Google has fired a Cloud engineer who interrupted Barak Regev, the managing director of its business in Israel, during a speech at an Israeli tech event in New York, according to CNBC. "I'm a Google software engineer and I refuse to build technology that powers genocide or surveillance!" the engineer was seen and heard shouting in a video captured by freelance journalist Caroline Haskins that went viral online. While being dragged away by security — and amidst jeers from the audience — he continued talking and referenced Project Nimbus. That's the $1.2 billion contract Google and Amazon had won to supply AI and other advanced technologies to the Israeli military. 

Last year, a group of Google employees published an open letter urging the company to cancel Project Nimbus, in addition to calling out the "hate, abuse and retaliation" Arab, Muslim and Palestinian workers are getting within the company. "Project Nimbus puts Palestinian community members in danger! I refuse to build technology that is gonna be used for cloud apartheid," the engineer said. After he was removed from the venue, Regev told the audience that "[p]art of the privilege of working in a company, which represents democratic values is giving the stage for different opinions." He ended his speech after a second protester interrupted and accused Google of being complicit in genocide.

The incident took place during the MindTheTech conference in New York. Its theme for the year was apparently "Stand With Israeli Tech," because investments in Israel slowed down after the October 7 Hamas attacks. Haskins wrote a detailed account of what she witnessed at the event, but she wasn't able to stay until it wrapped up, because she was also thrown out by security. 

The Google engineer who interrupted the event told Haskins that he wanted "other Google Cloud engineers to know that this is what engineering looks like — is standing in solidarity with the communities affected by your work." He spoke to the journalist anonymously to avoid professional repercussions, but Google clearly found out who he was. A Google spokesperson told CNBC that he was fired for "interfering with an official company-sponsored event." They also told the news organization that his "behavior is not okay, regardless of the issue" and that the "employee was terminated for violating [Google's] policies."

Russian hackers keep trying to infiltrate Microsoft, the company revealed in a blog post. These hacks follow a similar incident from November of last year, in which state-sponsored agents obtained the emails of Microsoft’s senior level managers. An internal investigation led by Microsoft identified the hackers in both instances as a Russian group called Midnight Blizzard.

It looks like Midnight Blizzard has gotten bolder in its approach. Last year’s attack seemed to prioritize the collection of email addresses, but this most recent attack finds the group repeatedly attempting to breach the company’s systems and gain access to source code. Microsoft has filed an incident report with the U.S. Securities and Exchange Commission.

We don’t know exactly what these hackers want, but Microsoft said they are likely using email addresses acquired during November’s attack to help gain access to internal systems. Midnight Blizzard “may be using the information it has obtained to accumulate a picture of areas towhich led to a breach of government networks. attack and enhance its ability to do so,” the company wrote. I know one thing. They had better leave Clippy alone.

Midnight Blizzard is believed to work directly for Russia’s Foreign Intelligence Service (SVR) and is said to operate at the behest of Vladimir Putin. The group is likely behind 2016’s hack of the Democratic National Committee and 2020’s hack of the software company SolarWinds, which led to a breach of government networks. 

A group of 41 state attorneys general are demanding that Meta step up its support services for users who have been victims of hacks and account takeovers. “We refuse to operate as the customer service representatives of your company,” the group writes in a letter addressed to Meta’s chief legal officer. “We request Meta take immediate action and substantially increase its investment in account takeover mitigation tactics, as well as responding to users whose accounts were taken over.”

The letter, which was first reported by Wired, pushes Meta to deal with an issue that has long been a source of frustration to Facebook and Instagram users: the difficulty in accessing support after you lose access to your account. Though the company has taken steps over the years to make it easier for people to recover lost accounts, Meta largely relies on automated systems and there are still many people who fall through cracks.

As the letter highlights, this can be especially devastating for people who lose access to business accounts and pages. Even high-profile creators can find themselves unable to get support from a human employee of the company unless they have a personal connection to someone who works at Meta. Other users sometimes resort to legal services or hiring their own hackers to get their accounts back.

While it may seem surprising that state law enforcement officials would get involved in this issue, Reddit and other online forums for hacking victims often advise people to lodge complaints with their local AG’s office as a last resort. Some users have reported regaining access to their accounts after a state attorney general’s office intervened on their behalf.

