Eshu Marneedi

Apple announced its earnings for the second quarter on Thursday, and Tim Cook, the company’s chief executive, interviewed with CNBC. CNBC wrote the following:

Cook also said Apple has “big plans to announce” from an “AI point of view” during its iPad event next week as well as at the company’s annual developer conference in June.

I don’t even understand why this was reported on, because artificial intelligence is the new craze both in Silicon Valley and Wall Street. Of course the chief executive of the world’s second-largest technology company — which reported revenue down 4 percent this quarter — would try to pump his stock price, and of course he would do that by saying there will be an AI-related announcement at next week’s hotly anticipated Apple event. It makes logical sense from a business perspective: If Cook can motion investors to hold off on dumping Apple stock this week, he can launch new iPads next week, point to the sales numbers, and watch the stock hike again. That is his job.

Later, CNBC retracted its original quote, but gave the full context to Zac Hall, editor at large at 9to5Mac, somehow:

We’re getting into a period of time here where we’re extremely excited like I’m in the edge of my seat literally because next week, we’ve got a product event that we’re excited about. And then just a few weeks thereafter, we’ve got the… Worldwide Developers Conference coming up and we’ve got some big plans to announce in both of these events. From an AI point of view…

Cook is not saying there will be AI-related announcements at these events, he is just saying (a) that there are “big plans” and (b) there will be announcements some time between now and the end of eternity “from an AI point of view.” Those are mutually exclusive statements — it is foolish to assume otherwise because Cook is well-trained before he sits in front of the media. Apple never reveals what it will announce before an event, even when it would be in the interest of the stock price.

So, that all begs the question: Will there be AI at next week’s event or not? It’s impossible to say conclusively, but I think there will certainly be mentions of AI during the presentation. However, I do not believe Apple will announce AI software of its own just a month before WWDC, where software is usually debuted. I imagine the AI references will be limited to passing mentions of how the new iPads Pro are “great for AI computing” and how you can run AI models with apps on the App Store, just like the “Scary fast” Apple event from October, where the company announced the M3 MacBooks Pro. The mentions will exist to please investors and to hold them off just a bit longer for WWDC, where the big-ticket AI features will be introduced via iOS 18.

Thursday’s keynote will not be a preview of AI features — or at least, so I think. Instead, it looks like it’ll serve as a filler event to build anticipation for the true announcements coming in the summer, while also finally refreshing the iPads, which is long overdue. This scenario also takes into account Mark Gurman’s report for Bloomberg on Sunday that said Apple will ship the M4 in the new iPads Pro: M4 or not, this event is slated to be hardware-focused, and I think the only AI references next week will exist to appease Wall Street. My final take: No AI at next week’s event.

Mishaal Rahman, reporting for Android Authority on Tuesday:

If everything an AI gadget like the Rabbit R1 can do can be replicated by an Android app, then why aren’t these companies simply releasing an app instead of hardware that costs hundreds of dollars, requires a separate mobile data plan to be useful, and has terrible battery life? It turns out that’s exactly what Rabbit has done… sort of.

See, it turns out that the Rabbit R1 seems to run Android under the hood and the entire interface users interact with is powered by a single Android app. A tipster shared the Rabbit R1’s launcher APK with us, and with a bit of tinkering, we managed to install it on an Android phone, specifically a Pixel 6a.

Once installed, we were able to set up our Android phone as if it were a Rabbit R1. The volume up key on our phone corresponds to the Rabbit R1’s hardware key, allowing us to proceed through the setup wizard, create a “rabbithole” account, and start talking to the AI assistant. Since the Rabbit R1 has a significantly smaller and lower resolution display than the Pixel 6a, the home screen interface only took up a tiny portion of the phone’s display. Still, we were able to fire off a question to the AI assistant as if we were using actual Rabbit R1 hardware, as you can see in the video embedded below.

The Rabbit R1, just like the Humane Ai Pin, is nothing more than a shiny object designed to attract hungry venture capitalists. The entire device is an Android app, a low-end MediaTek processor, and a ChatGPT voice interface wrapped up in a fancy orange trench coat — in other words, nothing more than a grift that retails for $200. I’ve said this time and time again: These artificial intelligence-powered “gadgets” are VC money funnels whose entire job is to turn profits then disappear six months later when Apple and Google add more broad AI functionality to their mobile operating systems. In the bustle of the post-October 2022 AI sphere, Rabbit raised a few million dollars in Los Angeles, built together an Android app with a rabbit animation, bulk bought some off-the-shelf cheap electronics from China, engineered a bright orange case, put the parts together, made its founder dress up like an off-brand Steve Jobs, and poof, orders started flooding in by the thousands. Ridiculous.

The Rabbit R1, in many ways, is more insulting than the Humane Ai Pin, which I’ve already bashed enough. It is significantly more affordable, priced at $200 with no subscription — unlike Humane’s $700, $24-a-month product — but it is quite literally worse than the Ai Pin from Rabbit’s chief rival VC funnel in every metric. The entire device, as Marques Brownlee, a YouTuber better known as MKBHD, demonstrated in his excellent review of the device, is a ChatGPT wrapper with an ultra-low-end camera and a knob — or wheel, rather — used in favor of a touch screen presumably to make it seem less like a smartphone. In practice, it is a bad, low-end smartphone that does one thing — and only one task — extraordinarily poorly, consistently flubbing answers and taking seconds to respond. It is a smartphone that does everything poorly aside from looking great. (Teenage Engineering designed the Rabbit R1; I’ll give the product design props.) I am astonished that we are living in a world where this $200 low-end Android smartphone is receiving so much media attention.

Rahman contacted Jesse Lyu, Rabbit’s chief executive and co-founder, for comment on his article, and Lyu, grifter-in-chief at Rabbit, naturally denied the accusations in the stupidest way possible. I don’t even understand how this made it to publication; it’s genuinely laughable. Lyu’s justification for the device is that Rabbit sends data and queries to servers — presumably its own servers — for processing. Here is a non-comprehensive list of iOS apps with large language models built in that send data to the web for processing: OpenAI’s ChatGPT, Microsoft Copilot, Anthropic Claude, and Perplexity — also known as every single AI processing app made by a large corporation because it is all but impossible to run LLMs on even the most sophisticated, powerful smartphone processors, let alone any random inexpensive MediaTek chip, such as found on the R1. The Rabbit R1 is an Android app that exchanges data with the internet with a cellular radio and some network calls. Any 15-year-old could engineer this in two weeks from the comfort of their bedroom.

I aggressively smeared the Humane Ai Pin not because I thought it was a grift, but because I thought it had no reason to exist. I thought and still think that Humane built an attractive piece of hardware and that the company still has conviction in creating a product akin to the smartphone in the hopes of eventually eclipsing it. (I think this entire idea is flawed, and that Humane will eventually go bankrupt, but at least Humane’s founders are set on their ambition.) Rabbit as an entire company, by stark contrast, is built on a throne of lies and scams: It came out of the woodwork randomly during the Consumer Electronics Show in January after raising $10 million the month prior from over-zealous VC firms, threw a launch party in New York with influencers and press alike, then shipped an Android app to consumers for $200. It’s a cheap smear of hard-working, dedicated hardware markers; it makes a mockery of true innovators in our very complicated technology climate in 2024. These “smartphone replacement” VC attractions ought to be bankrupt by, if not right after, June.

Ridiculous, but quite possible. Mark Gurman, reporting for Bloomberg in his Power On newsletter:

Earlier this month, I broke the news that Apple is accelerating its computer processor upgrades and plans to release the M4 chip later this year alongside new iMacs, MacBook Pros, and Mac minis. The big change with the M4: A new neural engine will pave the way for fresh AI capabilities. Now here’s another development. This year’s Macs may not be the only AI-driven devices with M4 chips.

I’m hearing there is a strong possibility that the chip in the new iPad Pro will be the M4, not the M3. Better yet, I believe Apple will position the tablet as its first truly AI-powered device — and that it will tout each new product from then on as an AI device. This, of course, is all in response to the AI craze that has swept the tech industry over the last couple years.

By introducing the new iPad Pro ahead of its Worldwide Developers Conference in June, Apple could lay out its AI chip strategy without distraction. Then, at WWDC, it could focus on how the M4 chip and new iPad Pros will take advantage of the AI software and services coming as part of iPadOS 18 later this year. I fully expect Apple to position the A18 chip in the iPhone 16 line as built around AI as well.

To be fair, though, these new products aren’t engineered and developed entirely around AI. This is partly about marketing. Hardware with even more impressive capabilities is further out. As I’ve reported, Apple is working on a table-top iPad connected to a robotic arm, as well as a home robot.

For context, the M3 line of processors debuted in late October 2023, so it has only been roughly six months since the latest generation of Apple’s high-end processors came into the market. Every single bone in my body disagrees viscerally with every aspect of this rumor — it does not make sense from any logical perspective whatsoever because there is no way Apple would sell an iPad Pro that is faster and more capable than the MacBook Air and base-model MacBook Pro. It would be genuinely embarrassing for it to sell a device that runs iPadOS — a moderately enhanced version of iOS — with a more powerful chip than the $1,600 MacBook Pro.

