself on both sides of the table, the for-profit and the non-profit, attempting to discount the value of the non-profit assets. And we made that bid in an attempt to properly account for the value the foundation had, and create a market bid that would need to be considered by the attorney general.”
Here’s some lore: in February 2025, a Musk-led coalition made a $97.4 billion bid for the non-profit that controls OpenAI. The bid was submitted by Marc Toberoff, one of Musk’s lawyers in the current case. This bid happened as OpenAI was restructuring itself so that the for-profit arm could be cleared to go public. In Birchall’s testimony, that bid was made because Musk, Birchall, and others, thought Altman might undervalue the nonprofit as the company restructured itself. (I’m not really sure why that would be a problem for Musk and xAI, frankly, but whatever.)
The defense counsel objected, and Birchall’s rant was struck for lack of foundation. So we did this piece by piece to establish the foundation, ending with Birchall saying, again, “Sam Altman was on both sides of the table.”
On cross-examination, Bradley Wilson from Wachtell Lipton — OpenAI’s lawyers — picked the thread back up. Wilson asked how much of this Birchall had learned from sources other than lawyers. Birchall said he’d have a hard time being able to untangle that. After a few more exchanges, Wilson moved to strike all of Birchall’s testimony about the xAI bid on grounds that would not be discussed in front of the jury.
“You must have been very convincing. You’re not very convincing today.”
The jury got to leave early while the lawyers duked it out, and this is where it got weird. Judge Yvonne Gonzalez Rogers started asking Birchall questions herself, and it clearly was making Birchall nervous. Birchall said he doesn’t remember discussing the xAI bid with Musk or Shivon Zilis or any other principal of the Musk organization. It sure sounded like Musk’s lawyers hadn’t given OpenAI proper discovery on this topic in the depositions, and so we were doing a fast and dirty deposition with the judge right then. At one point, Gonzalez Rogers told the plaintiff’s counsel to quit coaching the witness.
Birchall said he’d spoken to the other members of the consortium about the bid, but that he wasn’t involved in discussions with Musk about when to send the bid letter. He claimed he’d heard some things from Toberoff, but that he wasn’t aware that Toberoff represented some of the other bidders. He didn’t know if xAI was aware that Toberoff represented some of the other bidders, either.
Birchall didn’t know whether other investors had first-hand information about OpenAI, he claimed. No one had documents from inside OpenAI as far as he knew. Gonzalez Rogers remained unconvinced. “I’m still struggling with how you can have conversations with these individuals to raise $97.5 billion but have no recollections even in a general sense,” she said. Birchall said he had a general sense — he called each of the people involved to see if they were interested in joining Musk on the bid.
“Why would they do that?” Gonzales Rogers asked. Birchall said these were people with whom Musk et al had longstanding relationships. “You must have been very convincing,” she said. “You’re not very convincing today.”
Birchall said there were no numbers besides the topline one floated when he called prospective investors, and that after speaking with him, they were passed off to lawyers. He didn’t remember who chose the $97.4 billion number, and said he got it from the legal team, telling Gonzalez Rogers he didn’t get it from Musk. Gonzalez Rogers asked if that analysis was created by anyone besides Toberoff. Birchall said not that he could recall.
“Did a lawyer tell you this was part of litigation?” Gonzalez Rogers asked.
No, Birchall said. It was strictly a business deal.
Apparently Steven Molo, who’d been defending Musk during the deposition, had made multiple objections to questions about the deal, citing privileged communications. Business deals, apparently, aren’t privileged. But all discovery into the xAI bid for OpenAI had been blocked before the trial began. Unfortunately, by asking Birchall about the xAI deal at the very end of the direct examination, Musk’s team may have opened the door for more digging into it. You may be wondering, “open the door to what” and your guess is as good as mine. More discovery? Maybe something about anticompetitive behavior from Musk? It doesn’t sound like it’s going to be good for Musk, I can tell you that much.
Gonzalez Rogers then asked who’d passed the note, and all the lawyers just sat there like guilty children. Finally, the guy responsible said he’d passed it, but he didn’t write it; a junior lawyer did. Who wrote it? More silence. Finally Toberoff — hardly a junior lawyer — stood up and took responsibility. Why had he done it? “I thought it was appropriate.”
“Sounds like you wanted to open the door, then,” Gonzalez Rogers said. We adjourned while she said she’d consider what to do with this testimony. She will probably rule on it tomorrow.
Correction, April 30th: It is Shivon Zilis, not Sharon Zilis.
Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
Roger Linn is a legend in the world of musical instruments. He’s been at the cutting edge of music technology for decades. He created the LM-1, the first drum machine to use samples, and its successor, the LinnDrum, is one of the most iconic drum machines of all time. They were used on countless records in the 1980s, including hits by Tom Petty, Queen, and Tears for Fears. But the most notable fan was probably Prince, who used them extensively on Purple Rain and 1999.