Now, AGs are apparently flooded with such requests. “Our offices have experienced a dramatic and persistent spike in complaints in recent years concerning account takeovers that is not only alarming for our constituents but also a substantial drain on our office resources,” the letter states.

In addition to putting more resources into customer service, the letter asks Meta to provide more details on “the number of account takeovers over the past five years; suspected causes of the increase in account takeovers; safeguards currently in place to prevent account takeovers; current policies and procedures related to Meta’s response to account takeovers; and staffing related to safeguarding the platforms against account takeovers as well as responding to complaints.”

In a statement, a Meta spokesperson said that “scammers use every platform available to them and constantly adapt to evade enforcement” from the company. “We invest heavily in our trained enforcement and review teams and have specialized detection tools to identify compromised accounts and other fraudulent activity. We regularly share tips and tools people can use to protect themselves, provide a means to report potential violations, work with law enforcement and take legal action."

In 2022, Bloomberg reported that Meta was in the “early stages” creating a customer service division that would be able to help users with account issues. It's unclear what became of the plan. Later that year, the company laid off thousands of employees. In their letter, the state AGs notes that they saw an uptick in complaints “around the same time Meta announced a massive layoff of around 11,000 employees in November 2022, which reportedly focused on the ‘security and privacy and integrity sector.’”

The battle between AI chatbots is more than a two-horse race. Anthropic, the company formed by several ex-OpenAI employees, claims its new Claude 3 language model outperforms ChatGPT and Google's Gemini in several key industry benchmarks. It even hit "near-human" levels on some tasks, the company wrote in a blog. 

There are three new chatbots under the Claude 3 umbrella, including Haiku, Sonnet, and Opus. Sonnet powers the Claude.ai chatbot and is offered for free with an email sign-in. Meanwhile, Opus is the largest and most powerful LLM and will be available with a $20 per month subscription via the "Claude Pro" service. It's also multi-modal, so it can work with both text and image inputs, unlike past versions.

All Claude 3 models "can power live customer chats, auto-completions and data extraction tasks where responses must be immediate and in real-time," the company said. On top of promising "near-instant results," they can supposedly handle longer, multi-step instructions with increased accuracy.

Anthropic

Opus showed better graduate-level reasoning than GPT-4, scoring 14.7 percent higher in that test than GPT-4. It also beat OpenAI's chatbot in tasks involving math, coding, reasoning and knowledge. 

They also top past Claude models. "For the vast majority of workloads, Sonnet is 2x faster than Claude 2 and Claude 2.1 with higher levels of intelligence. It excels at tasks demanding rapid responses, like knowledge retrieval or sales automation. Opus delivers similar speeds to Claude 2 and 2.1, but with much higher levels of intelligence," according to Anthropic.

Meanwhile Haiku, the smallest version of Claude 3, is "the fastest and most cost-effective model on the market." To that end, it's capable of reading a dense research paper complete with charts and graphs in under three seconds. 

The company also noted that Claude 3 "can process a wide range of visual formats, including photos, charts, graphs and technical diagrams," aiding companies that use PDFs, flowcharts, or presentation slides. It'll also be less likely to refuse harmless content thanks to a more nuanced understanding of requests, while still recognizing "real harm."

Anthropic has said that Claude AI is guided by 10 secret foundational pillars of fairness. Claude 3 was trained on both nonpublic internal and public-facing data, using hardware from Amazon Web Services (AWS) and Google Cloud (Amazon recently invested $4 billion in Anthropic). 

Claude 3 Opus and Claude 3 Sonnet are available now through Anthropic's API, with Haiku set to follow soon. Sonnet is also accessible through Amazon Bedrock and in private preview on Google Cloud's Vertex AI Model Garden.

It’s 1995, and I’m trying to watch a video on the internet. I entered the longest, most complex URL I’d ever seen into AOL’s web browser to view a trailer for Paul W.S. Anderson’s long-awaited film adaptation of Mortal Kombat. I found it in an issue of Electronic Gaming Monthly, tucked away in the bottom of a full-page ad for the film. Online marketing at the time was such an afterthought, studios didn’t even bother grabbing short and memorable web addresses for their major releases, let alone dedicated websites. (Star Trek Generations and Stargate were among the few early exceptions.)

