As our reporting continues on how AI is changing the child safety landscape, I’d like to hear from people working in tech, trust and safety, law enforcement + child protection about how detection, reporting and investigations are evolving.
I can be reached on Signal at @rgriffin.42.
Above is a 🎁🔗to my story w/Matt Day, edited by @kurtwagner.bsky.social. It dives into why the National Center for Missing & Exploited Children calls Amazon an "outlier," both in the scale of what it detected in training data and in how little information it shared about where the data originated.
AI safety experts warn that quickly amassing large datasets without proper safeguards comes with grave risks.
As one expert put it: “If you hoover up a ton of the internet, you’re going to get [child sexual abuse material].”
The surge in Amazon's reports comes as technology companies race to gather enormous datasets to improve AI models. That scramble is already reshaping how child sexual abuse material is discovered, reported and investigated.
That lack of detail can limit efforts to remove the material from circulation, identify perpetrators and protect victims of abuse who may still be at risk, the child safety officials said.
The tech giant says it removed the material, which it obtained from outside sources, before actually training its models. But child safety officials told me Amazon has provided “very little to almost no information” about where the explicit content actually came from.
NEW: Amazon detected hundreds of thousands of pieces of suspected child sexual abuse material last year in data collected to train its artificial intelligence models. 🧵
If you’re involved in the nationwide AI infrastructure buildout or working on immigration policy right now, I’d like to hear how these competing priorities are playing out across the country. Please reach out! I can be reached on Signal from a nonwork device at @rgriffin.42
This week, ICE agents were present outside Meta’s Louisiana site and arrests were made, highlighting growing tension between the Trump administration’s push to drive investment in AI infrastructure and its immigration enforcement priorities.
GIFT LINK, here: shorturl.at/jdtLl
Projects of this scale, with accelerated timelines and massive workforce needs, have long relied on immigrant labor, a standard feature of large US construction builds.
The US push to rapidly expand AI infrastructure is colliding with realities on the ground.
I’ve been reporting on one of the most ambitious data center projects underway, what’s poised to become the largest in Meta’s fleet. At peak, the site is expected to require 5K+ workers.
This formal break between FWD.us and its biggest benefactor coincides with CZI's narrowed focus on science, rather than social advocacy, and Zuckerberg's broader recalibration toward the Trump administration.
For more detail on how FWD.us is navigating this change, check out our story.
This year, that support ended entirely. The Chan Zuckerberg Initiative provided no funding to FWD.us in 2025, relinquished its board seat, and was removed from the organization's bylaws.
That public advocacy faded as scrutiny of Facebook intensified, even as his philanthropy continued backing the group, including with a $100 million investment announced in 2021.
More context: Zuckerberg launched FWD.us in 2013 at the height of Silicon Valley’s push for expanded legal immigration, casting the issue as both a moral obligation + an economic necessity for the tech industry. He spoke out for Dreamers and was critical of the first Trump administration's policies.
Ben Steverman and I dove into the details of the split between the Chan Zuckerberg Initiative and FWD.us. You can find our reporting for @bloomberg.com, below. (GIFT LINK!)
NEW: Mark Zuckerberg once made immigration reform a signature cause. More than a decade later, his philanthropy has fully cut ties with the pro-immigration advocacy group he founded, FWD.us, underscoring how his posture has shifted over the years, and most recently as Trump returns to power.
Sources: The Chan Zuckerberg Initiative cuts ties with FWD.us, an immigration advocacy group founded by Mark Zuckerberg, eliminating FWD.us' main funding source (Bloomberg)
Main Link | Techmeme Permalink
🚨The FDA has put off a long-promised mifepristone safety review at Commissioner Marty Makary’s request, seeking to push the controversial study past the midterms to sidestep political fallout.
Gift link to my scoop w/ @rachelcohrs.bsky.social +Charles Gorrivan here: shorturl.at/ED0OI @bloomberg.com
Meta is moving into electricity trading, betting it can speed up construction of new US power plants to fuel its AI ambitions.
Our interview with Meta’s energy chief shows how Big Tech is reshaping America’s energy markets: www.bloomberg.com/news/article...
Meta is poised to announce the completion of 2Africa—a 45,000-km subsea cable looping Africa—but its Red Sea stretch remains stalled. Meta isn’t alone: regional security threats have sidelined multiple projects.
🔗SCOOP with Loni Prinsloo, Jillian Deutsch and Olivia Solon here: shorturl.at/loYga
Headline: Reddit seeks to strike next AI content pact with Google, OpenAI - Moving past a dollars-for-data model, Reddit seeks new deals - The social platform looks to bring new users into its forums
Reddit Inc. is in early talks to strike its next content-sharing agreement with Alphabet Inc.’s Google, aiming to extract more value from future deals now that its data plays a prominent role in search results and generative AI training. Reddit, more than a year and a half after its first data-sharing deal with Google for a reported $60 million, is in talks for deeper integration with Google’s AI products, according to executives familiar with the discussions. It’s proposing a new kind of partnership that would encourage users to become active contributors to Reddit’s popular online forums, so Google traffic could help the company grow and generate content for future training.
Scoop: Reddit is in talks with Google to strike a new AI content deal—one that would get the platform paid more for its data, which has become crucial for training Google's AI & powering search results
w/ @rileyraygriffin.bsky.social & @shiringhaffary.bsky.social
www.bloomberg.com/news/article...
Joined @vox.com epic pod, "Today, Explained" to chat AI talent wars, Zuckerberg's multi-hundred million dollar comp offers and ... that evasive concept of 🧠*superintelligence* 🧠
Listen up 🔊 open.spotify.com/episode/0fi5...
PLUS: It's a two-fer with my ol' pal,
@gerritd.bsky.social
SCOOP: President Trump threatened to inflict “substantial” tariffs on countries that impose digital taxes just days after Mark Zuckerberg raised concerns about the levies during a White House meeting.
Dive into the details: bloomberg.com/news/article... @bloomberg.com
The interview comes ahead of Clegg's new book, How to Save the Internet — and as he starts a new chapter, working to grow the tech sector in Europe and the UK.
More at the link below + in our coverage!
Clegg says he wasn't the right man to lead Meta into Trump 2.0.
And he warns Silicon Valley: The political pendulum will always swing back. Zigging & zagging won't serve them in the long term.
🎁🔗to the interview in @bloomberg.com's Tech in Depth newsletter: www.bloomberg.com/news/newslet...
Nick Clegg spoke with me and @kurtwagner.bsky.social a few months after stepping down from his longtime post at Meta—a job that saw him help steer the company through its biggest political and reputational "firestorms," from the Cambridge Analytica scandal to Russian election interference online.
A UK deputy PM turned Meta political fixer tells me Silicon Valley has overcorrected by leaning to hard into US politics and the "America first" agenda.
But he’s not surprised: “Silicon Valley is a place of stampedes and fads,” says Nick Clegg, Meta’s ex–global affairs chief.🧵
Scientists have released the closest images ever taken near the sun, captured by NASA’s Parker Solar Probe flying 3.8 million miles above the solar surface.
See more images of solar eruptions on the sun: wapo.st/3Irm3P3