That bit of illogical thinking is, however, small compared to the timeframes we’re working with here: Apple has never produced two full generations of Apple silicon only six months apart. With Taiwan Semiconductor Manufacturing Company’s factories booked trying to meet M3 3-nanometer process node demand, I have a tough time believing TSMC can fabricate more 3-nm chips to meet the demand for iPads Pro and Mac laptops. Also, the M3 processor is based on the architecture of the A17 Pro, which first debuted in September in iPhones 15 Pro, so the M4 would have to be based on the eventual A18 Pro (or A18, whatever it may be called), which has not even been announced yet — it will be announced this September. Historically, Apple has always based the Mac’s Apple silicon chip on that year’s iPhone chip.

And about the “artificial intelligence” focus: I genuinely can’t see Apple marketing the new iPads Pro — which are slated to debut at Apple’s virtual May 7 event — as “AI-focused” without first announcing AI features as part of iPadOS in June, at the Worldwide Developers Conference. What would potential buyers do with the AI-focused Neural Engine in iPadOS? This entire rumor is bending my mind because it seems so impossible and brazen. The M4, historically speaking, has no business being in these new iPads Pro — period — because no publicly available software can take advantage of the new Neural Engine on iPadOS, the M4 would stretch TSMC’s fabrication facilities to their max, and the M4’s debut in May would not line up with Apple’s product timelines. It’s completely nonsensical.

I trust Gurman — I truly do. There is not a time in recent memory when he has been wrong. But Gurman has been seesawing non-stop on this Apple event, which he earlier said wouldn’t even be an event in the first place. He also said that the iPads would be announced in March or April, but the event is taking place in May. Though he corrected some of these rumors later on, I heavily doubt his reporting here. If Apple truly does announce the M4 chip in the new iPads Pro in May, I won’t be shocked, because Gurman said it — but for now, I’m choosing to take this rumor with a grain of salt.

Also, one final note from Gurman’s newsletter this week, serving as some follow-up to my writing on Saturday about Apple Vision Pro:

Vision Pro demand has dropped considerably at many Apple stores. One retail employee says they haven’t seen one Vision Pro purchase in weeks and that the number of returns equaled the device’s sales in the first month that it was available.

Yikes.

Apple emailed me on Friday asking me how I’ve been enjoying my Apple Vision Pro. Here is what I wrote to the company.

Apple: Tell us why you’re not satisfied with your Apple Vision Pro.

Me: My main problem with Apple Vision Pro is its lack of content. Plain and simple, there is not much to do with it. I bought it because part of my job is to write about technology, but I probably wouldn’t have if I didn’t have that motive. Apple Vision Pro, a few months later, suffers from a lack of a use case. Everyone knows what to do with their iPhone, Mac, or iPad — but Apple Vision Pro? You might watch a movie, play a game or two, or fiddle around with the operating system, visionOS. But it’s restricted computing-wise in the same way the iPad is, which makes it impossible to use as a Mac replacement; it’s not sharable in the same way a TV is; and it’s not even nearly as easy to use as the iPad or iPhone, both of which can be used just by picking them up and tapping the screen.

Apple Vision Pro, from the get-go, is complicated to use. You have to ensure it has enough charge to use it since it doesn’t have very good standby battery life, then connect the battery, adjust the strap, place it on your head, adjust your hair, adjust the strap again, ensure it fits well, then unlock it and begin using it. For all of that to be worth it, there needs to be a seriously compelling reason to put it on. If something can be done 90 percent as enjoyably with an iPad or iPhone, most people — including myself — will just use that over Apple Vision Pro. Each time I use it, it’s a game of calculus: Is it worth it to do all this for the 10 percent of joy I’ll get?

What Apple wants people to think is that Apple Vision Pro does more than a traditional computer — not that Apple Vision Pro does what a traditional computer does but better. In practice, Apple Vision Pro does what a normal computer or mobile device does — but exponentially worse. Not only is it a hassle to use most of the time, but its software — based on iPadOS — functions in the same crippled ways iPadOS does. And with the lack of enthusiasm from third-party developers, the product is even more lackluster.

None of this is to say Apple Vision Pro is a lackluster product — it clearly isn’t. Every time I use it, I generally enjoy my time fiddling with new applications and experiences. But Apple sells many computers in various form factors, and most of those devices do the job of Apple Vision Pro just good enough that it’s not worth the effort to wear the headset most of the time. This is a solvable problem: Just make more content. Apple needs to incentivize third-party developers to make more experiences, produce more content itself, and improve the software experience to make computing better. For example, multitasking is impossible on visionOS, even though the inherent nature of Apple Vision Pro could make it a more capable Mac with an infinite amount of screen space. Why is using windows such a hassle on visionOS when they could be spectacular on this revolutionary spatial computer?

I particularly enjoyed the Major League Soccer highlight reel published in March and some of the other Apple-made immersive videos available through the TV app on visionOS. There should be much, much more of that kind of content available for paying subscribers. I know Apple has enthusiasm for this product, but looking at visionOS does not make that apparent.

Apple: What types of video content are you most interested in watching on Apple Vision Pro?

Me: Immersive video content, such as the MLS soccer highlight reel, available to all Apple Vision Pro users. Flat, 2D content isn’t as appealing because other devices can view it just fine, but immersive content is absolutely outstanding.

Apple: What one thing, if anything, would you add to or change about Apple Vision Pro?

Me: Make it lighter. The weight adds so much discomfort to using the product. It’s hard to use while lying down, uncomfortable while perfectly upright, and moderately uncomfortable while in a slightly reclined position — which is currently the most advisable way to use it. It rests on the cheeks and forehead evenly, but also terribly. It’s fatiguing to use for long periods of time. It needs to be lighter.

Sapna Maheshwari, David McCabe, and Cecilia Kang, reporting for The New York Times:

Just over a year ago, lawmakers displayed a rare show of bipartisanship when they grilled Shou Chew, TikTok’s chief executive, about the video app’s ties to China. Their harsh questioning suggested that Washington was gearing up to force the company to sever ties with its Chinese owner — or even ban the app.

Then came mostly silence. Little emerged from the House committee that held the hearing, and a proposal to enable the administration to force a sale or ban TikTok fizzled in the Senate.

But behind the scenes, a tiny group of lawmakers began plotting a secretive effort that culminated on Wednesday, when President Biden signed a bill that forces TikTok to be sold by its Chinese owner, ByteDance, or risk getting banned. The measure, which the Senate passed late Tuesday, upends the future of an app that claims 170 million users in the United States and that touches virtually every aspect of American life.

For nearly a year, lawmakers and some of their aides worked to write a version of the bill, concealing their efforts to avoid setting off TikTok’s lobbying might. To bulletproof the bill from expected legal challenges and persuade uncertain lawmakers, the group worked with the Justice Department and White House.

And the last stage — a race to the president’s desk that led some aides to nickname the bill the “Thunder Run” — played out in seven weeks from when it was publicly introduced, remarkably fast for Washington.

“Thunder Run” is a McCarthyist First Amendment violation straight from the Second Red Scare. This law was only able to pass because it was attached to a much-needed foreign aid appropriations bill, funding Ukraine and Israel and providing billions of dollars of humanitarian aid to vulnerable populations. It was included to ensure broad support within the Republican party — a compromise House Speaker Mike Johnson of Louisiana had to make to ensure members of his party would support the aid package. Republicans aren’t known for being the smartest of people, but it’s wrong to solely place the blame on their antics this time. Moderate Democrats played a hand in pushing the bill over the finish line, effectively stripping half the country of their First Amendment rights.

The government has yet to provide concrete evidence of a national security threat, which is strange, because the only sound legal argument for this law is national security. ByteDance is owned and effectively controlled by the Chinese Communist Party, and there is potential for the Chinese to compromise the security of the United States with access to hundreds of millions of Americans’ phones. Yet, there is zero evidence of this happening in actuality — I’m not saying that it isn’t happening, but there is no evidence for the public to see. When this law is challenged in court — and it absolutely will be — this will be the primary aspect of the case, as silencing speech because “terrorist content is promoted” is easily one of the most illegal things Congress can ever do. From the First Amendment of the Constitution:

Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.

That is the entire amendment, and there is no need for more because it’s extremely descriptive. Congress cannot make a law that prohibits the “free exercise of speech” — period, no matter what that speech is — unless it’s in the interests of the government’s job of protecting the American people. If there is no attempt to prove that this law is in support of that job, this bill should get tossed straight into a fire. There is a reason that yelling “I’m going to bomb this airplane!” in an airport is not considered “free exercise of speech,” so the same logic applies here: The government must provide tangible proof that TikTok’s continued Chinese ownership is an equivalent danger to someone yelling “I’m going to bomb this airplane!”

When the government makes that argument, it cannot use the speech on TikTok as proof, because no matter the speech, speech alone cannot impede national security. For example, 4chan is filled to the brim with unfavorable, illegal speech, but the government can’t directly punish the platform owner for “threatening national security” due to Section 230 of the Communications Acts of 1934 and 1996 which give platform owners immunity from what people say on them. Forcing the divestiture of a company because of speech on a platform — illegal or not; supporting terrorists or communists is not illegal — is a blatant violation of the First Amendment and has been proven case after case before the Supreme Court.

