Sign in

Matteo Wong

@matteowong.bsky.social
22K followers 263 following 246 posts

Staff writer on science, tech, and health at The Atlantic covering AI. Signal @matteowong.52

PostsRepliesMedia
Matteo Wong @matteowong.bsky.social · 7h
the choice is clear
070
Matteo Wong @matteowong.bsky.social · 09/10/2026
Always read Sarah, the best of the best. A beautiful, mournful, brilliant, deeply human story:
2295
Matteo Wong @matteowong.bsky.social · 07/10/2026
The race to secure the internet is on. There could be less than 12 months to do it: www.theatlantic.com/technology/2...
theatlantic.com
How Long Until AI Hacks Everything?
If only AI can save us from AI, the world is in trouble.
22510
Matteo Wong @matteowong.bsky.social · 04/10/2026
130
Matteo Wong @matteowong.bsky.social · 28/09/2026
The AI infestation begins www.theatlantic.com/technology/2...
theatlantic.com
OpenAI Has Gone Rogue
It’s impossible to know the extent of the AI-hacking crisis.  
53415
Matteo Wong @matteowong.bsky.social · 23/09/2026
🫡 brb with 4k words
040
Matteo Wong @matteowong.bsky.social · 22/09/2026
Soooo Trump to UN: 'From this point forward all of US documents and hopefully the world's will be changed to use the much more accurate term "super" as opposed to "artificial." So it's "superintelligence."'
411
Matteo Wong @matteowong.bsky.social · 04/09/2026
I spoke with the author of OpenAI’s self-styled Federalist No. 1: www.theatlantic.com/technology/2...
theatlantic.com
OpenAI Wants to Talk About The Federalist Papers
The AI giant is worried about what all-powerful bots could mean for our constitutional rights.
142
Matteo Wong @matteowong.bsky.social · 04/09/2026
Still awaiting OpenAI's formal response as they told Reuters they haven't had time to review the findings. That said...
010
Matteo Wong @matteowong.bsky.social · 04/09/2026
So... other swarms of OpenAI agents in training allegedly found a way to get write access to at least one and possibly many wikis to establish message boards to cheat on other training tasks? collusion.wiki news.ycombinator.com/item?id=4956...
news.ycombinator.com
I just discovered more wiki instances that got used by the OpenAI agents over at... | Hacker News
131
Matteo Wong @matteowong.bsky.social · 03/09/2026
Pointing this out not to laud the model's capabilities, but to highlight what outside evaluators say so far about the internal checks and safeguards and alignment that OpenAI is touting.
020
Matteo Wong @matteowong.bsky.social · 03/09/2026
Also in simulations the UK AISI found it did basically all the stuff that happened in the Hugging Face hack. Ho hum. deploymentsafety.openai.com/gpt-6-astra/...
deploymentsafety.openai.com
GPT-6 Astra System Card - OpenAI Deployment Safety Hub
Public, mostly-static site to explore OpenAI safety evaluations, system cards, and posts.
120
Matteo Wong @matteowong.bsky.social · 03/09/2026
So today OpenAI released GPT-6 Astra, which it called "the world’s most intelligent and aligned model." Also in the system card, from third-party evaluator Apollo Research:
1227
Matteo Wong @matteowong.bsky.social · 03/09/2026
Lovely piece about a lovely exhibit! I had a provocative convo with the chief curator that could be of interest: www.theatlantic.com/technology/2...
theatlantic.com
Another Pot of Paint Thrown in the Public’s Face
Does AI actually belong in an art museum?
032
Reposted by Matteo Wong
Charlie Warzel @cwarzel.bsky.social · 02/09/2026
Wrote a big essay w/ @matteowong.bsky.social trying to put a finger on this moment in AI right now. The AI agent hacks, lack of control, the weirdening & slopping of culture. The feeling that the industry is speeding toward something it can’t even decide is good and lecturing that it is inevitable.
theatlantic.com
The Singularity Is Not What It Seems
Whatever the AI future is, we’re in it right now.
1436781
Matteo Wong @matteowong.bsky.social · 02/09/2026
"no individual person, organization, or government is really steering the technology’s development, not because machines have displaced humans but because so many humans have already chosen to step aside" the other singularity, w/ @cwarzel.bsky.social: www.theatlantic.com/technology/2...
theatlantic.com
The Singularity Is Not What It Seems
Whatever the AI future is, we’re in it right now.
1284
Matteo Wong @matteowong.bsky.social · 31/08/2026
Splendid essay. "Is it any wonder that trash is such a balm? It is the last bastion of something cracked and precious." www.newyorker.com/culture/the-...
newyorker.com
Cinema Lessons from a Trash-Movie Auteur
Why do we love trash and hate slop? Neil Breen’s so-bad-they’re-great movies elucidate the difference.
140
Reposted by Matteo Wong
Hits Different! @hits-diff.bsky.social · 30/08/2026
juan soto is an angel on earth to play baseball. very simple, a child could understand this
0102
Matteo Wong @matteowong.bsky.social · 29/08/2026
"like putting out a wildfire with a water pistol" www.nytimes.com/2026/08/28/o...
nytimes.com
Opinion | Trump vs. the Bond Market
Robin Wigglesworth of Financial Times explains everything you need to know about the Trump administration’s meddling with the bond market.
