Tag: big tech

Lawsuit: Elon Musk’s xAI Trained Grok AI Using Child Pornography

A woman identified as “Jane Doe” sued Elon Musk’s xAI this week, alleging the company trained its Grok AI chatbot on child pornography depicting her, in what appears to be the first case accusing xAI of training its AI on child sexual abuse material (CSAM).

Ars Technica reports that the proposed class-action lawsuit filed against Musk’s xAI, now part of SpaceX, centers on abuse Doe suffered as a preschooler in the early 2000s, when adult men raped her to produce images later sold to pedophiles online. Those images were hashed by the National Center for Missing and Exploited Children (NCMEC) and the Canadian Centre for Child Protection, groups that track known child pornography so it can be identified and removed wherever it resurfaces.

Doe gets alerts through the U.S. Department of Justice Victim Notification System whenever her abuse material turns up somewhere new. The Canadian Centre for Child Protection told her that AI-generated CSAM depicting her had shown up on xAI. According to the complaint, offenders on online forums discussed “creating AI generated CSAM of Plaintiff and other similarly situated known, legacy, victims of CSAM.”

The lawsuit claims xAI stores images Grok generates and reuses them to further train the model. A press release from Doe’s lawyers described the material as “that same material,” referring to the CSAM depicting her that investigators say fed into Grok’s outputs. The complaint itself alleges that “CSAM depicting Plaintiff with its longstanding well-known hash values has been used as a part of the dataset used by xAI.”

Breitbart News previously reported on AI training datasets that were found to contain child pornography:

The Stanford Internet Observatory, in collaboration with the Canadian Centre for Child Protection and other anti-abuse charities, conducted a study that found more than 3,200 images of suspected child sexual abuse in the AI database LAION. LAION, an index of online images and captions, has been instrumental in training leading AI image-makers such as Stable Diffusion.

This discovery has raised alarms across various sectors, including schools and law enforcement. The child pornography has enabled AI systems to produce explicit and realistic imagery of fake children and transform social media photos of real teens into deepfake nudes. Previously, it was believed that AI tools produced abusive imagery by combining adult pornography with benign photos of kids. However, the direct inclusion of explicit child images in training datasets presents a more direct and disturbing reality.

Much of Doe’s legal argument turns on how Grok’s terms of service handle user content. The complaint says Grok treats public posts on X, along with the outputs Grok itself generates, as training data by default. As the filing puts it, “Because Grok’s terms treat public X posts and Grok’s own outputs as training data by default, publicly posting an image does not just expose it to viewers, but also feeds [it] directly into the pipeline xAI uses to train and improve its model and thereby generate further images.” xAI filters violent content out of its training data, but its terms do not specifically exclude CSAM, non-consensual intimate imagery, or other sexual or inappropriate material.

Keep reading

The Guest List Economy: How Big Tech, Big Pharma, and Big Real Estate Lock You Out

Who, exactly, manages the U.S. economy? And what happens when the people who run it convert it into their own private club — one that has a guest list that somehow never includes you?

Big Tech giants like Apple decide which apps you’re allowed to see and use. Remember when it shut down Parler? 

One day Parler was up. The next day Parler was gone — just because Apple decided the wrong people were talking too much.  

That’s a very convenient arrangement, don’t you think? 

Big banks and payment processors often play the role of gatekeepers too — deciding which businesses get accounts, which transactions get processed, and, in some cases, who gets access to the financial system at all. Conservatives learned just how much power those institutions wield during the years-long fight over politically motivated “debanking.”

Then there are the pharmacy benefit managers — middlemen who stand between you and the medicine your doctor prescribed. They decide which drugs get covered by insurance, which pharmacies you get to use, and how much you pay.

Do we honestly believe these people have the patients’ best interests at heart?

Of course we don’t.  

Average people lose choices and smaller competitors must follow the rules as written by the powerful and well-connected.

Their latest racket is in housing. 

Giant real estate players are creating preferred networks, where the well-connected get the first look at available homes on the market while regular buyers are left to scour through a public market that no longer always presents everything that’s actually for sale.  

House Judiciary Antitrust Subcommittee Chair Scott Fitzgerald is asking the right questions. He has demanded answers from Compass — the country’s biggest real estate brokerage — and Midwest Real Estate Data (MRED) MLS system about the private listings partnership they have popularized together, which only MRED members can see. Because the MLS controls nearly all the listings that are visible in one particular region, this anti-competitive behavior has real consequences.  