After the interminable process of transcribing the URL from print, I gathered my family around our Packard Bell PC (powered by an Intel 486 DX and, let’s say, 8MB of RAM), hit return and waited as the video slowly came down our 33.6kbps dial-up connection. And waited. It took 25 minutes for it to fully load. After corralling my family once again, I hit play and was treated to an horrendously compressed, low-resolution version of the trailer I’d been dreaming about for months. It was unwatchable. The audio was shit. But that was the moment I became obsessed with online video.

I imagined a futuristic world beyond my boxy CRT set and limited cable TV subscription. A time after VHS tapes when I could just type in a URL and enjoy a show or movie while eating one of those rehydrated Pizza Hut pies from Back to the Future 2. The internet would make it so.

Looking back now, almost 30 years later, and 20 years after Engadget sprung to life, I realize my 11-year-old self was spot on. The rise of online video transformed the internet from a place where we’d browse the web, update our LiveJournals, steal music and chat with friends on AIM to a place where we could also just sit back and relax. For Millennials, it quickly made our computer screens more important than our TVs. What I didn’t expect, though, was that streaming video would also completely upend Hollywood and the entire entertainment industry.

If my experience with the Mortal Kombat trailer didn’t make it clear enough, video was a disaster on the internet in the ’90s. Most web surfers (as we were known as the time) were stuck with terribly slow modems and similarly unimpressive desktop systems. But really, the problem goes back to dealing with video on computers.

Apple’s Quicktime format made Macs the ideal platform for multimedia creators, and, together with its Hypercard software for creating interactive multimedia databases, it spawned the rise of Myst and the obsession with mixed-media educational software. PCs relied on MPEG-1, which debuted in 1993 and was mainly for VCDs and some digital TV providers. The problem with both formats was space: Hard drives were notoriously small and expensive at the time, which made CDs the main option for accessing any sort of video on your computer. If your computer only had a 500MB hard drive, a slim disc that could store 650MB seemed like magic.

But that also meant video had no place in the early internet. RealPlayer was the first true stab at delivering streaming video and audio online — and while it was better than waiting 20 minutes for a huge file to download, it was still hard to actually stream media when you were constrained by a dial-up modem. I remember seeing buffering alerts more than I did any actual RealPlayer content. It took the proliferation of broadband internet access and one special app from Adobe to make web video truly viable.

While we may curse its name today, it’s worth remembering how vital Macromedia Flash was to the web in the early 2000s. (We’ve been around long enough to cover Adobe’s acquisition of Macromedia in 2005!) Its support for vector graphics, stylized text and simple games injected new life into the internet, and it allowed just about anyone to create that content. HTML just wasn’t enough. Ask any teen or 20-something who was online at the time, and they could probably still recite most of The End of the World by heart.

With 2002’s Flash MX 6, Macromedia added support for Sorenson’s Spark video codec, which opened the floodgates for online video. (It was eventually replaced in 2005 by the VP6 codec from On2, a company Google acquired in 2009.) Macromedia’s video offering looked decent, loaded quickly and was supported on every browser that had the Flash plugin, making it the ideal player choice for video websites.

The adult entertainment industry latched onto Flash video first, as you’d expect. Porn sites also relied on the technology to lock down purchased videos and entice viewers to other sites with interactive ads. But it was YouTube (and, to a lesser extent, Vimeo) that truly showed mainstream users what was possible with video on the internet. After launching in February 2005, YouTube grew so quickly it was serving 100 million videos a day by July 2006, making up 60 percent of all online videos at the time. It’s no wonder Google rushed to acquire the company for $1.65 billion later that year (arguably the search giant’s smartest purchase ever).

After YouTube’s shockingly fast rise, it wasn’t too surprising to see Netflix announce its own Watch Now streaming service in 2007, which also relied on Flash for video. At $17.99 a month for 18 hours of video, with a library of only 1,000 titles, Netflix’s streaming offering didn’t seem like much of a threat to Blockbuster, premium cable channels or cinemas at first. But the company wisely expanded Watch Now to all Netflix subscribers in 2008 and removed any viewing cap: The Netflix binge was born.