Somehow, those are not his greatest contributions to the music world. That would, undoubtedly, be the MPC. Linn partnered with Akai to create one of the most popular and important samplers ever. The MPC60 and its successors became the tool of choice for countless hip-hop and house producers. J Dilla’s MPC 3000 even sits in the Smithsonian.
Roger Linn was also an early adopter of MPE, or MIDI polyphonic expression. It’s a key feature of his LinnStrument, an expressive 3D controller released in 2014 — three years before the Association of Musical Electronics Industry (AMEI) officially released the MPE standard. Turns out the man stays so innovative by keeping things simple and focused.
What is your most indispensable tool?
My MacBook Pro.
Which is the most underappreciated?
My Vision Pro. I called it the most amazing product I rarely use.
What is the first app you install on a new phone or computer?
On a computer, Rhino3D.
What is one thing you wish you could change about your phone?
Apple Mail’s bugs.
What sites do you have pinned to your tab bar?
New York Times.
How many tabs do you have open right now?
One. This document.
Which social media platform do you use the most (if any)?
I don’t use social media except to announce my monthly “All Things LinnStrument” email newsletter.
What is your happy place online?
A VR app for the Meta Quest called Walkabout Mini Golf. It was a large number of artistically created open VR worlds that offer a surprising level of beauty from the Quest 3’s limited power. I go there to play a game of mini golf, fly around, or meet friends in a private instance of a particular world.
What is your favorite gadget you’ve ever owned?
I don’t know about “ever”, but these days it’s VR headsets, currently the Meta Quest 3 or Apple Vision Pro.
Which was the most disappointing?
In general, I’m disappointed by products that are designed by engineers who assume their customers are engineers.
What game do you have the fondest memories of?
Myst.
Which tech trend do you wish would go away?
Spam.
What creation are you most proud of?
LinnStrument.
What’s the best piece of advice you’ve ever received?
Keep it simple.
What is your current obsession?
VR.
What do you do when you need to focus?
Breathe. Calm down.
What do you do when you’re feeling stuck?
I try to shift my perspective.
When was the last time you went somewhere without your phone?
I never go anywhere without my phone. Maybe swimming.
What’s the last piece of physical media you bought?
That would be a long time ago. I’ve only bought books, music, films, etc. in digital form for a long time.
What do you think is worth splurging on?
If someone made a VR headset with retina resolution, very high power, lots of beautiful open worlds, but it was expensive, I’d probably buy it.
What would the tagline for your biopic be?
“He created tools that allowed musicians to make better music.”
What’s the last GIF or meme you used?
This isn’t a GIF, but maybe it’s a meme:
( ͡° ͜ʖ ͡°)
Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.
NEWYou can now listen to Fox News articles!
A robot mower sounds like the kind of yard tool that should make life easier. It cuts the grass, saves you time and quietly handles a chore most people would rather avoid.
But a new independent security report raises a bigger concern about what may be happening behind the scenes. Security researcher Andreas Makris says Yarbo robots, which include autonomous lawn mowers and snow blowers, contained serious flaws that could expose owners to remote access, live camera viewing and Wi-Fi credential theft. The report says roughly 6,000 robots are currently affected.
Yarbo has since responded through its Security Center, saying the core technical findings are accurate and that it has started rolling out security fixes. Still, the report raises important questions about how much access smart yard devices should have inside your home network.
Sign up for my FREE CyberGuy Report
- Get my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox.
- For simple, real-world ways to spot scams early and stay protected, visit CyberGuy.com – trusted by millions who watch CyberGuy on TV daily.
- Plus, you’ll get instant access to my Ultimate Scam Survival Guide free when you join.
SMART HOME HACKING FEARS: WHAT’S REAL AND WHAT’S HYPE
A robot mower connected to home Wi-Fi can create security risks if remote access controls are weak or unclear. (Yarbo)
Yarbo robot security risk: What the report claims
Makris says Yarbo robots ship with a persistent remote access setup that uses a tunnel to reach the robot over the internet. According to the report, the robots also include a hardcoded root password shared across the fleet and a remote connection method tied to the robot’s serial number. That is important because “root” access gives someone deep control over the device. In simple terms, it can mean administrator-level access to the system inside the robot. The report also says the remote tunnel runs automatically, can restart itself if stopped and may return if removed. That raises a major concern for owners because they may not have a simple switch in the app to shut it off.