But the American people have many allies in their corner. 

Just like Rep. Fitzgerald and the rest of his colleagues on the congressional antitrust subcommittee, the Justice Department and Federal Trade Commission have also never been afraid to enforce the competition laws on the books to protect

Populist conservatives should always do antitrust this way — enforcing the law, knocking down the barriers anti-free market actors put up, and giving regular people a fair shot at competing. But then getting the hell out of the way.

The goal should be to protect competition, not to remake the economy to fit a more egalitarian, socialistic model — like how the Biden administration prevented low-cost airlines Spirit Airlines and Jet Blue from merging, leading to Spirit’s bankruptcy earlier this year. That didn’t help consumers. That led to less choices and higher fare costs. 

What helps consumers is a government willing to take down the corporate actors who violate the consumer welfare standard and leave Americans with fewer choices, not more. Right now, the most urgent target for that kind of enforcement is the housing industry.

The hallmarks of a free marketplace are competition, choice, and the promise that an outsider can still knock the people on top off their perch by building a better mousetrap. 

America’s economy is supposed to be a marketplace open to everyone, not a managed economy that has its rules set by the members of a members-only club. The more Congress, Attorney General Todd Blanche, and FTC Chair Andrew Ferguson can do to keep it that way, the better off we’ll all be.

Keep reading

Google To Expand Play Age Signals API to All Users Worldwide

Just in time for the global digital ID push, Google will roll out its Play Age Signals API to all Google Play users worldwide by the end of 2026. This gives Android apps an easy way to ask Google how old its users are.

Google Play’s vice president of product management, Paul Feng, announced the expansion on July 29. Australia and Canada come first by mid-August, followed by “a full global rollout to all users later this year.” Apple launched its own version, the Declared Age Range API, worldwide in February.

An app calls the API while it’s running and Play answers with the band the user falls into, 0-12, 13-15, 16-17, or 18+ by default. Developers can redraw those bands in the Play Console. A developer who sets minimum ages of 13 and 17 gets users sorted into 0-12, 13-16, and 17 and over.

The feedback arrives as a lower and an upper bound, and the top band carries no upper bound, so an adult in that setup comes back as ageLower = 17, an age floor with no ceiling. A user who declined to share gets no band at all, only a NOT_SHARED status. Google’s rules bar any other use of the answer, “including, but not limited to, advertising, marketing, user profiling or analytics.”

To parents, Google says the API as “a privacy-preserving tool that puts parents in the driver’s seat.” A parent enters a child’s range once in the Family Link app, 16-17 rather than an exact birth date and every app that has built in the API can read it. Google says sharing is off until a parent opts in and that the setting can be changed or switched off at any time.

That default lasts until a law overrides it and Texas already has and you only have to look at the latest senate bill that we just covered to know exactly how this “privacy” preserving plan can end up being anything but. Adults can share their own range when an app asks. “Providing a safe online experience and protecting users from harm is a top priority at Google Play,” Feng wrote in his announcement.

Apps receive more than a band. Google’s developer pages say an app can receive “users’ age verification or supervision status, age ranges, and other applicable signals,” and the status field can also come back VERIFICATION_REQUIRED. A developer knows when a user declined to share and when an age was verified rather than declared, and Google leaves it to each app to decide what to accept. TechRadar, citing reports, says users who fail to complete verification can be blocked from downloading a wide range of apps, not just those with adult ratings.

The API went live in Brazil on March 17, the day the Digital ECA took effect. That law bans the “I am over 18” checkbox and threatens fines of up to 50 million reais, about US$9.44 million, or 10 percent of a company’s Brazilian revenue. Texas followed.

Play began returning ages and running an age verification flow for Texans who created accounts after May 28, once a federal appeals court stayed the December 2025 injunction that had blocked the state’s App Store Accountability Act.

Keep reading

Tech Evangelist Bill Gates Suddenly Argues AI Needs to Slow Down to Protect Humanity

Bill Gates, the Microsoft co-founder and leftist billionaire who has spent years predicting that AI would free people from drudgery and speed cures for disease and climate change, published an essay Wednesday warning that without urgent government intervention, AI could cause more harm than good.

Bill Gates, who transformed himself from a software tycoon into a leftist icon of environmentalism and health philanthropy,  has now published an essay calling for an AI slowdown. This is a massive shift is a shift for a tech evangelist who has built his post-Microsoft career on faith in AI’s ability to fix hard problems. Gates said this is the first time in his life he wishes a new technology would advance more slowly.