That leaves just one more question: Why won’t ByteDance divest TikTok? It’s a good question that I’ve also been pondering because I’m not one to favor Chinese control over one of the largest social media platforms in the United States. I want a TikTok divestiture, but I don’t want it to be forced by the government. I’ve come to this conclusion: ByteDance won’t ever divest TikTok not because it doesn’t want to, but because it legally cannot due to its Chinese ownership and control. The CCP is truly an authoritarian government, and TikTok is its best way of manipulating the public’s image of it, so it’s willing to let a Chinese company suffer financially if it means displaying a facade of strength in front of the United States. The CCP doesn’t care about the money ByteDance makes — it’s communist — but it does care about the data ByteDance generates. It wants power, and the best way for it to truly change the public’s perception of it is by threatening the public’s favorite social media platform.

Naturally, if TikTok vanishes in a year — a prospect that I think is still thoroughly unlikely — Americans will solely place the blame on their government, not on TikTok or China. And that point of contention between Americans and their government is exactly the reason why China doesn’t want to divest TikTok. The Chinese government wants power and strength; it wants to change the way Americans perceive it across the Pacific. This bill just gave China a brand new, effective strategy. Nice work, Washington — you’ve been outsmarted by Beijing again.

Viola Zhou on Tuesday reported an extremely thorough, quite lengthy piece about the cultural differences and headaches in Taiwan Semiconductor Manufacturing Company’s new semiconductor fabrication plant in Arizona, typically called a “fab.” This was my favorite part that truly exemplifies the differences between work culture in the East and the West:

At Fab 18, nearly all communication took place in Taiwanese and Mandarin Chinese, the two most widely spoken languages in Taiwan. The Americans found it difficult to understand meetings, production guidelines, and chatter among local engineers. In theory, every American was supposed to have a Taiwanese buddy — a future Arizona worker who would help them navigate the workplace. But the Americans said their buddies were often too busy to help with translations, or else not familiar enough with the technical processes because they were freshly transferred from other production lines.

Many trainees, including Bruce, relied on Google Translate to get through the day, with mixed results. Technical terms and images were hard to decipher. One American engineer said that because staff were not allowed to upload work materials to Google, he tried to translate documents by copying Chinese text into a handwriting recognition program. It didn’t work very well…

TSMC’s work culture is notoriously rigorous, even by Taiwanese standards. Former executives have hailed the Confucian culture, which promotes diligence and respect for authority, as well as Taiwan’s strict work ethic as key to the company’s success. Chang, speaking last year about Taiwan’s competitiveness compared to the U.S., said that “if [a machine] breaks down at one in the morning, in the U.S. it will be fixed in the next morning. But in Taiwan, it will be fixed at 2 a.m.” And, he added, the wife of a Taiwanese engineer would “go back to sleep without saying another word.”

During their visit, the Americans got a taste of the company’s intense work culture. To avoid intellectual property leaks, staff were banned from using personal devices inside the factory. Instead, they were given company phones, dubbed “T phones,” that couldn’t be connected to most messaging apps or social media. In one department, managers sometimes applied what they called “stress tests” by announcing assignments due the same day or week, to make sure the Americans were able to meet tight deadlines and sacrifice personal time like Taiwanese workers, two engineers told Rest of World. Managers shamed American workers in front of their peers, sometimes by suggesting they quit engineering, one employee said.

This story reported a challenge with chip manufacturing in the United States that I hadn’t considered until now: cultural differences. Semiconductor manufacturing — like any manufacturing — is a very male-dominated industry, and also one that happens to be low-paying in the East with frankly atrocious working conditions, so bringing that industry to the West, where workers expect shorter work hours and more humane treatment, is difficult. Later in the story, Zhou reports how female co-workers were mistreated, how employees weren’t permitted to bring their phones to work, and how there was a disconnect between the Taiwanese managers and American workers. I suggest you read the piece in its entirety.

The obvious solution to this problem is for the managers themselves to be American, but that’s unfeasible as of now because someone needs to train those managers for the job. Unlike other multinational corporations that have operated in the United States for decades, TSMC needs to first train the Americans to a level of seniority before they can take over the plants entirely. This creates a major bottleneck in the form of a chicken-and-egg problem: TSMC needs American workers to function as supervisors, but Americans want to leave their jobs at TSMC because of the Taiwanese managers. It’s a tough problem to solve, but one that I think can be ironed out with some changes to C-suite leadership.

Unlike the rank-and-file managers who presume authority over the day-to-day operations of the Arizona plant, C-suite executives in Taiwan should be able to rectify this issue by training the Taiwanese managers better. The onus shouldn’t be on the Americans to change — the Taiwanese need to better adapt to the Americans’ way of work. The United States will never be Taiwan or China, and I think TSMC management understands that. The U.S. government is providing the funding, the clients are providing the orders — now, it’s time for TSMC to change how it manages its employees to better its recruiting strategy.

These managers need a change of attitude because that’s inherently the main job of management — to adapt to workers’ preferences. If TSMC doesn’t change the way it controls its workers, the projects will fall apart fast. Funding comes once projects get off the ground, especially if President Biden wins re-election this fall, but for the projects to succeed, workers need to be satisfied. TSMC’s Glassdoor ratings are not desirable for the No. 1 semiconductor manufacturer in the world. Americans care a lot about the culture of the company for which they work, and TSMC needs to understand that and better adapt to American work culture.

Alex Heath, reporting a politically heated story for The Verge on Wednesday:

Google fired 28 employees in connection with sit-in protests at two of its offices this week, according to an internal memo obtained by The Verge. The firings come after 9 employees were suspended and then arrested in New York and California on Tuesday.

The fired employees were involved in protesting Google’s involvement in Project Nimbus, a $1.2 billion Israeli government cloud contract that also includes Amazon. Some of them occupied the office of Google Cloud CEO Thomas Kurian until they were forcibly removed by law enforcement. Last month, Google fired another employee for protesting the contract during a company presentation in Israel.

In a memo sent to all employees on Wednesday, Chris Rackow, Google’s head of global security, said that “behavior like this has no place in our workplace and we will not tolerate it…”

He also warned that the company would take more action if needed: “The overwhelming majority of our employees do the right thing. If you’re one of the few who are tempted to think we’re going to overlook conduct that violates our policies, think again. The company takes this extremely seriously, and we will continue to apply our longstanding policies to take action against disruptive behavior — up to and including termination.”

I say this as a liberal, somewhat progressive person politically: Google had every single right to fire each last one of these “protestors.” Gaby Del Valle for The Verge also reported earlier on Wednesday that the demonstrators occupied Google’s offices illegally, even after they were asked to leave by management, which led the company to call the police to arrest nine of them. A man who works for Google said to the protestors, as quoted by The Verge: “We’re asking you to leave again for the last time.” Then, when they stayed, a police officer offered the demonstrators a plea deal of sorts: “Listen, we’ll let you walk out the door right now — it’s a non-issue if you’re willing to go. If not, you’re going to be arrested for trespassing.”

Every one of the 28 employees who was fired Wednesday evening was given multiple chances to leave a secured, locked building that they were not permitted to use for demonstrations, but they flagrantly violated orders given by a representative for the owner of the building, who also happens to be their employer. If that isn’t a reason for termination of employment, I do not know what is. This type of lunatic behavior would get any employee fired because the protestors engaged in illegal activity. It’s trespassing — illegally occupying a building when the owner gives repeated instructions to leave. When that owner is your employer, termination is a fair punishment.

Those complaining about “free speech,” like many right-wingers, do not actually understand what free speech means in this context. Employees are permitted to protest, especially against their employer, for a variety of reasons. That is protected speech under the Constitution as long as the protestors don’t cause a disturbance, also known as “disorderly conduct” in criminal law. Trespassing goes beyond disorderly conduct; it’s a felony offense to occupy a building when disallowed. From what is known currently, it doesn’t seem that any protestors were charged with crimes — they were simply fired for staging a rogue protest against their employer. Trespassing and causing a disturbance is enough of a reason to fire an employee according to the law. If an employee randomly stood up from their desk and began shouting, they would be reprimanded.

Online activists are calling this protest “peaceful,” when breaking the law is exactly what makes a protest the opposite of “peaceful.” No matter what the protest was for — Black Lives Matter, LGBTQ rights, or Israel’s military campaign in Gaza — a protest in a building where demonstrators are unauthorized to be is illegal, and therefore, punishable. Punishing an employee for failing to obey orders is a right given by the Constitution’s First Amendment to Google, and a corporation disciplining an employee or contractor for psychotic behavior is protected legally under free speech rights in the United States. Nobody has a right to be upset about how the protestors were treated in this case — while their cause’s importance can be debated until the end of time, their actions are undoubtedly flawed.

I truly cannot believe people who managed to land a job at one of the world’s largest technology firms are so stupid that they occupied the private office of their chief executive as if they were rioters on January 6, when a mob of Republican supporters stormed the Capitol in Washington to stop the certification of the 2020 election. The blatant lawlessness exhibited in this protest is appalling and should be condemned in the strongest terms. A functioning democracy necessitates the right to protest, but this wasn’t any ordinary protest — it was a stunning spectacle of incompetence, mindlessness, and arrogance unlike one displayed in Silicon Valley before. “Big Tech” employees have protested via many walk-outs, sit-ins, and other protests to reject their employers' policies, but they have always done so peacefully and respectfully, inspiring change for everyone in a dignified manner. This was the complete opposite.