Why a robot mower could put your home network at risk
Smart devices often need internet access to work. App controls, software updates, diagnostics and support all depend on that connection. However, Makris claims Yarbo’s setup creates a much riskier situation. He says remote access appears to be built into every robot, rather than turned on only when an owner asks for help. The report says an attacker with the right information could potentially reach a robot remotely, access internal functions and use it as a foothold on the owner’s network. So while a robot mower may seem harmless as it cuts grass, rolls through the yard or parks near the garage, that same machine can also connect to your Wi-Fi, carry cameras and sit close to your home every day.
5 WORRISOME PRIVACY CLAUSES HIDDEN IN SMART HOME DEVICES
The Yarbo report raises concerns about remote access, live camera feeds and saved Wi-Fi credentials on connected yard robots. (Yarbo)
Yarbo camera access concerns for homeowners
According to the report, Yarbo robots can have multiple camera feeds. Makris says that if someone gained root access through the remote tunnel, they could potentially view the robot’s surroundings remotely. That could include a driveway, backyard, entryway, garage area or outdoor space where your family spends time. For homeowners, this concern goes beyond a glitch. A camera-equipped device outside your home deserves the same scrutiny as a camera inside your home.
How saved Wi-Fi passwords could be exposed
The report also says an attacker with root access could retrieve saved Wi-Fi credentials from the robot’s system. That would be a serious issue because many homes use one main Wi-Fi network for phones, laptops, tablets, smart TVs, security devices and more. Once someone has your Wi-Fi password, the risk can spread. They may try to reach other connected devices or look for weak spots that were never meant to face the internet. This is why connected outdoor equipment should never get a free pass. A lawn robot may be housed outside or in the garage, but its network access can reach inside.
What Yarbo says now
After Makris published his report, Yarbo posted a response to its Security Center page on its website. The company said the report identified serious vulnerabilities in its remote diagnostic, credential management and data-handling systems. Yarbo co-founder Kenneth Kohlmann also said the “core technical findings are accurate” and acknowledged that the company’s initial response did not reflect the seriousness of the issues.
Yarbo says the problems primarily involved historical design choices in parts of its remote diagnostic, access management and data-handling systems. The company also said some legacy support tools did not give users enough visibility or control. Yarbo said some authentication and credential systems did not meet its current security expectations.
A NEW SECURITY SEAL OF APPROVAL IS COMING TO YOUR SMART HOME GADGETS
Security experts recommend keeping smart yard devices on a guest network instead of your main home Wi-Fi. (Yarbo)
What Yarbo says it has fixed
Yarbo says it has taken several remediation steps since the report was published. According to the company, it has retired historical fleet-level root credentials, revoked shared FRP remote-access credentials and disabled related FRP server-side connection paths.
The company also says updated versions of the Yarbo mobile app no longer contain static credentials or embedded access mechanisms capable of directly authenticating against backend services. Yarbo says it has removed reporting scripts, legacy dependencies and non-essential network configurations that no longer served a necessary product function.
However, Yarbo says more work remains. The company says it is rebuilding its credential management system so any remaining shared-credential models can be replaced with individually scoped, per-device credentials. Each credential would support independent rotation and revocation.
Why Yarbo data connections raise privacy questions
The report also points to connections involving Hanyangtech, Yarbo’s Shenzhen-based parent company, along with ByteDance Feishu, Tencent TDMQ and Chinese DNS resolvers. Makris says some robot telemetry can be sent to ByteDance’s Feishu platform and that certain infrastructure choices are built into the firmware.
Yarbo now says it has removed reporting scripts, legacy dependencies and non-essential network configurations that no longer served a necessary operational or product function. The company also says historical servers and legacy access channels will continue to be phased out as part of its remediation work.
The core issue is transparency. Owners should know where their devices send data, which companies can access it and whether those connections are essential for normal use. That level of clarity matters even more for devices with cameras, location data and access to home networks.
What this means for you
If you own a Yarbo robot, this report means you should treat it like any other connected device with cameras, location data and access to your home Wi-Fi. Yarbo says it is pushing security updates automatically to connected devices. That means owners should connect their Yarbo long enough to receive the latest security update. After that, consider moving it back to a guest network or an isolated smart-device network.
CyberGuy reached out to Yarbo, and a representative said the company encouraged readers to refer to the Security Center at yarbo.com/pages/yarbo-security-center for the latest verified information and ongoing updates.
How Yarbo owners can reduce the risk
You may not be able to control everything happening inside the robot, but you can take a few practical steps to limit what it can reach on your home network.
1) Put the robot on a guest network
Do not keep your robot mower on the same network as your laptop, phone or security cameras. Use a guest network or a separate smart-device network if your router supports it.
2) Change your main Wi-Fi password if you are concerned
If your robot has connected to your main Wi-Fi and you are worried about exposure, change the Wi-Fi password. Use a strong, unique password and store it in a trusted password manager so you do not have to reuse or remember it. Then reconnect only trusted devices. Check out the best expert-reviewed password managers of 2026 at Cyberguy.com
3) Check your router for unknown devices
Open your router app or admin page and review connected devices. Look for anything unfamiliar. Remove devices you do not recognize.
