Michael Utz
banner
theutz.com
Michael Utz
@theutz.com
🇺🇸 → 🇫🇮 → 🇹🇷
👨🏻‍💻 + 👩🏻‍⚕️ → 👦🏻 + 👱🏻‍♀️
Reposted by Michael Utz
This is well put. So *many* technologies are joyous and progressive. The web was invented in a PUBLICLY FUNDED UNIVERSITY! It is, in fact, an aberration that the fascist VCs think they're the default with their approach to tech. They were only ever supposed to be a tiny fraction of the ecosystem.
Once, technology solved problems. People liked having problems solved, so they liked technology.

Tech execs started to think of that sentiment as their due. So when they stopped solving problems, and people stopped liking them, they became outraged. "How dare you not love whatever we give you?"
I am a luddite and to me there is a lot of joy in technology. In technology that we deploy for all our wellbeing. Solar panels and vaccines, high-speed rail and wikipedia.

But they joy stems from the feeling of being able to be part of humanity in an embedded, meaningful way. Not from buying shit.
February 19, 2026 at 5:42 PM
Reposted by Michael Utz
is the problem really that lefties aren't enthusiastic about technology, or is the problem that "technology" has broadly lost its valence as an apolitical and universally positive force in public life, and must now be re-justified through a culture war lens?
February 18, 2026 at 2:28 PM
Reposted by Michael Utz
So after Trump pardoned Honduras's narco-president, the admin didn't just let him out--they canceled an immigration hold and sent a specialized team to **chauffeur him to the $1k/night Waldorf Astoria.**

“It's "absolutely fucking nuts,” said one official

www.propublica.org/article/trum...
February 18, 2026 at 1:21 PM
One of my favorite things about using open-source software is they’re much more likely to be developed with sensible engineering priorities.

Corporate APIs often suffer from the 1000 papercuts of a product manager not prioritizing technical debt. All kinds of weird, undocumented behavior.
February 17, 2026 at 10:24 AM
🟪 Purple Word #34

✅ Solved in 10 guesses

🏆 Top 20% of players today

🔥 1 day streak

purpleword.com
Purple Word - Daily Word Puzzle Game
A daily word puzzle game. Can you find the purple word? Play free online word game with daily challenges!
purpleword.com
February 16, 2026 at 8:04 AM
Reposted by Michael Utz
This great piece resonates with my own sense of loss and mourning of my social and professional identity as a computer programmer as the community of software engineers embraces LLMs and continues to sink further into surveillance and behavioral manipulation-based technofascistic business models.
A programmer's loss of identity - ratfactor
ratfactor.com
February 14, 2026 at 6:11 PM
Selfish AI | GarfieldTech
www.garfieldtech.com
February 14, 2026 at 6:49 AM
Reposted by Michael Utz
Analysis | The fact that Joe Biden isn't in the Epstein files shows how politically and culturally irrelevant he has become.
February 13, 2026 at 1:09 PM
Reposted by Michael Utz
Notice how the popular media narrative has moved on, while people on the ground in Minneapolis are still struggling and crying out that things are as bad as ever. This is how fascist regimes operate — overreach, recalibration, normalization. Get louder, not quieter.
February 11, 2026 at 3:54 AM
Reposted by Michael Utz
Hold on

I just watched a video where someone said the "why did the chicken cross the road?" joke is because the chicken wanted to die - the "other side" - but I have been alive for a very long time and this is news to me
February 11, 2026 at 8:55 PM
Reposted by Michael Utz
⚠️ Despite all the hype, chatbots still make terrible doctors. Out today is the largest user study of language models for medical self-diagnosis. We found that chatbots provide inaccurate and inconsistent answers, and that people are better off using online searches or their own judgment.
February 9, 2026 at 5:08 PM
Reposted by Michael Utz
This is because a statistical model of word frequency is not a useful tool for modeling the complex interactions of the human body
⚠️ Despite all the hype, chatbots still make terrible doctors. Out today is the largest user study of language models for medical self-diagnosis. We found that chatbots provide inaccurate and inconsistent answers, and that people are better off using online searches or their own judgment.
February 10, 2026 at 2:27 PM
Reposted by Michael Utz
A pyramid scheme built on a Lin Manuel Miranda chat bot

An MLM LMM LLM
February 10, 2026 at 3:14 PM
Reposted by Michael Utz
In the unlikely event of cabin depressurization, AI agents will automatically drop from the ceiling. Please ask it what to do before helping any other passengers. In the event of a water landing, your seat cushion can be used to place up-to-the-minute bets on Draft Kings. We now accept crypto.
February 10, 2026 at 2:08 PM
Reposted by Michael Utz
this was surprisingly good (it's about ai for programming in case you don't want to read that)
www.lelanthran.com/chap14/conte...
C and Undefined Behavior
www.lelanthran.com
February 8, 2026 at 8:17 PM
Reposted by Michael Utz
the vibe
February 7, 2026 at 7:26 PM
Reposted by Michael Utz
Van Hollen has been much better than most Democrats, but once again, I am asking elected leaders to stop providing running commentary like a common pundit or beat reporter and instead tell *us* what to do about it or tell us what *they* are doing about it.
Foreign countries are bribing our president to sell out the American people.
 
Trump family businesses made $187 MILLION from this deal, and just months later he gave the UAE some of our most top-secret AI tech.
 
They are selling our national security to the highest bidder.
February 2, 2026 at 3:13 PM
Reposted by Michael Utz
Yes. Every comment or README (or any other files) of every dependency, and sub-dependency and sub-sub-dependency... ad infinitum. Along with any docs, man pages or examples about the code or APIs that the Agent looks at online. Anything an Agent even looks at is a potential attack vector.
February 1, 2026 at 7:05 PM
Given the way prompt injection works, am I to understand that every COMMENT in every package of every dependency ever written is now a potential attack vector? Do I understand this correctly @internetofbugs.com @edzitron.com ?
February 1, 2026 at 4:02 PM
Reposted by Michael Utz
I'd like experts to talk about the cognitive load AI forces even non-AI users to take on, because we're now forced to filter through scads of junk info daily without the social and structural supports that once guided us. The decision fatigue alone could be causing serious social and health fallout.
January 29, 2026 at 8:05 PM
Reposted by Michael Utz
blogged: Mozilla Slopaganda
dbushell.com/2026/01/28/m...

— I read it so you don’t have to.
Mozilla Slopaganda
The one where I question reality
dbushell.com
January 28, 2026 at 8:40 AM
www.shipmap.org
This is pretty fascinating!
Shipmap.org
An incredible visualisation of global shipping traffic, created by Kiln.digital and the UCL Energy Institute.
www.shipmap.org
January 21, 2026 at 9:48 AM
Reposted by Michael Utz
One of the most exciting ways AI is changing the world is accelerating the velocity of absolute horseshit
January 6, 2026 at 9:30 PM
Reposted by Michael Utz
In the future, you won't even need to cook food. You'll just ask the AI and it'll describe the food to you
January 6, 2026 at 1:16 AM
Reposted by Michael Utz
I wonder if the big push toward requiring ID for all the centralized sites will be how we finally see people break out into their own spaces again? like, I'm not about to hand a tech company that's mandating all employees vibe code harder my government ID just so I can post jokes on the timeline
January 5, 2026 at 2:40 AM