Condemning the protestors isn’t an endorsement of Israel’s actions in Gaza, Google’s deal with the Israeli government, or the U.S. government’s foreign policy with respect to Jerusalem. Anyone who appreciates dignity and the right to protest in the workplace should be ashamed of Wednesday’s events because they demonstrate the rogue, nonsensical mentality of pro-Palestinian mobsters who are taking rights away from peaceful protestors with their illegal actions. In addition to breaking trespassing laws, they chanted “From the river to the sea, Palestine will be free!” a slogan deemed antisemitic by many, presumably including Jewish people at Google, and pinned banners to the wall with antisemitic language. Google has the right to maintain company policy and remove employees who disrupt the workplace with hateful messages, regardless of what political ideologies those messages are linked to. Google is a company made of people, and if they feel disrespected, they have the right to take action.

For the sacred right to peacefully protest in the United States to remain intact, protestors need to remain respectful and mindful of their neighbors. If they aren’t, the country risks another January 6 — but this time, much, much worse. Political violence and lawlessness are never acceptable.

Gaby Del Valle, reporting for The Verge:

Sens. Wyden and Lummis introduce an amendment limiting FISA’s warrantless wiretapping powers. The amendment would reverse a provision included in the recent House bill reauthorizing Section 702 of FISA that expands the definition of “electronic communications service provider,” which critics say would force Americans to essentially spy for the government.

“Forcing ordinary Americans and small businesses to conduct secret, warrantless spying is what authoritarian countries do, not democracies,” Wyden said in a statement.

The House of Representatives recently re-authorized Section 702 of the Foreign Intelligence Surveillance Act, which, ironically, allows the Federal Bureau of Investigation to perform clandestine searches of American citizens on American soil without a warrant or their consent. The government justifies this by saying FISA is a critical national security tool to prevent foreign attacks on the United States, but the act is mostly used to surveil Americans, not foreigners. If FISA is not re-authorized by Saturday, the government can no longer spy on its citizens — foreign and abroad — however it pleases without even a warrant to justify its actions.

Senator Ron Wyden, Democrat of Oregon, and many progressive Democrats in the House who unsuccessfully voted against re-authorizing FISA have expressed concerns about this warrantless searching, but nevertheless, it seems increasingly likely that the original version of Section 702 will be re-authorized by the weekend. But the new version of FISA, which extends the program for two years, also includes an amendment courtesy of House Republicans to expand the definition of “electronic communications providers,” i.e., companies that are required to provide data when the government requests it. This definition, until now, has only included smartphone makers and other telecommunications companies, like Apple and Google, but the latest House amendment that passed also requires cloud computing firms to provide data.

This effectively means that the government, from now on, will be permitted to ask Amazon Web Services for all the user data associated with one account, which could include practically an entire person’s business life because most people and small businesses use one cloud provider to host their website or other business tools. Google Cloud, AWS, and even Apple’s iCloud servers are all susceptible to this unprecedented, warrantless searching, which, as Wyden and Senator Cynthia Lummis, Republican of Wyoming, say, is precisely what authoritarian regimes do. Russia and China employ this exact kind of surveillance to silence their people, and it is absolutely astounding to me that nobody has taken this law to court over a violation of the Fourth Amendment.

If the new amendment is passed — which seems increasingly likely based on the bipartisan support the latest bill has enjoyed in both the House and Senate — it would open Americans up to a brand new front of dangerous government surveillance akin to communist regimes like China. I don’t think Congress should axe FISA entirely, unlike many progressives and far-right “Make America Great Again” Republicans, but I think there should be an amendment to prevent warrantless searches. And Wyden and Lummis’ amendment in the Senate, in an ideal world, should pass, because allowing the government to surveil data stored in the cloud is blatant government overreach.

Foo Yun Chee, reporting for Reuters:

Meta Platforms and other large online platforms should give users an option to use their services for free without targeted advertising, EU privacy watchdog the European Data Protection Board said on Wednesday.

The EDPB’s opinion came after it was asked by national privacy regulators in the Netherlands, Norway, and Germany to look into consent or pay models adopted by large online platforms such as Meta.

“If controllers do opt to charge a fee for access to the ‘equivalent alternative’, they should give significant consideration to offering an additional alternative. This free alternative should be without behavioural advertising,” EDPB said in a statement.

The board’s ruling on Wednesday gives the European Commission, the European Union’s executive body, the ability to force Meta and “other large online platforms,” like Google, to provide their services for free without any targeted advertising. In other words, this outlandish decision allows the commission to dictate how a corporation makes money, even if the “recommended” method is non-viable. The non-technical equivalent of this ruling is the government telling a baker they can’t price their bread at $5 because it thinks that is too expensive. The government doesn’t actually know that the ingredient cost per loaf is $4.75, but it also doesn’t care to find out, so it just punishes the baker for selling the bread at a 25-cent profit even though the baker needs the 25 cents to continue their operations. It is entirely unfair.

The European Commission has been waiting on this ruling since March so it can begin to force Meta to offer non-targeted advertising for free, instead of forcing consumers to subscribe to the ad-free versions of Instagram and Facebook as Meta currently does in response to the Digital Markets Act, which went into effect in early March. E.U. users can choose to pay Meta 10 euros (around $11) monthly to remove all ads — including the targeted ones — because the bloc’s DMA forces “Big Tech” to offer users the ability to disable targeted advertising somehow. The subscription is Meta’s way of sneaking around the true intention of the legislation, which is for companies to offer a simple toggle switch for users to disable targeting for free. The commission didn’t like Meta’s clever idea, so it complained to the EDPB, which, naturally, ruled in the commission’s favor.

It’s important to keep in mind that the DMA does not specifically state that this scheme — which is known as a “pay or OK” tactic inside Brussels — is illegal, or that it shouldn’t be employed. The law simply requires there to be some way for users to opt out of targeted advertising, even if that method is asking for payment since the legislation says nothing about payment entirely. But commissioners, prominently Margrethe Vestager, the commission’s antitrust chief, have decided Meta’s perfectly legal compliance with the DMA is too unacceptable, so they have begun re-interpreting the law to enforce it at their whim. It’s the same old tactic the European Union has been playing for months. As I wrote in March, the commission is “playing a one-sided, rigged game while laughing manically in the corner at everyone falling face-flat on the ground.”

Back to Wednesday: Now, with pesky legality out of the way, the commission is free to push Meta against the wall and choke its neck until it offers its services practically for free, since Meta’s primary revenue source is effective advertisement targeting. Without targeted advertisements, Meta’s average revenue per user drops significantly since advertisers want to ensure their products get placed in front of prospective buyers’ eyes — but thanks to the European Union, if Meta wants to continue operating in the bloc, it needs to take a loss on offering services to Europeans. And if Meta does the algebra and determines it’s not worth it to stay in the European Union, the commission will surely take Meta to court for leaving for it being somehow anti-consumer, even though it’s the commission’s fault in the first place for forcing the company to leave.

The only way for Meta to remain profitable in the European Union is for the company to inundate users with terrible advertisements in the hopes that it can make up for the lack of targeted advertising by simply selling more advertisements at a lower price. That move, however, would also probably draw the ire of regulators in Brussels, who clearly have nothing better to do than go after technology companies which not one ordinary European is complaining about. Europeans can already opt out of targeted advertising for free by using App Tracking Transparency, Apple’s technology for disguising a tracking identifier companies like Meta use to keep track of users across the web. But privatization does not seem to be in the commission’s interests, so it has instead opted to bully any company doing business in its sacred bloc for no reason other than politics. Is there seriously no way for Europeans to object to this madness?

Julian Barnes, reporting for The New York Times:

The threat against U.S. elections by Russia and other foreign powers is far greater today than it was in 2020, the chairman of the Senate Intelligence Committee said on Tuesday.

Senator Mark Warner, the Virginia Democrat who leads the committee, said the danger had grown for multiple reasons: Adversarial countries have become more adept at spreading disinformation, Americans are more vulnerable to propaganda, communication between the government and social media companies has become more difficult and artificial intelligence is giving foreign powers new abilities…

“With polarization in this country, and the lack of faith in institutions, people will believe anything or not believe things that come from what used to be viewed as trusted sources of information,” Mr. Warner said. “So there’s a much greater willingness to accept conspiracy theories.”

Vulnerability to influence operations, Mr. Warner said, is not confined to the United States. In Slovakia, for example, Russian information operations influenced views of Russia’s war in Ukraine.

“Slovakia was 80 percent pro-Ukraine,” he said. “Two years later, with massive amounts of Russian misinformation and disinformation, you have a pro-Russian government and 55 percent of Slovaks think America started the war in Ukraine.”