4) Limit what the robot can access
Some routers let you isolate guest devices. Turn that on when available. This can keep the robot from seeing other devices on your network.
5) Ask Yarbo for specific answers
Owners should ask what remote diagnostic access remains, whether credentials are now unique per robot and whether the company will provide a true off switch for remote diagnostics.
6) Keep the robot updated, but stay cautious
Yarbo says security updates are delivered automatically once devices connect to the internet. Connect the robot through a guest network or an isolated smart-device network so it can receive the latest update without giving it access to your main devices.
Join CyberGuy Live: Lock Down Your Phone in 30 Minutes (Saturday, June 13, 10 am ET)
Your phone holds your email, passwords, photos, banking apps and personal data. In this free, live online class, Kurt the CyberGuy will walk you step by step through simple phone security fixes you can do in real time. You’ll learn how to improve your privacy settings, spot the latest phone scams, use trusted security tools and walk away with a simple checklist to stay protected. Register here: CyberGuyLive.com
Kurt’s key takeaways
The Yarbo report is a reminder that convenience can come with hidden access. A robot mower may seem like a helpful yard tool, but under the hood, it can act like a connected computer with cameras, location data and a path into your network. The biggest concern is control. Owners need to know who can reach their devices, when remote access turns on and whether they can shut it off. A company should not expect you to trust a black box sitting on your Wi-Fi. If you own one of these robots, isolate it from your main network and push Yarbo for clear answers. If you are shopping for any smart yard device, ask about security before you ask about battery life.
Would you let a smart yard robot onto your Wi-Fi if the company could not clearly explain who can access it and when? Let us know by writing to us at Cyberguy.com
CLICK HERE TO DOWNLOAD THE FOX NEWS APP
Sign up for my FREE CyberGuy Report
- Get my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox.
- For simple, real-world ways to spot scams early and stay protected, visit CyberGuy.com – trusted by millions who watch CyberGuy on TV daily.
- Plus, you’ll get instant access to my Ultimate Scam Survival Guide free when you join.
Copyright 2026 CyberGuy.com. All rights reserved.
Google’s AI Overviews are running into an interesting problem right now. Earlier on Friday, if you searched for the term “disregard,” the AI Overview section would include a response like what you’d see from a more traditional AI chatbot instead of the typical AI summary, as spotted on X. As you can see in the image at the top of this story, I got an AI Overview response that said, “Got it. If you need anything else or have a new question later, just let me know!”
As of Friday afternoon, however, Google isn’t showing an AI Overview for the term “disregard” at all — instead, it shows a list of news stories about the issue first. Google hasn’t replied to our requests for comment. In a statement to Android Authority, a spokesperson said that “We’re aware that AI Overviews are misinterpreting some action-related queries, and we’re working on a fix, which will roll out soon.”
AI Overviews haven’t just been tripping up over the word “disregard.” When searching for “ignore,” Google’s AI Overview section showed the following message to a Verge colleague:
Message received! I’m here and ready to help. What would you like to focus on today? Just let me know if there’s a specific topic, task, or question you’d like to explore.
When they searched “skip,” the AI Overview section said:
It looks like your message was just a test or a typo! Feel free to ask a question, share a prompt, or let me know how I can help you with your tasks today. I’m ready whenever you are!
As of Friday afternoon, Google is still showing me AI Overviews with broken responses when I search for “ignore” and “skip.”
As funny as this all is, it’s almost certainly just some kind of bug — I expect Google will fix it soon enough. Maybe Google Search itself is tired after everything that happened at Google I/O.
Updates, May 22nd: Google now isn’t showing AI Overviews for “disregard.” Also added a Google statement.
-
New York21 minutes agoBrooklyn Man Wedged in Upstate New York Cave Is Rescued After 6 Hours
-
Los Angeles, Ca27 minutes agoGrandmother hospitalized after hit-and-run slams into her in San Fernando
-
Detroit, MI51 minutes ago3 injured in separate weekend shootings in Detroit, police say
-
San Francisco, CA1 hour agoThe Store Cats of San Francisco
-
Dallas, TX1 hour agoFC Dallas vs Colorado Rapids: Lineup notes 📝
-
Miami, FL1 hour agoDeputies investigating shooting in NW Miami-Dade that left a man in critical condition – WSVN 7News | Miami News, Weather, Sports | Fort Lauderdale
-
Boston, MA1 hour agoMan killed after MBTA commuter rail train struck his vehicle at Braintree crossing – The Boston Globe
-
Denver, CO1 hour agoDenver weather: Warmer weather for Memorial Day with an isolated storm