It’s 2007, and I’m trying to watch a video on the internet. In my post-college apartment, I hooked up my desktop computer to an early-era (720p) Philips HDTV, and all of a sudden, I had access to thousands of movies, instantly viewable over a semi-decent cable connection. I didn’t need to worry about seeding torrents or compiling Usenet files (things I’d only heard about from dirty pirates, you see). I didn’t have to stress about any Blockbuster late fees. The movies were just sitting on my TV, waiting for me to watch them. It was the dream for digital media fanatics: Legal content available at the touch of a button. What a concept!

Little did I know then that the Watch Now concept would basically take over the world. Netflix initially wanted to create hardware to make the service more easily accessible, but it ended up spinning off that idea, and Roku was born. The company’s streaming push also spurred on the creation of Hulu, announced in late 2007 as a joint offering between NBCUniversal and News Corp. to bring their television shows online. Disney later joined, giving Hulu the full power of all the major broadcast TV networks. Instead of a stale library of older films, Hulu allowed you to watch new shows on the internet the day after they aired. Again, what a concept!

Amazon, it turns out, was actually earlier to the streaming party than Netflix. It launched the Amazon Unbox service in 2006, which was notable for letting you watch videos as they were being downloaded onto your computer. It was rebadged to Amazon Video On Demand in 2008 (a better name, which actually described what it did), and then it became Amazon Instant Video in 2011, when it was tied together with premium Prime memberships.

As the world of streaming video exploded, Flash’s reputation kept getting worse. By the mid-2000s, it was widely recognized as a notoriously buggy program, one so insecure it could lead to malware infecting your PC. (I worked in IT at the time, and the vast majority of issues I encountered on Windows PCs stemmed entirely from Flash.) When the iPhone launched without support for Flash in 2007, it was clear the end was near. YouTube and other video sites moved over to HTML5 video players at that point, and it became the standard by 2015.

By the early 2010s, YouTube and Amazon weren’t happy just licensing content from Hollywood, they wanted some of the action themselves. So the original programming boom began, which kicked off with mostly forgettable shows (anyone remember Netflix’s Lillyhammer or Amazon’s Alpha House? Hemlock Grove? They existed, I swear!).

But then came House of Cards in 2013, Netflix’s original series created by playwright Beau Willimon, executive produced (and partially directed) by renowned filmmaker David Fincher and starring Oscar winner Kevin Spacey (before he was revealed to be a monster). It had all of the ingredients of a premium TV show, and, thanks to Fincher’s deft direction, it looked like something that would be right at home on HBO. Most importantly for Netflix, it got some serious awards love, earning nine Emmy nominations in 2013 and walking away with three statues.

By that point, we could watch streaming video in many more places than our computer’s web browser. You could pull up just about anything on your phone and stream it over 4G LTE, or use your smart TV’s built-in apps to catch up on SNL over Hulu. Your Xbox could also serve as the centerpiece of your home entertainment system. And if you wanted the best possible streaming experience, you could pick up an Apple TV or Roku box. You could start a show on your phone while sitting on the can, then seamlessly continue it when you made your way back to your TV. This was certainly some sort of milestone for humanity, though I’m torn on it actually being a net win for our species.

Instant streaming video. Original TV shows and movies. This was the basic formula that pushed far too many companies to offer their own streaming solutions over the past decade. In the blink of an eye, we got HBO Max, Disney+, Apple TV+, Peacock, and Paramount+. There’s AMC+, powered almost entirely by the promise of unlimited Walking Dead shows. A Starz streaming service. And there are countless other companies trying to be a Netflix for specific niches, like Shudder for horror, Criterion Channel for cinephiles and Britbox for the tea-soaked murder-mystery crowd.

And let’s not forget the wildest, most boneheaded streaming swing: Quibi. That was Dreamworks mastermind Jeffrey Katzenberg’s nearly $2 billion mobile video play. Somehow he and his compatriots thought people would pay $5 a month for the privilege of watching videos on their phones, even though YouTube was freely available.