These statements from Warner aren’t shocking even in the slightest. If you look at social media now, it’s filled with disinformation from self-proclaimed Americans, both on the left and right of the political spectrum. Left-wing nuts continue to push pro-China, pro-Iran propaganda in the name of “progressivism,” while right-wingers brazenly post pro-Kremlin, anti-Ukraine rhetoric. As I’ve written before, I don’t think these are actually Americans, but rather, I surmise the ideas from Russian and Chinese bots have entirely taken over the minds of Americans, including those in Congress. Young Americans are more likely than ever to distrust institutions and their government — and rightfully so — but they are also more likely to subscribe to foreign propaganda that advances flawed ideologies.

Russia and China continue to flood social media websites with fictitious misinformation to influence the 2024 election. The two countries employ bots, and sometimes even human labor, to publish websites full of wrong information, social media posts expressing dissatisfaction with the current administration, and launch advertising campaigns to create distrust in the U.S. government. These tactics prove themselves every day to be overwhelmingly successful; a message disguised as an “America First” endorsement is much more likely to be listened to than one directly opposing American military efforts overseas. A Russian operative asking “Why are we sending our money to Ukraine when we should be securing our southern border?” has a more striking effect on right-leaning Americans than “We should stop sending money to Ukraine because it’s none of our business.”

These operations are not covert or minuscule in scale — they’re entirely widespread on the internet today, on social media websites like X and Threads. People with “radical” political views might not actually be expressing any beliefs at all — instead, they are probably just a Russian asset. My advice is not to interact with these foreign influence accounts whatsoever, and I further demand that social networks like Meta take more action to combat misinformation and perform a mass deletion of spam accounts with outrageous beliefs. This is happening in the United States, in Europe, and in many Asian countries, and internet citizens must be more vigilant in reporting it and combating the spread of dangerous propaganda that has the power to threaten our respective democracies.

Andrew Hawkins, reporting for The Verge:

Rivian is pushing a new software update that will give its customers better insight into which EV chargers to visit — and which to avoid…

“Our North Star is charging and trip planning in EVs should just work,” Wassym Bensaid, Rivian’s head of software, told The Verge. “You should not think about it.”

I had the chance to test out Rivian’s new software update during a recent road trip in an R1S SUV. Inputing a destination brought up dozens of chargers on the vehicle’s navigation, each of which displayed a letter grade. An “A” grade is a sign that the charger was in good working condition, while an “F,” well, speaks for itself.

“Surprisingly, actually, there’s multiple chargers rated F,” Bensaid said. “That was one of the ‘a ha’ moments as we went through the data.”

The new ranking system is determined by a host of data collected by Rivian’s customers, Bensaid said. Each vehicle is connected and constantly sending data back to the company’s headquarters, which then gets processed to remove “noise” that’s not essential to the decision-making algorithm.

This is extremely clever. While Tesla Superchargers are notoriously reliable and display stall information in Tesla vehicles’ infotainment systems, that functionality isn’t available for other electric vehicle manufacturers to integrate themselves, even though Rivian and other automakers have recently opened support for Tesla Superchargers, working with Tesla to develop adapters for their vehicles. (Rivian’s integration launched in March.) This new Rivian software feature, however, not only adds reliability information to the map in the vehicle itself from Tesla Superchargers, but also collects information from other brands, like Electrify America, the United States’ largest DC fast charging network outside of Tesla’s Superchargers.

Electrify America stalls are often plagued with reliability issues, and Electrify America itself doesn’t have the ability to monitor how its chargers are operating. The only way for customers to check if an Electrify America unit is functioning is to drive to one and hope for the best. Rivian’s new software system will automatically collect analytics from the chargers whenever a Rivian driver charges at one; after enough users have charged at a destination, results will appear for all other drivers through Rivian’s in-car map. While this information isn’t real-time, per se, unlike Tesla’s feature which automatically notifies drivers if a stall is broken, it is better than going to a charger without knowing anything about its reliability.

Hawkins points out that this software is limited due to how few Rivian cars traverse the roads of the United States, which is somewhat detrimental to the usefulness of the feature. Underused chargers located in rural areas are significantly less likely to ever be touched by a Rivian driver, let alone however many it requires for the software to begin calculating reliability information for it, so only busy chargers in metropolises will benefit from the software update. I still think that is better than nothing, but it also adds pressure on Electrify America and ChargePoint, another EV charger brand, to add public uptime data via an application programming interface or integrations with carmakers for them to integrate the statistics into their vehicles.

Google, writing in an email to me Monday morning:

Hi Eshu,

Thank you for being a Google One member where you enjoy extra storage, family sharing, and more. We’re writing to let you know about some updates coming to your Google One subscription starting on May 15. These changes are designed to streamline your benefits while ensuring you have a valuable subscription experience…

Phasing out two benefits: With a focus on providing the most in-demand features and benefits, we’re discontinuing free shipping for select print orders from Google Photos (in Canada, the UK, US, and EU) starting on May 15 and VPN by Google One later this year.

My subscription price is still the same, so I guess it’s time to add another product to the Google Graveyard. Speaking of Google One and the Google Graveyard, here is Abner Li, reporting for 9to5Google:

Google is now “discontinuing the VPN feature as [they] found people simply weren’t using it.” The company tells 9to5Google that the deprecation will let the team “refocus” and “support more in-demand features with Google One.”

Earlier this year, Google One hit 100 million subscribers and CEO Sundar Pichai teased it as a future growth area driven by AI. Today’s change follows this week’s news about AI editing tools in Google Photos going free in the coming months and no longer requiring a subscription save for unlimited Magic Editor usage.

Google One included a virtual private network service, which the company launched in 2020 for some odd reason, but now that’s gone too. And again, users’ subscription prices are staying the same. Anyone who subscribed to Google One, especially to a yearly plan, is now just getting prematurely cut off because Google doesn’t care about supporting products.

Again, add it to the Google Graveyard.

Richard Stengel, writing for The Atlantic:

How many times has it happened? You’re on your computer, searching for a particular article, a hard-to-find fact, or a story you vaguely remember, and just when you seem to have discovered the exact right thing, a paywall descends. “$1 for Six Months.” “Save 40% on Year 1.” “Here’s Your Premium Digital Offer.” “Already a subscriber?” Hmm, no.

Now you’re faced with that old dilemma: to pay or not to pay. (Yes, you may face this very dilemma reading this story in The Atlantic.) And it’s not even that simple. It’s a monthly or yearly subscription—“Cancel at any time.” Is this article or story or fact important enough for you to pay?

Or do you tell yourself—as the overwhelming number of people do—that you’ll just keep searching and see if you can find it somewhere else for free?

According to the Reuters Institute for the Study of Journalism, more than 75 percent of America’s leading newspapers, magazines, and journals are behind online paywalls. And how do American news consumers react to that? Almost 80 percent of Americans steer around those paywalls and seek out a free option.

Paywalls create a two-tiered system: credible, fact-based information for people who are willing to pay for it, and murkier, less-reliable information for everyone else. Simply put, paywalls get in the way of informing the public, which is the mission of journalism. And they get in the way of the public being informed, which is the foundation of democracy. It is a terrible time for the press to be failing at reaching people, during an election in which democracy is on the line. There’s a simple, temporary solution: Publications should suspend their paywalls for all 2024 election coverage and all information that is beneficial to voters. Democracy does not die in darkness—it dies behind paywalls.

I’d go one step further than Stengel did: I think paywalls entirely are mostly unnecessary for regular news websites. There should be an exception for small firms or publications that mainly focus on analytical coverage or opinion pieces, such as The New Yorker or The Atlantic, even though there is irony in an article about paywalls being hidden behind a paywall, as Stengel rightly points out. The reason I say this is because analysis isn’t information per se in the same way regular, hard news and reporting is. The people who are interested in analysis or a columnist’s views on a topic are much more likely to pay for that information — it’s just news that needs to be balanced and free.

Paywalls are a relic from a time when newspapers were bought from newsstands daily and read in coffee shops across the country. Every day, people would buy a newspaper for a quarter, read it, then leave it for the next person to read — and so on, every day. Newspaper publishers saw an opportunity to sell subscriptions to these papers to frequent readers, who perhaps could save some money by buying the newspaper at a discounted monthly or yearly rate as opposed to buying a new paper every day. When the online era took over and as people bought fewer physical newspapers, publishing houses moved to sell those subscriptions online.

As inflation hiked, however, newspaper subscriptions became more expensive. Now, The New York Times costs $25 a month because the physical copy costs $4 per paper ($4 a day for 30 days is $120 a month, so the digital version is still cheaper, obviously). In parallel, news became more affordable due to the advent of online advertising and free journalism, causing many to drop their subscriptions and just consume free news published by digital-only websites, like The Verge or NBC News. The result of this schism is that there is a divide between free and paid journalism, as well as the quality of information that emanates from each source.

This is the core of Stengel’s piece: The online media landscape gives credence to clickbait because it spreads so quickly due to it being free. Free articles metastasize through the internet so rapidly because everyone can read them. But this abundance of free, semi-factual journalism is spreading and warping the public’s perception of the news media because good journalism just isn’t being sold and marketed properly. These online media companies often make more money than the print publishers that sell subscriptions because of how quickly the information spreads on social media. Why aren’t print publishers picking up on this?

