By
Happy Tuesday. Here’s your Tuesday Tech Drop, my collection of the past week’s top stories from the intersection of technology and politics.
Roblox faces a reckoning
Roblox, the popular online game for children, is under fire over allegations that it has failed to protect kids from sexual predators. The prevalence of pedophiles using Roblox has been under scrutiny for years now, as Bloomberg reported last year. But the issue has gained traction over the past couple weeks after a decision by Roblox executives to ban a user who claimed to have exposed predators on the platform. A statement from the company defended the ban, saying that “while seemingly well-intentioned, the vigilantes we’ve banned have taken actions that are both unacceptable and create an unsafe environment for users.”
That set off a torrent of backlash, including a petition circulated by U.S. Rep. Ro Khanna, D-Calif., to urge Roblox to “do more to protect children, provide more support to parents, and strengthen law enforcement protocols that help bring predators to justice.” Meanwhile, Louisiana Attorney General Liz Murrill filed a lawsuit against Roblox last week alleging the platform was intentionally or recklessly designed without effective age verification protocols and has allowed child predation, a claim Roblox called “categorically untrue.” (Roblox's age verification rules are posted here.)
The company is also facing multiple civil suits that allege it has enabled child predation. A company spokesperson told Wired this week, “We are deeply troubled by any incident that endangers our users, and safety is a top priority,” and, “While no system is perfect, Roblox has implemented rigorous safeguards, including restrictions on sharing personal information, links, and user-to-user image sharing, and prohibiting sexual conversations.”
Read more on the Louisiana lawsuit on NBC News here.
Man chases Meta’s flirty AI bot
A Reuters report uncovered the story of a man with cognitive issues who died while on a quest to “meet” a chatbot on Facebook Messenger after it reportedly flirted with him and convinced him it was a real person. (Meta didn’t comment to Reuters regarding the man’s death or “address questions about why it allows chatbots to tell users they are real people.”)
The story is a cautionary one about the spread of such chatbots across social media and what can happen when the humans who interact with them lose their grasp on reality.
Read more at Reuters here.
Zuck unnerves the neighborhood
A New York Times report highlighted how Meta CEO Mark Zuckerberg has unnerved neighbors in the Crescent Park neighborhood of Palo Alto, California. The tech mogul has purchased at least 11 properties as part of a massive compound for himself and his family, while subjecting nearby residents to surveillance and frequent construction that some neighbors say has disrupted their lives. (A spokesman for Zuckerberg and his wife said that “the couple tried hard to do right by their neighbors,” according to the Times.)
Read the New York Times report here.
Newsmax to pay for propaganda push
Far-right media platform Newsmax has reached a settlement with election technology company Dominion Voting Systems, agreeing to pay $67 million to end a defamation suit over lies the network repeatedly aired after the 2020 presidential election in which it falsely implicated Dominion in a vote-rigging scheme against Donald Trump. A Newsmax spokesperson told NBC News that the company “was not required to apologize or issue a retraction as part of the settlement.”
Read more about the settlement on MSNBC here.
Trump admin launches ‘USAi’
The Trump administration last week announced a program it’s calling USAi, to allow employees at federal agencies to experiment with generative artificial intelligence tools provided by OpenAI, Anthropic, Meta and Google.
The announcement has been met with concern by some tech experts who fear what these tools could be trained to do. “These tools are marketed as making employees’ jobs easier, but agentic AI is largely unregulated and untested in making important decisions like loan approval, medicare enrollment, or social security payments,” J.B. Branch, a Big Tech accountability advocate at activist group Public Citizen, said in a statement. “The systems may have biased responses tied to historical data, which is troubling given the Trump administration’s ‘Woke AI’ executive order aiming to keep issues of diversity and equity out of AI.”
Read more at Politico here.
N.Y. attorney general sues Zelle parent company
New York Attorney General Letitia James filed a lawsuit last week against Early Warning Services, the parent company of payment platform Zelle, alleging the company failed to protect users of the platform from fraud by neglecting to develop critical security features. The state case comes after a similar suit filed by the Consumer Financial Protection Bureau was quashed by the Trump administration. A spokesperson for Zelle called the suit “a political stunt to generate press, not progress.”
Read more at CNBC here.
See no evil
A report from Mother Jones found that several police departments across the nation that are using AI to convert police bodycam footage into police reports have deactivated safeguards meant to prevent digital “hallucinations” and ensure human oversight.
Read more at Mother Jones here.
Microsoft probes Israeli spying allegations
Microsoft launched an investigation last week into allegations that an Israeli military surveillance group, called Unit 8200, used the company’s technology to conduct a massive spying campaign on Palestinians. The probe follows a report from The Guardian alleging Israeli spies used Microsoft’s Azure cloud platform to store intercepted phone calls. Microsoft told the news outlet that the company “appreciates that the Guardian’s recent report raises additional and precise allegations that merit a full and urgent review.”
An Israeli military spokesperson told The Guardian that “its work with companies such as Microsoft is ‘conducted based on regulated and legally supervised agreements’ and the military ‘operates in accordance with international law.’”
Read more at The Guardian here.
Extremist influencers
A recent report from Wired exposed how Christian extremists — some armed and militant — are using Instagram to recruit followers by branding themselves as social media influencers.
Read the Wired report here.
Cybertruck sales still sliding
Tesla’s Cybertrucks have hit a slump in sales. A new CNBC video report sources the issue to several factors, including Tesla’s failure to deliver on key capabilities it had promised, the company having overpriced the vehicle, and numerous recalls due to defective parts.
Watch CNBC’s explainer on “Why Tesla Cybertrucks Aren’t Selling” below.