Every entertainment company thinks it can be as successful as Disney, which has a vast and beloved catalog of content as well as full control of Lucasfilm and Marvel’s properties. But, realistically, there aren’t enough eyeballs and willing consumers for every streaming service to succeed. Some will die off entirely, while others will bring their content to Netflix and more popular services (like Paramount is doing with Star Trek Prodigy). There are already early rumors of Comcast (NBCUniversal’s parent company) and Paramount considering some sort of union between Peacock and Paramount+.

Online video was supposed to save us from the tyranny of expensive and chaotic cable bills, and despite the messiness of the arena today, that’s still mostly true. Sure, if you actually wanted to subscribe to most of the major streaming services, you’d still end up paying a hefty chunk of change. But hey, at least you can cancel at will, and you can still choose precisely what you’re paying for. Cable would never.

It’s 2024, and I’m trying to watch a video on the internet. I slip on the Apple Vision Pro, a device that looks like it could have been a prop for The Matrix. I launch Safari in a 150-inch window floating above my living room and watch the Mortal Kombat trailer on YouTube. That whole process takes 10 seconds. I never had the chance to see the trailer or the original film in the theater. But thanks to the internet (and Apple’s crazy expensive headset), I can replicate that experience.

Perhaps that’s why, no matter how convoluted and expensive streaming video services become, I’ll always think: At least it’s better than watching this thing over dial-up.

To celebrate Engadget's 20th anniversary, we're taking a look back at the products and services that have changed the industry since March 2, 2004.

Meta and LG have partnered up to “expedite” the former company’s extended reality (XR) business. What does that mean exactly? We don’t know, but Meta’s current VR/XR business is fairly robust, with the recent release of the Quest 3 headset.

LG says the ultimate goal of the partnership is “to combine the strengths of both companies across products, content, services and platforms to drive innovation in customer experiences within the burgeoning virtual space.”

Meta CEO Mark Zuckerberg traveled to LG’s headquarters in Seoul to announce the collaboration. During this visit, LG CEO William Cho tried out the Quest 3 and the recently-released Ray-Ban Meta smart glasses. The business leaders discussed “business strategies and considerations for next-gen XR device development.” LG’s CEO also seemed to take a particularly keen interest in Meta’s large language models and the potential to further integrate AI into standalone devices.

As stated above, we don’t know exactly what this partnership will entail. LG says it hopes to bring together “Meta’s platform with its own content/service capabilities” from its TV business. That sounds pretty boring, but LG also said the partnership will combine “Meta’s diverse core technological elements with LG’s cutting-edge product and quality capabilities.”

This leads to the lens-shaped elephant in the room. Meta XR and VR devices require displays and LG makes displays. It could be just that simple. After all, even Apple relied on Sony for the micro-OLED displays inside of the Vision Pro headset.

This news follows LG creating a dedicated XR business unit last year, which was founded to “accelerate the pursuit of new ventures in the virtual space arena.” This led to rumors that the company was planning to launch its own VR/XR headset, which could still happen.

Video doorbells manufactured by a Chinese company called Eken and sold under different brands for around $30 each come with serious security issues that put their users at risk, according to Consumer Reports. The publication found that these doorbell cameras are sold on popular marketplaces like Walmart, Sears and Amazon, which has even given some of their listings the Amazon Choice badge. They're listed under the brands Eken, Tuck, Fishbot, Rakeblue, Andoe, Gemee and Luckwolf, among others, and they're typically linked to a user's phone via the Aiwit app. Outside the US, the devices are sold on global marketplaces like Shein and Temu. We found them on Chinese website Alibaba and Southeast Asian e-commerce website Lazada, as well. 

Based on Consumer Reports' investigation, these devices aren't encrypted and can expose a user's home IP address and WiFi network name to the internet, making it easy for bad actors to gain entry. Worse, somebody with physical access to the doorbell could easily take control of it by creating an account on the Aiwit app and then pressing down on its button to put it into pairing mode, which then connects it with their phone. And, even if the original owner regains control, the hijacker can still get time-stamped images from the doorbell as long as they know its serial number. If they choose "to share that serial number with other individuals, or even post it online, all those people will be able to monitor the images, too," Consumer Reports explains. 