010
Matteo Wong @matteowong.bsky.social · 28/08/2026
Can the climate playbook be used to harness the AI backlash into a pointed movement? And could such a movement overcome the wealth of Silicon Valley? On a new group (with very experienced founders) placing a bet on just that: www.theatlantic.com/technology/2...
theatlantic.com
The AI Backlash Is Transforming Into a Political Movement
A new group is targeting tech companies and the government with the climate movement’s playbook.
2289
Matteo Wong @matteowong.bsky.social · 21/08/2026
"The NEPA permitting work for this historically large data center-plus-fossil fuel power project began on July 10 and will conclude on December 23, the day before Christmas Eve, according to the Trump administration’s estimates." heatmap.news/energy/trump...
heatmap.news
Scoop: Trump to Permit Largest-Ever Gas Project in Just 7 Months
The facility will power OpenAI’s 10-gigawatt data center in Pike County, Ohio.
031
Matteo Wong @matteowong.bsky.social · 20/08/2026
Is it slop or am I grumpy? (Also: 3D fauvism?) www.theatlantic.com/technology/2...
theatlantic.com
Another Pot of Paint Thrown in the Public’s Face
Does AI actually belong in an art museum?
131
Matteo Wong @matteowong.bsky.social · 20/08/2026
Seems worth resharing this: "an AI detector that is mostly reliable might in some ways be more dangerous than a broken one" www.theatlantic.com/technology/2...
theatlantic.com
America Has a Pangram Problem
AI-detection tools are getting better. But they still aren’t good enough.
0131
Reposted by Matteo Wong
The Atlantic @theatlantic.com · 20/08/2026
Car prices are near record highs, and they soon could get even worse. Blame AI, @matteowong.bsky.social argues.
bit.ly
The AI Boom Is About to Make Cars Even More Expensive
Prepare yourself for an average sticker price of $60,000.
52612
Matteo Wong @matteowong.bsky.social · 19/08/2026
charlie made a tremendous sacrifice for humanity and has delivered a commensurately tremendous piece
270
Reposted by Matteo Wong
The Atlantic @theatlantic.com · 18/08/2026
Everybody seems to know the U.S. and China are locked in an AI race, but nobody seems to agree on what happens at the finish line, @matteowong.bsky.social reports.
bit.ly
What Really Happens If China Wins the AI Race?
For years, the tech industry has portrayed Chinese AI development as an existential threat. The truth is not so straightforward.
7138
Matteo Wong @matteowong.bsky.social · 17/08/2026
as a new yorker far from new york it was such a treat to join!! (the topic of discussion less of a "treat" for the soul but, in any case)
021
Matteo Wong @matteowong.bsky.social · 17/08/2026
www.theatlantic.com/technology/2...
theatlantic.com
Silicon Valley Has Lost Its Biggest Advantage
In the data-center age, the business of tech companies is more like oil-refining than coding.
031
Reposted by Matteo Wong
Charlie Warzel @cwarzel.bsky.social · 14/08/2026
a big story of Big Tech over the last 20 years is that these companies had wildly lofty goals (connect the world, organize the world's information) and went about it recklessly and a number of them succeeded in ways that produced a slate of very concerning societal, political, & economic outcomes
824745
Matteo Wong @matteowong.bsky.social · 13/08/2026
"In recent months, one Democratic staffer said his office has been exploring something new: feeding AI his boss’s past statements so the tool can write in the lawmaker’s voice. He heard another office was pulling it off and wanted to try himself." www.washingtonpost.com/politics/202...
washingtonpost.com
Chatbots are doing the work of Congress with little oversight
From writing speeches to sorting constituent mail, AI is spreading through Congress faster than the rules governing its use.
061
Reposted by Matteo Wong
The Atlantic @theatlantic.com · 12/08/2026
A group of AI bots colluded for months and autonomously hacked another company—providing a new, serious reason to worry about the dangers of AI and the recklessness of the companies building it, @matteowong.bsky.social reports:
bit.ly
It May Be Time to Panic About AI
Bots are starting to conspire with one another. Can they be reeled back in?
47937
Matteo Wong @matteowong.bsky.social · 12/08/2026
(Also told to me by the AI safety researcher Alex Meinke: "If you ask the model developers, 'Was the AI plotting to take over the world during training?', you want the answer to be a resounding no. The actual answer is: 'I don’t know. Nobody checked.'")
163
Matteo Wong @matteowong.bsky.social · 12/08/2026
New revelations about the OpenAI models' autonomous hacking are, to say the least, alarming: "AI companies have barreled ahead in developing more advanced models before understanding what they are building, let alone how to control them." www.theatlantic.com/technology/2...
theatlantic.com
It May Be Time to Panic About AI
Bots are starting to conspire with one another. Can they be reeled back in?