I firmly believe more websites should drop the paywall entirely and instead opt to sell effective, non-intrusive advertising, which can often be lucrative. Newspapers themselves are already fantastic examples of how lucrative advertising can be, as they sell full-page ads in innovative formats, which advertising companies are itching to cash in on. This is part of the game of journalism in 2024 — if we want the media to be a reliable arbiter of information without relegating the ever-important job of reporting to citizen journalists on social media websites like X or Threads, the corporate overlords who control the media should get better at making it profitable.

The news industry is at a crossroads, with the hastened development of generative artificial intelligence in newsrooms and the increased popularity of foreign-owned video websites like TikTok, where more Americans are getting their information from than ever before. At a critical time like this, journalism should become more accessible, innovative, and forward-facing — and removing paywalls is a key step in that direction.

I have written about the Humane Ai Pin twice before, once when the product was first announced in November, and the other in March, when the company released a video walkthrough of the device’s features. Now, reviews are in, both on YouTube and the web, and they’re scathing. The Ai Pin is — and I cannot stress this enough — utter garbage, and that isn’t me jesting. To support this claim, here are some quotes from reviewers who have had time with the pin:

“It’s a nightmare.” — Arun Maini, Mrwhosetheboss

“It’s just frustrating.” — Michael Fisher, MrMobile

“It’s futuristic if the future sucked.” It “solves nothing and makes me feel stupid.” — Cherlynn Low, Engadget

“It just doesn’t work.” — David Pierce, The Verge

That is just a small snippet of criticism this device has received in just one day of reviews. I agree with these reviewers — this product lacks conviction, lacks a path to success, and doesn’t even do what it is advertised to do. As one commenter on YouTube put it, it feels like a late April Fools’ Day joke. It costs as much as a mid-range smartphone at $700, requires a mandatory $24 monthly subscription to function as anything more than a paperweight¹, and relies on the whims of artificial intelligence to do literally anything. And the times when it does do something, it does it wrongly, misidentifying landmarks, such as in Pierce’s review, or making up information, as shown in Fisher’s video.

AI software can be refined and tweaked over time². What can’t, however, is the very design premise of this gadget. Its primary method of interaction is a loudspeaker that is bound to annoy everyone around, and connecting to a Bluetooth headset requires interacting with a flawed, frustrating laser projector. Reviewers have described the projector as annoying, hard to use, and simply impossible to view while in daylight. Interacting with content requires learning unintuitive gestures and flailing movements of the arms. Due to this oversight, which I picked up on in November, interacting with this menace of a lapel pin causes arm strain and annoyance.

Also, the battery dies quite frequently, making Humane’s “Perpetual Power System,” i.e., extra batteries, essential to use the product for anything longer than a few hours, according to early tests. And if you do use it for extended periods, it’ll overheat, as Fisher demonstrated during a call with his mother and as Pierce encountered while using the laser projector. These occurrences show that the product is impossible to use most of the time, and is useless when it is possible to use. The onboard camera, used for first-person shots, is low-quality and lackluster, though it is neat in a pinch, but no reasonable nor sane person would say that it is worth $700 and $24 a month.

Humane’s founders, Bethany Bongiorno and Imran Chaudhuri, have marketed the Ai Pin as a smartphone companion at the very least, and at times have even come out as more confident after the flopped November launch, calling the product a smartphone replacement at a time of increased distraction. To back up this moot point — people love their phones — Bongiorno has been reposting accounts citing a book with debatable sourcing that claims smartphones are the sole cause of childhood and adolescent depression, when in fact, the rises in these numbers are correlated with the uptake in smartphone use, not caused by it. When confronted with her earlier claims that she marketed the Ai Pin as a replacement for the phone, she flat-out denied it.

Nobody should take these claims seriously because this entire project feels like a scheme for investor money. When the original plan for a more ambitious product failed, Humane pivoted to AI large language models and built this device in four months after the release of ChatGPT in October 2022. Trust the reviewers because they’re experts — and the experts say that the Humane Ai Pin is a worthless piece of garbage. So do I.

Thomas Grove, Nicholas Bariyo, Micah Maidenberg, Emma Scott, and Ian Lovett, reporting for The Wall Street Journal:

A salesman at Moscow-based online retailer shopozz.ru has supplemented his usual business of peddling vacuum cleaners and dashboard phone mounts by selling dozens of Starlink internet terminals that wound up with Russians on the front lines in Ukraine.

Although Russia has banned the use of Starlink, the satellite-internet service developed by Elon Musk’s SpaceX, middlemen have proliferated in recent months to buy the user terminals and ship them to Russian forces. That has eroded a battlefield advantage once enjoyed by Ukrainian forces, which also rely on the cutting-edge devices.

The Moscow salesman, who in an interview identified himself only as Oleg, said that most of his orders came from “the new territories”—a reference to Russian-occupied parts of Ukraine—or were “for use by the military.” He said volunteers delivered the equipment to Russian soldiers in Ukraine.

On battlefields from Ukraine to Sudan, Starlink provides immediate and largely secure access to the internet. Besides solving the age-old problem of effective communications between troops and their commanders, Starlink provides a way to control drones and other advanced technologies that have become a critical part of modern warfare.

The proliferation of the easy-to-activate hardware has thrust SpaceX into the messy geopolitics of war. The company has the ability to limit Starlink access by “geofencing,” making the service unavailable in specific countries and locations, as well as through the power to deactivate individual devices.

Russia and China don’t allow the use of Starlink technology because it could undermine state control of information, and due to general suspicions of U.S. technology. Musk has said on X that to the best of his knowledge, no terminals had been sold directly or indirectly to Russia, and that the terminals wouldn’t work inside Russia.

The Wall Street Journal tracked Starlink sales on numerous Russian online retail platforms, including some that link to U.S. sellers on eBay. It also interviewed Russian and Sudanese middlemen and resellers, and followed Russian volunteer groups that deliver SpaceX hardware to the front line.

Anyone who seriously believes Musk doesn’t have the ability to properly restrict these terminals to specific airspace via geofencing is genuinely stupid, and probably a Russian asset. Here is how Russians get access to Starlink, even though President Vladimir Putin’s propaganda would lead you to believe Starlink terminals aren’t authorized for use in Russia: First, smugglers buy Starlink hardware in Middle Eastern countries, like the United Arab Emirates, for example, then activate those terminals for use anywhere in the world — a subscription SpaceX, the company that makes Starlink, sells. Then, those smugglers market the terminals on sites like eBay so that Russians can have the hardware shipped to nearby, Kremlin-friendly countries, sold at a markup. Then, smugglers bring the terminals and receivers over the border as if they were drugs or any other illegal items. “Patriotic” Russians then wheel them over to the front lines, where idiotic Russian soldiers are so stupid that they don’t even camouflage the bright white plastic terminals.

Ukraine, which is currently fighting a brutal war with Russia, also has access to Starlink, provided due to SpaceX’s contractual obligations with the U.S. Defense Department, which requires SpaceX to deploy Starlink hardware to U.S. allies in need, like Taiwan and Ukraine. Granted, Musk neutered Ukraine’s access to these important terminals, which provide internet access in Russian-occupied areas, when needed the most, but the service still remains available there in some capacity due to U.S. contracts. Bloomberg reported Wednesday that the Defense Department pays $23 million for this deal, but the U.S. official who leaked the information to Bloomberg declined to say whether the United States would renew it with Musk or not. This type of smuggling would be concerning to the United States due to these deals, so, in March, House Democrats sent a letter to SpaceX over the illegal import of Starlink terminals by Russia, which aid Putin in controlling drones and commanding troops. Apparently, nothing came of that letter, and just a month later, The Journal reported on the continued illegal use of Starlink terminals in Russia.

Again, only a fool would believe Musk and his company have no clue about the illegal use of these products in Russia — smart people work for SpaceX and upwards of 5,000 satellites roam Earth tracking the position of the terminals in real-time, according to The Journal’s report. Musk has the tools at his disposal to halt Russia’s unlawful use of terminals to destroy civilian buildings in Ukraine and illegally occupy sovereign territory, but he never will, because he himself is a Russian asset parroting propaganda straight from Moscow on X, his social media website. However many pro-Kremlin Republican puppets there are in Congress, however, the United States should exercise its leverage and contracts with SpaceX to force Musk’s firm to comply with U.S. law and disable enemy use of Starlink satellites. The United States has given Musk a free pass on contracts for too long — if it wants to continue doing business with the world’s richest Russian propagandist, it needs to shove him down on his knees and make him beg for the money he wants.

Katie Notopoulos, reporting for Business Insider regarding Google possibly scraping Google Docs for use as training data for Gemini, Google’s artificial intelligence chatbot:

A representative for Google confirmed to Business Insider that simply changing the share settings to “anyone with the link” did not mean that a document was “public” and would be used for AI training.

To be “publicly available,” that document would need to be posted on a website or shared on social media. Basically, some kind of web crawler would need to be able to find it. That can’t happen with a file just emailed back and forth between two people — like if you send your friend a link over Gmail, for instance, Google said.

Breathe a sigh of relief.

I still think this is mildly disingenuous since there is no way to opt out of AI training, unlike on a traditional website, where disallowing AI crawlers to index the page is as simple as adding a line to the website’s robots.txt file. And, not to mention, there is no clarification in either Google Docs’ privacy policy or terms of use on how Google might use your documents to train its AI models. I hope Google makes improvements and clarifies its policy in the near future. But for now, unless you’re publicly publishing a Google Doc to the web, there is no need to worry.