Based on the ratings these doorbells' listings got on Amazon, the platform has sold thousands to people who were probably expecting the devices to be able to provide some form of security for their homes. Instead, the devices pose a threat to their safety and privacy. The doorbells could even put people's well-being and lives at risk if, say, they have stalkers or are domestic violence victims with dangerous exes who want to follow their every move. 

People who own one of these video doorbells can protect themselves by disconnecting it from their WiFi and physically removing it from their homes. Consumer Reports said it notified the online marketplaces selling them about its findings in hopes that their listings would get pulled down. Temu told the publication that it's looking into the issue, but Amazon, Sears and Shein reportedly didn't even respond. 

If you work over public Wi-Fi, need to access geo-restricted content or just want to add an extra layer of privacy to your internet connection, you may want to use a VPN service. NordVPN is one of the most popular providers out there and right now, a digital code giving you a year of access to NordVPN Plus is going for $55 at Amazon. The plan also throws in one of our top password managers, NordPass. For comparison, right now a year of the Plus service is $72 directly from Nord. Of course, the best deals the company offers is on its two-year plans. Right now two years of the Plus service is $60 from Nord — so you're still saving $5 with Amazon's deal, plus you're not locked into a full two-year commitment. 

If you just want the VPN coverage without the password manager, you can get the standard service. It's $45 for a year of access, a savings of $15 over buying from Nord directly. And if you're just interested in the password manager, two-years of NordPass is down to $35, which is $5 less than going through Nord's site. 

We named Nord's password manager one of the best for cross-platform use in our guide to those services. The service keeps your credentials safe while making it easy to access your vault from whichever device or operating system you happen to be on. It also allows for biometric sign ins, making it even easier to get at your saved passwords. 

As for Nord's VPN, like all such services, it masks your IP address and encrypts your data to and from its destination. It also blocks your ISP from seeing data about your browsing. Just keep in mind that VPNs can't protect against other security risks like phishing and identity theft. Nord's VPN service didn't make the cut in our testing of such services, partly because we thought the price was a bit too high for the features provided. This deal removes some of the hesitation and we did find the service to be speedy and like that it's based on WireGuard, one of the more secure protocols. But the lack of open source software for most of its products and a less-than-stellar record of customer data privacy prevents it from being the best we can recommend. The service that did top our list was ProtonVPN. It's currently $72 for one year of the service alone. A bundle that includes a password manager, email and other services is currently $120 for the year. 

Follow @EngadgetDeals on Twitter and subscribe to the Engadget Deals newsletter for the latest tech deals and buying advice.

Tumblr and WordPress are reportedly set to strike deals to sell user data to artificial intelligence companies OpenAI and Midjourney. 404 Media reports that the platforms’ parent company, Automattic, is nearing completion of an agreement to provide data to help train the AI companies’ models.

It isn’t clear which data will be included, but the report suggests Automattic may have overreached initially. An alleged internal post from Tumblr product manager Cyle Gage suggests Automattic prepared to send private or partner-related data that wasn’t supposed to be included in the deal. The questionable content reportedly included private posts on public blog posts, deleted or suspended blogs, unanswered (therefore, not publicly posted) questions, private answers, posts marked explicit and content from premium partner blogs (like Apple’s former music site).

The internal post suggests Automattic’s engineers are preparing a list of post IDs that should have been excluded. It isn’t clear whether the data had already been sent to the AI companies. Engadget emailed Automattic to ask for comment on the report, and we’ll update this article if we hear back.

The company reportedly plans to launch a new opt-out tool on Wednesday that claims to allow users to block third parties — including AI companies — from training on their data. 404 Media reviewed an alleged internal FAQ Automattic prepared for the tool, which includes the answer, “If you opt out from the start, we will block crawlers from accessing your content by adding your site on a disallowed list. If you change your mind later, we also plan to update any partners about people who newly opt-out and ask that their content be removed from past sources and future training.” 

The phrasing, describing it as “asking” the AI companies to remove the data, may be relevant.