“I’m surprised to feel this way,” Gates said. “I’ve never seen a problem that innovation can’t solve.”

Gates grouped his worries into three categories: safety threats such as AI-enabled hacking, biological weapons and fraud; job losses driven by AI and AI-powered robots; and risks to children, including stunted learning and damaged human relationships. Writing software code and enabling new kinds of cyberattacks are the two areas where he says AI has recently outpaced his own expectations. Asked about the latter, he used one word: “mind-blowing.”

Gates argued the AI industry cannot police itself and called on the U.S. government to take the lead in regulating it. He wants broad national and international frameworks to address AI’s risks to security, safety, jobs and other areas, built around collaboration between elected officials, subject-matter experts and ordinary citizens. The idea echoes a proposal from AI leader Demis Hassabis, who has called for international AI safety standards overseen by an expert body.

“My basic view is nobody’s shown up,” Gates said of the lack of government action so far. “So this is a shrill call, and I want this to be bipartisan.”

Keep reading

Mark Zuckerberg’s Meta Agrees to $18 Billion Settlement with 29 States to End Teen Social Media Addiction Lawsuit

Mark Zuckerberg’s Meta agreed to pay roughly $18 billion to settle a lawsuit brought by 29 states over allegations that Facebook and Instagram harmed children through addictive design.

NBC News reports that the settlement, which still needs a judge’s approval, came one day after Instagram chief Adam Mosseri testified in a California courtroom. Meta CEO Mark Zuckerberg had also been expected to take the stand before the deal was reached.

In a statement, Meta said “the agreement includes a payment of approximately $18 billion, which can be used to fund youth online safety initiatives, among other state priorities.” The company said the deal is part of a broader agreement with 52 attorneys general nationwide, extending well beyond the 29 states that sued in California.

Meta did not concede wrongdoing. A court filing states the company “denies the allegations against it and that it has any liability to the Plaintiffs.”

Attorneys general from California, Colorado, Kentucky, New Jersey and other states had accused Meta of designing addictive products that harmed children and of violating federal privacy and consumer protection laws. Meta pushed back, arguing the states were cherry-picking features while ignoring safety tools it already offers, including teen accounts that default to private, time-limit reminders, parental supervision options, and restrictions on who can contact minors and what content they see.

Under the settlement, Meta will build in daily limits and nighttime blocks for teenage users. The court filing says the company “commits to establishing daily limits and blocks on nighttime use for teenage users.” Teen accounts on Facebook and Instagram will default to a combined two-hour daily use limit, and teens will need parental permission to turn that limit off.

Keep reading

Gaming Giant Roblox Says Teen Suicide Victim Waived Right to Sue When She Was Just 8 Years Old

Gaming platform Roblox is arguing that an 8-year-old girl gave up her right to sue the company by clicking “I agree” on its terms of service, according to court filings in a wrongful death lawsuit filed by her mother. The girl tragically took her own life at just 13 year old after being introduced to dark and violent subcultures on Roblox, TikTok, and Discord.

The Louisville Courier Journal reports that Jaimee Seitz filed the lawsuit in October 2025 in federal court in the Eastern District of Kentucky, alleging her daughter, Audree Heine, was exposed to “harmful and violent influences” on Roblox after joining the platform at age eight, despite parental controls Seitz says she had set on the account. Audree also used Discord and TikTok to communicate with other Roblox users, and the suit names both companies as defendants alongside Roblox.

Audree tragically died by suicide in December 2024, one week after her 13th birthday. According to the lawsuit, investigators later found a journal in her school locker showing she had been introduced to a community that glorified a mass school shooter and other violent ideologies through her interactions on Roblox, Discord, and TikTok.

Roblox, Discord and TikTok have each asked the court to dismiss the claims. Roblox also filed a motion to compel arbitration, arguing that Audree and her parents waived her right to a trial no fewer than 28 times, including by creating her account, accepting six updates to Roblox’s terms, redeeming gift cards twice, and buying the platform’s virtual currency, Robux, 19 times. “Roblox made the Terms plain and Audree repeatedly manifested her assent, as courts have routinely held in examining similar agreements and as the one at issue here,” the company’s attorneys wrote.

Seitz disputed that argument in comments to the Courier Journal. “So let’s be serious about an argument that if my child clicking a tiny ‘I agree’ box somehow means she knowingly agreed to arbitration,” she said. “She was a child, she didn’t understand arbitration. She didn’t understand contracts. She didn’t understand what the little box could mean years later. But now they want a court to treat that click as though an 8-year-old knowingly negotiated away her right to have these issues heard. It’s insulting.”