Cristiano Lima-Strong, reporting for The Washington Post:

Key federal lawmakers Sunday unveiled a sweeping proposal that would for the first time give consumers broad rights to control how tech companies like Google, Meta, and TikTok use their personal data, a major breakthrough in the decades-long fight to adopt national online privacy protections. The bipartisan agreement, struck by Senate Commerce Committee Chair Maria Cantwell (D-Wash.) and House Energy and Commerce Committee Chair Cathy McMorris Rodgers (R-Wash.), marks a milestone in the congressional debate over data privacy. The issue has befuddled lawmakers despite near-universal agreement — in Silicon Valley and in Washington — on the need for federal standards to determine how much information companies can collect from consumers online.

The measure, a copy of which was reviewed by The Washington Post, would set a national baseline for how a broad swath of companies can collect, use, and transfer data on the internet. Dubbed the American Privacy Rights Act, it also would give users the right to opt out of certain data practices, including targeted advertising. And it would require companies to gather only as much information as they need to offer specific products to consumers, while giving people the ability to access and delete their data and transport it between digital services.

Significantly, the deal — one of Washington’s most significant efforts to catch up to privacy protections adopted in Europe nearly a decade ago — would resolve two issues that have bogged down negotiations for years: whether a federal law should override related state laws and whether consumers should be permitted to sue companies that violate the rules.

Europe’s General Data Protection Regulation, commonly known as GDPR, aimed to achieve exactly what Cantrell and McMorris Rodgers’ bill now aims to do in the United States, but GDPR just makes users’ experience on the internet worse. The U.S. bill, which has not been fully written, would give users the right to request companies delete their user data — a crucial measure for data privacy in 2024. If you’re European, that might seem like common sense, but in the United States, consumers can only ask companies to delete their data, not force them to. This bill would change that.

Otherwise, the legislation is light on details, though I assume that will change as it gets written. It would allow users to opt out of targeted advertising, which is a potential cause for concern, although I imagine there will be a carveout for paid, advertisement-free subscriptions like the one Meta sells in Europe to comply with the Digital Markets Act and that is currently being challenged by the European Commission for some nonsense reason. And the mandate about restricting companies to only necessary data collection is essential to keep data brokers in check — brokers that collect massive amounts of data and sell it however they would like without any oversight or consumer consent.

Data transportation regulation is also important, though it concerns me how the measure will be written in this regard. Consumers should have the right to request copies of their data in easily accessible formats — not proprietary ones like the businesses that allow consumers to receive their data at all usually supply — and import that data in the finance and technology sectors. This regulation, however, should only apply to large corporations, as it may hinder innovation amongst smaller ones that need the data advantage the larger companies currently possess. (The new act does change rules depending on how large a company is, calculated via annual income.)

Color me skeptical, since data privacy regulation and monopoly checks are challenging pieces of legislation for a dysfunctional Congress to pass, but I think this new bill looks promising. Granted, the longer a bill is in committee, the worse it will become since most senators really have no grasp on technology. That combined with the looming election in November when Democrats might lose the Senate, and I am unsure if this bill will ever get to the House and be signed by the president. But it goes without saying that there is a crucial need for good, knowledgeable data privacy regulation at the federal level, bypassing a patchwork of poorly written state legislation that makes companies’ lives more difficult and confuses unwitting consumers.

Relatedly, Maryland passed similar data privacy regulations for its consumers Sunday, as well.

Cade Metz, Cecilia Kang, Sheera Frenkel, Stuart Thompson, and Nico Grant, reporting for The New York Times:

In late 2021, OpenAI faced a supply problem.

The artificial intelligence lab had exhausted every reservoir of reputable English-language text on the internet as it developed its latest A.I. system. It needed more data to train the next version of its technology — lots more.

So OpenAI researchers created a speech recognition tool called Whisper. It could transcribe the audio from YouTube videos, yielding new conversational text that would make an A.I. system smarter.

Some OpenAI employees discussed how such a move might go against YouTube’s rules, three people with knowledge of the conversations said. YouTube, which is owned by Google, prohibits use of its videos for applications that are “independent” of the video platform.

Ultimately, an OpenAI team transcribed more than one million hours of YouTube videos, the people said. The team included Greg Brockman, OpenAI’s president, who personally helped collect the videos, two of the people said. The texts were then fed into a system called GPT-4, which was widely considered one of the world’s most powerful A.I. models and was the basis of the latest version of the ChatGPT chatbot… Like OpenAI, Google transcribed YouTube videos to harvest text for its A.I. models, five people with knowledge of the company’s practices said. That potentially violated the copyrights to the videos, which belong to their creators.

Last year, Google also broadened its terms of service. One motivation for the change, according to members of the company’s privacy team and an internal message viewed by The Times, was to allow Google to be able to tap publicly available Google Docs, restaurant reviews on Google Maps, and other online material for more of its A.I. products.

As I’ve said many times previously, I do not think scraping content — even non-consensually — from the web to train AI models is illegal, since I think AI large language models are transformative. Granted, if an LLM reproduces text one-for-one, that is a concern because it is not fair use according to U.S. copyright law. But transformative use of copyrighted works is permitted under the law for a good reason. The best way to solve the kerfuffle between publishers, authors, and other creators and the AI companies hungry for their data is via comprehensive regulation written by experts — but knowing Congress, that will never happen. The current law is the law we will always have, and while it is not sufficient to address this new era of copyrighted works, it’s what we’re stuck with.

With that said, I am not necessarily upset at OpenAI for scraping public YouTube videos using Whisper to train GPT-4 — GPT-4 is not quoting YouTube videos verbatim and provides helpful information, which is more than enough to qualify as “fair use.” What I do have a problem with is Google’s implementation — in its reaction to OpenAI’s scraping, its own scraping, and its use of private Google Docs data. Google is the owner of YouTube, and YouTube users sign a contract with Google in order to use the service: the terms of service. So, due to this relationship between YouTube users and Google, Google has a responsibility to inform its users about how it’s using their data in the form of a privacy policy.

Unlike the terms of use, YouTube’s privacy policy says nothing about how Google can use YouTube videos to train Bard and its other LLMs. (I’ll get to Google Docs momentarily.) This creates two issues: (a) it makes it fair game for any person or company to scrape YouTube content since Google did not provide YouTube users with an explicit guarantee that it would be the only company that scrapes their videos (or that their videos wouldn’t be scraped at all), and (b) it, unbeknownst to YouTube’s users, compromises their data without their knowledge. Neither of these issues put Google in a legally compromising scenario due to fair use, but it is not a good look for Google.

Looks matter for a company like Google, which must create the illusion of privacy for users to feel confident giving the company so much data. Unlike Meta, people use Google services for a variety of private matters, sharing personal family videos on YouTube and writing sensitive notes in Google Docs. On Meta services, every user has the expectation — aside from on services like WhatsApp and Messenger — that their data will be shared with the world, available for anyone to view or use however they would like. Google promises privacy and security, and, for the most part, has delivered on that promise — but it can’t continue selling users on privacy when its actions directly contradict that pitch.

And about OpenAI: YouTube likes to say that OpenAI’s scraping is against its terms of use, which any person who uses YouTube — including OpenAI’s employees who scraped the data — has implicitly agreed to, but YouTube doesn’t have the right to enforce that specific rule in the terms of use because the same usage terms also give YouTube creators the ownership rights to the content they publish on YouTube. It cannot be against the terms of service for creators to do what they want with their content; what if a creator wants OpenAI to have access to their videos? YouTube cannot meaningfully enforce this rule, and even if it wanted to, the argument would be unstable because YouTube (Google) does the same thing even though it has not given itself rights via the terms of service it claims OpenAI has broken.

And then, there is Google Docs. Unlike the issue of YouTube, this one is legally concerning. Google claims that it only trains on data from users who opt into “experimental features,” which is to say, the features that allow users to use Bard to help them write documents. That part of the agreement is well advertised, but the part where Google grants itself the ability to access private user data to train AI models is implicitly stated. Google does not tell users to sign a new service agreement to use Bard in Google Docs — it just includes in the main terms of service that if a user were to sign up for experimental features, their data may be used for training purposes. That is sleazy.

It might not be illegal, but, as I said earlier, immorality is harmful enough. This creates one more unnecessary problem for Google in the form of a question: How is Google gaining access to private Google Docs data? Most users are under the assumption that what they write in Google Docs/Sheets/Slides etc., is for their eyes only — that it is private data that is most likely encrypted at rest. If Google can mine and use it however it wants, it’s just being stored as plain text somewhere. LLMs do not need encrypted data because it is illegible, so Google is either decrypting Google Drive data for users who have opted in or is storing everyone’s files in some unencrypted format.

Whatever the case is, it is deeply concerning because it breaks trust. What happens to all of the people who no longer use Google Docs but have signed the terms of use that permit the usage of their old data that was written before the new agreement? Millions — hundreds of millions — of people are unwittingly sending their data straight to Bard’s language bank. The usage of the data itself may not be illegal, but the collection of it is immoral and might be a breach of the “contract” between Google and its users. I’m not particularly concerned about my data being used to train an LLM as long as that data is anonymized and obfuscated — and I think many people are the same — but it is wrong for Google to harvest this data and use it in ways users are unaware of.