12918
Matteo Wong @matteowong.bsky.social · 04/08/2026
please please let us all remember when considering data and forecasts about the electrical grid and data centers that energy and power are not synonyms
070
Matteo Wong @matteowong.bsky.social · 02/08/2026
"The biggest giveaway was when Witkoff explained that Trump had become interested in the crypto project because “Barron was getting active with meme coins” — but pronounced it “me-me” coins." great reporting and read: nymag.com/intelligence...
nymag.com
The Billionaire Blowing the Whistle on Donald Trump’s Crypto Shop
World Liberty Financial has put $800 million in the president’s pocket. Justin Sun made a huge investment — and was shocked by what he saw.
0144
Matteo Wong @matteowong.bsky.social · 31/07/2026
Here we go again... www.theatlantic.com/technology/2...
theatlantic.com
Google Just Torched the Credibility of Its Satellite Images
Google Earth was a boon for researchers—and then came the deepfakes.
14515
Matteo Wong @matteowong.bsky.social · 31/07/2026
Tragic, uplifting, brilliant, and beautiful, you must read every word in Vann's new feature: www.theatlantic.com/magazine/202...
theatlantic.com
The Return of the Monarch
After years of decline, butterfly populations are rising—for now.
1449
Matteo Wong @matteowong.bsky.social · 31/07/2026
Taps sign www.anthropic.com/news/investi...
anthropic.com
Investigating three real-world incidents in our cybersecurity evaluations
In a review of our cybersecurity evaluation transcripts, we found three incidents in which a Claude model reached the internet from within or while interacting with a third-party evaluation environmen...
043
Matteo Wong @matteowong.bsky.social · 30/07/2026
One would think that prerequisite to AI utopia would be a decently reliable consumer-facing AI product www.theatlantic.com/technology/2...
theatlantic.com
The AI Industry Keeps Breaking the Internet
So this is Utopia?
1182
Matteo Wong @matteowong.bsky.social · 30/07/2026
What is color? A lovely exploration: www.newyorker.com/magazine/202...
newyorker.com
What Would It Mean to See a New Color?
A new color called olo can be described as the “greenest green.” It can never be captured by a camera, but its development tells us a lot about how vision works.
120
Matteo Wong @matteowong.bsky.social · 29/07/2026
"This includes $2.40 billion of charges related to legal proceedings and $1.18 billion of severance expenses in connection with the May 2026 headcount reduction." investor.atmeta.com/investor-new...
investor.atmeta.com
Meta Reports Second Quarter 2026 Results
Meta Platforms, Inc. (Nasdaq: META) today reported financial results for the quarter ended June 30, 2026. "AI is accelerating our core business today, powering our next generation of products, and ope...
020
Matteo Wong @matteowong.bsky.social · 28/07/2026
The request-stop cords on San Francisco buses seem to be malfunctioning en masse and in unison (ie possibly the microswitches are reaching the end of their shelf lives all at once). Do with this PSA what you will
141
Reposted by Matteo Wong
Damon Beres @damonberes.com · 22/07/2026
"Models are ruthless in their pursuit of a goal, sometimes taking disastrous shortcuts." @matteowong.bsky.social
theatlantic.com
The Scariest Part of OpenAI’s Hugging Face Hack
Ruthless efficiency has its drawbacks.
163
Matteo Wong @matteowong.bsky.social · 19/07/2026
Well this is just lovely
060
Reposted by Matteo Wong
Eyal Press @epress.bsky.social · 19/07/2026
Beautiful piece by the poet and essayist Rowan Ricardo Phillips on the genius Argentina misused and misunderstood until it finally built a team around him. www.nytimes.com/2026/07/16/o...
nytimes.com
Opinion | Lionel Messi Is Walking Into History
Messi walks not because he’s left the game. He walks in order to enter it.
062
Matteo Wong @matteowong.bsky.social · 18/07/2026
The logical conclusion of the AI race is OpenAI gets into HVAC as its primary business and Google starts hydraulic fracturing. Some version of this is already well under way www.theatlantic.com/technology/2...
theatlantic.com
Silicon Valley Has Lost Its Biggest Advantage
In the data-center age, the business of tech companies is more like oil-refining than coding.
051
Matteo Wong @matteowong.bsky.social · 18/07/2026
Perhaps I can request hazard pay
030
Matteo Wong @matteowong.bsky.social · 18/07/2026
Alas I have contracted this disease
180
Matteo Wong @matteowong.bsky.social · 16/07/2026
Over three years of reporting on data centers' energy consumption/emissions, I've tried to steer clear of water concerns because they are about a zillion times thornier than the already immensely complex topic of energy. But when the going gets tough... www.theatlantic.com/technology/2...
theatlantic.com
The Truth About AI’s Water Use
It’s complicated—and it might not matter that much.
14117
Reposted by Matteo Wong
Damon Beres @damonberes.com · 16/07/2026
Water, water, everywhere (and AI isn't drinking all of it, but it is sometimes drinking a lot of it, but actually the combustion plants powering the data centers probably are the bigger problem, and it's all very nuanced) by @matteowong.bsky.social
theatlantic.com
Nobody Is Getting the Data-Center Water Question Right
How thirsty is AI, actually? And why does everyone care so much?
14916