Alex Walsh, an attorney representing Seitz, said Roblox’s filing seeks to move the case into a “secret arbitration process” rather than a jury trial. “Why would the companies want that? For a very simple reason,” Walsh said. “They do not want the truth about how dangerous their platforms are to come out. They don’t want there to be a light shined on what they’ve done wrong and how many children have been harmed.”

Keep reading

Tech Giants Scramble to Calm Public Anger over AI Data Centers

In July, roughly 1,000 Georgia residents gathered at the Effingham County College & Career Academy to hear what OpenAI could offer their community. They also faced protesters angry about the impacts of AI data centers.

The Wall Street Journal reports that a day after announcing plans for a new data center near Savannah, Georgia, OpenAI held an open house for residents’ questions. Attendees passed protest signs reading “I didn’t vote for AI” and “You can’t drink data!!!” Inside were OpenAI-staffed booths with facility information, QR codes to a job board, site renderings and a taco bar. OpenAI pledged the community $80 million, plus up to $71 million in coding credits for local students.

OpenAI Chief Global Affairs officer Chris Lehane called the open house “an important piece” of winning community support. “People really do want to know, like, ‘Are my electricity bills gonna go up or not? Is this gonna impact my water supply? Am I gonna pay more or less in taxes?’” he said.

Open houses, multimillion-dollar pledges and free food were not standard for an industry that typically won approval through speed, confidentiality and political connections. That’s changing as grassroots backlash threatens the buildout’s momentum. Companies add job guarantees and water investments to deals, cite facilities that cut tax burdens, and frame AI’s impact around opportunity, not job losses.

“The politics of data centers have become more prominent and unambiguous across the country,” said Microsoft Vice Chair and President Brad Smith. Microsoft, he said, saw the need for a community-focused approach after the issue surfaced in last November’s elections, including New Jersey’s governor’s race, which Democrat Mikie Sherrill won after campaigning on electricity prices. “As soon as, especially, the New Jersey race was over, it was clear that this was on a different trajectory,” he said. Two months later, Microsoft announced commitments on electricity, water use, jobs and taxes.

Opponents cite rising electricity prices, environmental effects and noise; some object to AI itself over job-loss fears or broader social harm. Clifford Young, chair of Ipsos Public Affairs, said the backlash reflects a “broader-based belief that the system is broken and no longer working for the average person… it’s become, I would say, even a poster child for it.”

Keep reading

Meta Whistleblower Testifies Meta Took ‘Don’t Ask, Don’t Tell’ Approach to Child Safety

A former Meta safety engineer told a federal jury this week that company leaders repeatedly received warnings about harm to children on Facebook and Instagram and largely failed to act.

Quartz reports that Arturo Béjar, who worked as a safety engineer at Meta, testified that the company operated under what he described as a “don’t ask, don’t tell” philosophy toward child safety. He said internal studies showed children encountering harmful material at elevated rates, including recommendations that surfaced content from sexual predators and graphic violent images. Béjar said he brought these findings to Facebook and Instagram executives multiple times, and those conversations produced little meaningful change.

Béjar also testified that during his second stint at Meta, from 2019 to 2021, the company replaced the term “addiction” with “problematic use,” a framing he said did not meet the thresholds academic researchers use to define addiction. He said the metrics Meta made public gave a false picture of safety because they measured violations of content policies rather than actual harm experienced by users.

Béjar told the jury that in 2021 he emailed Meta CEO Mark Zuckerberg after Zuckerberg publicly stated that the company does not prioritize profit over safety. “I felt that he created a false and misleading impression of Facebook’s commitment to young people,” Béjar testified. He said he briefed Zuckerberg on product issues at least 100 times during his tenure at the company.

Béjar was the first witness called in the trial, which opened Tuesday in federal court in Oakland, California, before U.S. District Judge Yvonne Gonzalez Rogers. The case was brought by 29 state attorneys general who allege Meta intentionally designed Facebook and Instagram to be addictive to minors, collected data on children under 13 without parental consent in violation of the federal Children’s Online Privacy Protection Act, and misled the public about platform safety. Opening arguments were delivered by lawyers representing California, Colorado, Kentucky and New Jersey.

Meta attorney Paul Schmidt argued that the states had selectively quoted internal documents to build a misleading narrative. He pointed to the company’s safety measures, including parental controls, privacy defaults for teenage users, and reminders encouraging users to limit time on the app.