An alleged internal document from Automattic’s AI head, Andrew Spittle, replying to a staff question about data-removal assurances when using the tool, explains, “We will notify existing partners on a regular basis about anyone who’s opted out since the last time we provided a list. I want this to be an ongoing process where we regularly advocate for past content to be excluded based on current preferences. We will ask that content be deleted and removed from any future training runs. I believe partners will honor this based on our conversations with them to this point. I don’t think they gain much overall by retaining it.”

So, if a Tumblr or WordPress user requests to opt out of AI training, Automattic will allegedly “ask” and “advocate for” their removal. And the company’s AI boss “believes” the AI companies will find it in their best interest to comply “based on our conversations.” (How’s that for reassurance!)

AI data training deals have become a lucrative opportunity for websites treading water in today’s slippery online publishing landscape. (Tumblr’s staff was reportedly reduced to a skeleton crew in late 2023.) Last week, Google struck a deal with Reddit (ahead of the latter’s IPO) to train on the platform’s vast knowledge base of user-created content. Meanwhile, OpenAI rolled out a partnership program last year to collect datasets from third parties to help train its AI models.

The Federal Trade Commission (FTC) concluded Elon Musk ordered Twitter (now X) employees to take actions that would have violated an FTC consent decree regarding consumers’ data privacy and security. The investigation arose from the late 2022 episode informally known as “The Twitter Files,” where Musk ordered staff to let outside writers access internal documents from the company’s systems. However, the FTC says Twitter security veterans “took appropriate measures to protect consumers’ private information,” likely sparing Musk’s company from government repercussions by ignoring his directive.

FTC Chair Lina Khan discussed the conclusions in a public letter sent Tuesday to House Judiciary Committee Chair Jim Jordan (via The Washington Post). Jordan and his Republican colleagues have tried to turn the FTC’s investigation into a political wedge issue, framing the inquiry as a free speech violation — perhaps to shore up GOP support from Musk’s legion of rabid supporters. Jordan and his peers previously described the investigation as “attempts to harass, intimidate, and target an American business.”

Khan’s response to Jordan adopts a tone resembling that of a patient teacher explaining the nuance of a complicated situation to a child who insists on seeing simplistic absolutes. “FTC staff efforts to ensure Twitter was in compliance with the Order were appropriate and necessary, especially given Twitter’s history of privacy and security lapses and the fact that it had previously violated the 2011 FTC Order,” Khan wrote.

“When a firm has a history of repeat offenses, the FTC takes particular care to ensure compliance with its orders,” the FTC Chair wrote.

Tom Williams via Getty Images

The FTC’s investigation stemmed from allegations that Musk, newly minted as Twitter’s owner, ordered staff to give outside writers “full access to everything” in late 2022. Had staff obeyed Musk’s directive, the company likely would have violated a settlement with the FTC (originally from 2011 but updated in 2022) requiring the company to tightly restrict access to consumer data.

In November 2022, the FTC said publicly it was monitoring Twitter’s developments following Musk’s acquisition with “deep concern.” That followed the resignation of chief information security officer Lea Kissner and other members of the company’s data governance committee. They expressed concerns that Musk’s launch of a new account verification system didn’t give them adequate time to deploy security reviews required by the FTC.

Ultimately, Twitter security veterans ignored Musk’s “full access to everything” order. “Longtime information security employees at Twitter intervened and implemented safeguards to mitigate the risks,” Khan wrote in the letter. “The FTC’s investigation confirmed that staff was right to be concerned, given that Twitter’s new CEO had directed employees to take actions that would have violated the FTC’s Order.”

Slaven Vlasic via Getty Images

Rather than supplying outside writers with the “full access” Musk wanted them to have, Twitter employees accessed the systems and relayed select information to the group of outsiders. “Ultimately the third-party individuals did not receive direct access to Twitter’s systems, but instead worked with other company employees who accessed the systems on the individuals’ behalf,” Khan wrote.

The FTC says it will continue to monitor X’s adherence to the order. “When we heard credible public reports of potential violations of protections for Twitter users’ data, we moved swiftly to investigate,” FTC spokesman Douglas Farrar said in a statement to The Washington Post. “The order remains in place and the FTC continues to deploy the order’s tools to protect Twitter users’ data and ensure the company remains in compliance.”