Obviously, the best way to solve this problem is for Google to stop collecting Google Docs data — and perhaps YouTube data, though that is less pressing because it’s public, unlike private documents — or amend its privacy policy to include third parties like OpenAI in the mix, but all of that ignores a larger question: Where will training data for LLMs come from? Reputable websites such as The Times have blocked ChatGPT’s crawlers from ingesting their articles to use as training data, and eventually, these robots will run out of the internet to train on. That poses a large problem for LLMs, which require training data to function entirely.

The solution proposed by some is to prompt LLMs to create data for themselves, but immediately, anyone who knows how transformer models work will know that it will lead to heavily biased, inaccurate data. LLMs are not perfect now, and if they are trained on imperfect data, they will just become more imperfect, repeating the cycle. The only plausible solution I find to this is to make LLMs more efficient. Currently, AI companies are relying on the findings of research from 2020 — that research said, plainly, that the more data a model is fed, the more accurate it will be. But transformer models have improved since then, to the point where they can even correct themselves using data from the web to prevent “hallucinations,” a phenomenon where a chatbot creates information that doesn’t exist or is wrong.

I predict that in the next few years, researchers will stumble upon a breakthrough: LLMs will be able to do their weighting and prediction without as much data, using the web to fact-check their findings. I’m not a scientist, but this industry is booming right now, and new ideas will come to the table soon. But for 2024, perhaps AI firms should look elsewhere than private user data to train their models.

Mark Gurman, reporting for Bloomberg:

Apple Inc. has teams investigating a push into personal robotics, a field with the potential to become one of the company’s ever-shifting “next big things,” according to people familiar with the situation.

Engineers at Apple have been exploring a mobile robot that can follow users around their homes, said the people, who asked not to be identified because the skunk-works project is private. The iPhone maker also has developed an advanced table-top home device that uses robotics to move a display around, they said.

Though the effort is still in the beginning stages — and it’s unclear if the products will ultimately be released — Apple is under growing pressure to find new sources of revenue. It scrapped an electric vehicle project in February, and a push into mixed-reality goggles is expected to take years to become a major moneymaker.

I heavily doubt this is anything moronic like Tesla’s humanoid robot, which was actually just a man in a jumpsuit, but rather a product akin to Amazon’s Astro robot or a robot vacuum. Later in the article, Gurman writes that the company was exploring a product that could wash dishes in a sink, but I think a product like that would be too audacious and wouldn’t fit in with Apple’s existing products.

Apple manufactures personal computers, whether they be strapped to the face, on a desk, strapped to the arm, or in a pocket. A robot vacuum is a personal computer, but a humanoid contraption isn’t because the graphical user interface isn’t as important there — dexterity and motion are. Just as Apple doesn’t manufacture its own production equipment that it (Foxconn) uses to make its iPhones, it also won’t make a fully functioning robot straight out of a science-fiction film.

I’d be interested to see what Apple makes in this product category, but I’m also skeptical since these products haven’t been blockbuster hits before. Astro, for example, doesn’t have a killer use case since there aren’t that many things that a floor robot with wheels can do. I think Apple should spend less time engineering these experimental technologies and should instead focus on making its existing products more advanced, moving manufacturing away from China, and improving pitfalls Apple Vision Pro has.

These products, according to Gurman’s reporting, are aimed at the home only, which makes sense coming from a company like Apple. Instead of whatever Gurman outlines in his reporting, however, I think the way to approach this problem is to partner with existing home robotics companies, like iRobot, which manufactures the Roomba robot vacuum, and develop software that can integrate with HomeKit, Apple’s smart home platform. Imagine CarPlay but for other smart home products; they can run their own software made by their original manufacturers, but Apple can also contribute if users would like it to.

Apple makes software for its major hardware products, like the iPhone and the Mac, but these robots wouldn’t be their own stand-alone products — they would most likely need an iPhone or other Apple product to connect to and be controlled by. So in a way, it makes sense: Apple should work with third-party makers to make their products work better for Apple users, which is exactly how HomeKit operates now. That way, if the effort fails — as HomeKit has in many ways — it’s not on Apple to craft a remedy in the way it would be if Apple announced a robot of its own.

When Apple makes something, expectations are high, but if it works with another company, it isn’t much of a deal if the partnership fails. It will make headlines, but it will not be a failure for the company. Partnering with another firm may give Apple less to tout, but it also puts less pressure on the it to innovate, something that is sorely needed after the relative bust of Apple Vision Pro, which received a grand total of two weeks of media coverage.

David Pierce, writing for The Verge:

Google Podcasts is dead. It has been dying for months, since Google announced last fall that it was killing its dedicated podcast app in order to focus all its podcasting efforts on YouTube Music. This is a bad idea and a big downgrade, and I’d be more mad if only I were more surprised.

The Podcasts app is just the latest product to go through a process I’ve come to call The Google Cycle. It always goes the same way: the company launches a new service with grandiose language about how this fits its mission of organizing and making accessible the world’s information, quickly updates it with a couple of neat features, immediately seems to forget it exists, eventually launches a competitor out of some other part of the company, obviously begins to deprecate it and shift focus to the new competitor, and then, years later, finally shuts it down for real.

This is easily the best description of Google’s ridiculous sunsetting of products: advertise it to infinity as being the next best thing, completely neglect to update it, then axe it years later or build some of its functionality into an existing, more popular product — that’s the Google way. Google Podcasts’ funeral on Tuesday reminded me of my favorite killed Google product: Inbox by Gmail. When it first came out, it was a new, futuristic, mobile-first client for Gmail, not meant to replace it, but to supplement it — and it was fantastic. In the early months, it required an invite from another Gmail user, just like Gmail first did in 2004, building anticipation and excitement for a novel email client that was designed for smartphones first.

Google marketed Inbox to the fullest extent possible, and it had every reason to do so. Inbox pulled information from your Gmail inbox, then surfaced that information as if you searched for it on Google – in other words, organizing the world’s information and making it universally accessible and useful, Google’s mission statement. But, slowly, either due to the lack of adoption or just budget cuts from within the company, Google began haphazardly adding Inbox’s features to the normal Gmail applications, showing small thumbnails for tickets or events in the email list. This by all intents and purposes wasn’t Inbox, but it looked like Inbox, even though it didn’t even come close to replacing it. People began using those features, naturally.

Then, Google said Inbox had run its course and it was time for it to go, redirecting people back to Gmail. Google just wasted five years of everyone’s lives by making them switch to a new email client when it could have simply added the new features to Gmail from the very beginning. This is not just one case — time and time again, Google has continuously wasted customers’ time by announcing and marketing products that will eventually meet the fate of death in the end. And these products are often unceremoniously discontinued, just like in the case of Google Domains, which for years was a competitor to Squarespace that offered lower prices, but in the end, was sold to the very company it aimed to eclipse. This behavior weakens consumer trust.

After just a few of these incidents — Google abandoning or neglecting a product enjoyed by so many — customers no longer trust Google. Case in point, when users on the social media website X circulated a false message “from Google” saying that Google was sunsetting Gmail, and users actually believed it so much that Google had to issue a clarification that Gmail was indeed not going away. Google suffers from such terrible mismanagement that users are no longer incentivized to rely on Google services, and that is a terrible thing for the market and for Google, whose services are relied on by hundreds of millions of people a day.

I don’t think Google will ever discontinue its core products, like Google Search, Gmail, Google Drive, and Android, purely because regulators would probably intervene in those decisions because of how much of an astronomical impact they would have on the economy and society. But Google will continue to disappoint unsuspecting users and funnel them into more lucrative services to extract advertising revenue from them. That is why Google discontinued Podcasts, Inbox, and so many more of its services — it didn’t want to spend money on development and wanted more advertising revenue. Think about it this way: If users are scattered across three different services, selling advertisements on each one becomes (a) more difficult due to the lack of users on each service and (b) less lucrative because those placements will be viewed by fewer people each. Concentration is in Google’s interests.

For now, nobody should place their trust in a new Google product due to Google’s lack of corporate management. Sundar Pichai, Google’s chief executive, has continuously proven himself to be a servile, incompetent leader, and the only reason Google’s market capitalization has grown over the past nine years is due to the market’s expansion — more people are buying smartphones and computers, and more people now need Google services. Pichai has not done a single good thing for Google post-Chrome, a project he helped lead. Silicon Valley start-ups like OpenAI and Anthropic are running laps around Gemini in the artificial intelligence chatbot department — Google’s entry was rushed out of the door to compete with Microsoft, for heaven’s sake — Google Search is overridden with robot-written spam, and Google has no answer to Apple and Meta’s virtual reality products.

Google, by every measurement, is losing its dominance in the technology sector because it has a lack of corporate conviction. The job of a chief executive at such a large corporation as Google isn’t to write code in the break rooms — it’s to inspire the company to do good things, and Pichai has spectacularly failed in this regard. Pichai’s Google is slow to innovate, makes mediocre products, and needs a morale boost. Whoever is able to do that should probably take the top post within the company as soon as possible.