Keep reading

Amazon ‘Prime Air’ To Expand Drone Delivery To Nearly 500 US Cities This Year

Amazon’s Prime Air drone delivery service, first teased by Jeff Bezos in a 2013 60 Minutes interview as a way to get packages under 5lbs to customers in 30 minutes or less, represents the broader push toward autonomous aerial last-mile logistics that could bypass traffic and cut delivery times dramatically. 

After years of regulatory hurdles with the FAA, technical iterations, and limited trials, the service has matured into a commercial offering using the MK30 drone, which hovers to drop packages.

A major breakthrough came in 2024, when the Federal Aviation Administration (FAA) granted Amazon a waiver allowing its drones to fly beyond the visual line of sight of their operators, and has already completed hundreds of thousands of deliveries in 2026 across 11 sites in seven states. 

Today Amazon announced plans to expand it to nearly 500 US cities and towns by year-end – a roughly sixfold increase – bringing ultrafast options (as quick as 30 minutes) to tens of millions more customers in places like the Chicago, Atlanta, Cleveland, Syracuse, and Boise metro areas.

The drones will primarily operate in suburban areas, away from skyscrapers and major airports that could complicate operations.

Amazon describes the aircraft as “highly autonomous,” with onboard cameras and sensors for navigation, obstacle detection, and safe delivery.

The cameras do not transmit a live video feed, according to the company.

The company also sought to address the potential concern about noise.

“During drop-off, the sound level is below that of an idling delivery truck parked curbside and lasts about 30 seconds,” Amazon said.

Someone standing outside may hear a sound “comparable to a window fan on low” as the drone arrives, while people indoors may not hear it at all, the company added.

Prime Air delivery is free for Prime members on orders of at least $50, but orders below that threshold carry a $2.99 delivery fee, while customers without a Prime membership are charged $4.99.

Keep reading

States take Meta to trial in California in the biggest fight yet over social media harms to children

Of the thousands of lawsuits Meta faces over child safety on its platforms, none may be more consequential than one going to trial this week in California.

States are seeking extensive financial damages that could, in theory, total as much as $1.4 trillion, plus changes to how the company operates Facebook and Instagram.

The lawsuit accuses the social media giant of contributing to the youth mental health crisis by knowingly and deliberately designing features that get children addicted to its platforms. It also claims that Meta routinely collects data on children under 13 without their parents’ consent, in violation of federal law.

“Meta has harnessed powerful and unprecedented technologies to entice, engage, and ultimately ensnare youth and teens. Its motive is profit, and in seeking to maximize its financial gains,” the lawsuit says.

Dozens of states filed the lawsuit three years ago. The trial set to begin Tuesday in federal court in Oakland, California, features four of the states as plaintiffs — California, Colorado, Kentucky and New Jersey. The other 25 states are expected to have trials later.

Meta said it disputes the allegations, and the trial evidence will show its commitment to supporting young people. “We’ve listened to parents, worked with experts and law enforcement, and conducted in-depth research to understand the issues that matter most,” the company said in a statement.

States seek to land a major blow against Meta
For Meta, which already lost two pivotal cases over harms to children and teens this year, the stakes are high. The company reported a rare profit decline last month, in part due to $2.4 billion in legal expenses.

The $1.4 trillion figure, which Meta disclosed in a legal filing, is almost as high as the Menlo Park, California, company’s entire market capitalization — that is, the value of all its outstanding shares on the stock market. Paying it would inevitably put Meta Platforms in bankruptcy and perhaps put the company under state ownership.

“The state attorneys general are going for the gusto,” said Eric Goldman, a professor and co-director of the High Tech Law Institute at Santa Clara University School of Law. “They are trying to set the definitive precedent in this case and they have asked for extraordinary damages and they are going to seek extraordinary structural remedies if they succeed.”

Meta calls the possible penalty “untethered to any claimed violation” by the states.

“A sanction of that size has no analog in the history of consumer protection enforcement,” Meta said in a July 6 filing with the U.S. District Court for the Northern District of California.

If Meta loses the trial, the court would have wide discretion over the size of any financial penalty, and legal experts say anything close to $1.4 trillion would be unlikely.

“It’s not plausible in the sense that Meta doesn’t have that much money and could not get it,” said James Grimmelmann, a law professor at Cornell Law School and Cornell Tech. “An award that large would put Meta into bankruptcy, wipe out its owners, and effectively result in the states owning Meta.”

Keep reading