Janosch Delcker
@janoschdelcker.bsky.social
AI demystifier, tech journalist, futurist
Previously: Chief Technology Correspondent, Deutsche Welle; AI
Correspondent, Politico; Columbia Journalism School.
http://www.janoschdelcker.com
Previously: Chief Technology Correspondent, Deutsche Welle; AI
Correspondent, Politico; Columbia Journalism School.
http://www.janoschdelcker.com
Pinned
Reposted by Janosch Delcker
«Silicon Valley isn’t building apps anymore.
It’s building empires.»
The Authoritarian Stack: A mapping of firms, funds, and political actors turning core state functions into private platforms, by @francescabria.bsky.social et. al. for @fesonline.bsky.social
www.authoritarian-stack.info
It’s building empires.»
The Authoritarian Stack: A mapping of firms, funds, and political actors turning core state functions into private platforms, by @francescabria.bsky.social et. al. for @fesonline.bsky.social
www.authoritarian-stack.info
The Authoritarian Stack
How Tech Billionaires Are Building a Post-Democratic America — And Why Europe Is Next
www.authoritarian-stack.info
November 2, 2025 at 10:00 AM
«Silicon Valley isn’t building apps anymore.
It’s building empires.»
The Authoritarian Stack: A mapping of firms, funds, and political actors turning core state functions into private platforms, by @francescabria.bsky.social et. al. for @fesonline.bsky.social
www.authoritarian-stack.info
It’s building empires.»
The Authoritarian Stack: A mapping of firms, funds, and political actors turning core state functions into private platforms, by @francescabria.bsky.social et. al. for @fesonline.bsky.social
www.authoritarian-stack.info
“Microsoft and OpenAI have now extended that agreement to 2032. The new contract provides Microsoft with access to technologies after OpenAI has reached A.G.I., provided that the technology is equipped with “appropriate safety guardrails.” www.nytimes.com/2025/10/28/t...
OpenAI Restructures to Become a More Traditional For-Profit Company
www.nytimes.com
October 28, 2025 at 4:06 PM
“Microsoft and OpenAI have now extended that agreement to 2032. The new contract provides Microsoft with access to technologies after OpenAI has reached A.G.I., provided that the technology is equipped with “appropriate safety guardrails.” www.nytimes.com/2025/10/28/t...
Reposted by Janosch Delcker
Die Insides von Janosch Delcker @janoschdelcker.bsky.social auf der @re-publica.com Hamburg haben mir Angst gemacht. Der Affe in meinem Kopf wollte die ganze Zeit dagegen ansingen. "Die Gedanken sind frei, wer kann sie erraten..." Ihr ahne es schon, die KI, die KI die kann sie erraten...
Wie entschlüsseln neue Technologien unser Denken – und wie behalten wir die Kontrolle? In seinem neuen Buch ruft Janosch Delcker (@janoschdelcker.bsky.social) dazu auf, unsere „Neurorechte“ zu schützen. Auf der #rpHH diskutiert er, was es dafür braucht.
🔗 re-publica.com/de/session/m...
🔗 re-publica.com/de/session/m...
September 22, 2025 at 3:03 PM
Die Insides von Janosch Delcker @janoschdelcker.bsky.social auf der @re-publica.com Hamburg haben mir Angst gemacht. Der Affe in meinem Kopf wollte die ganze Zeit dagegen ansingen. "Die Gedanken sind frei, wer kann sie erraten..." Ihr ahne es schon, die KI, die KI die kann sie erraten...
"Former Commissioner Thierry Breton, who helped draft the bloc’s digital rulebook, does not appear to have confirmed attendance as he is listed as an “invited witness”."
www.euractiv.com/news/former-...
www.euractiv.com/news/former-...
Former Commissioner Thierry Breton invited to MAGA DSA hearing | Euractiv
Hearing will also target the EU's Digital Markets Act and similar UK laws – with UK far right politician Nigel Farage listed as a confirmed witness
www.euractiv.com
September 15, 2025 at 8:29 AM
"Former Commissioner Thierry Breton, who helped draft the bloc’s digital rulebook, does not appear to have confirmed attendance as he is listed as an “invited witness”."
www.euractiv.com/news/former-...
www.euractiv.com/news/former-...
Reposted by Janosch Delcker
Wie entschlüsseln neue Technologien unser Denken – und wie behalten wir die Kontrolle? In seinem neuen Buch ruft Janosch Delcker (@janoschdelcker.bsky.social) dazu auf, unsere „Neurorechte“ zu schützen. Auf der #rpHH diskutiert er, was es dafür braucht.
🔗 re-publica.com/de/session/m...
🔗 re-publica.com/de/session/m...
August 26, 2025 at 2:06 PM
Wie entschlüsseln neue Technologien unser Denken – und wie behalten wir die Kontrolle? In seinem neuen Buch ruft Janosch Delcker (@janoschdelcker.bsky.social) dazu auf, unsere „Neurorechte“ zu schützen. Auf der #rpHH diskutiert er, was es dafür braucht.
🔗 re-publica.com/de/session/m...
🔗 re-publica.com/de/session/m...
“SpaceX has most likely paid little to no federal income taxes since its founding in 2002 and has privately told investors that it may never have to pay any, according to internal company documents reviewed by The New York Times.”
www.nytimes.com/2025/08/15/t...
www.nytimes.com/2025/08/15/t...
SpaceX Gets Billions From the Government. It Gives Little to Nothing Back in Taxes.
www.nytimes.com
August 15, 2025 at 12:40 PM
“SpaceX has most likely paid little to no federal income taxes since its founding in 2002 and has privately told investors that it may never have to pay any, according to internal company documents reviewed by The New York Times.”
www.nytimes.com/2025/08/15/t...
www.nytimes.com/2025/08/15/t...
“Geoff Hinton is a super genius … However, he is a genius of machine learning, not governance.”
Fully agree; said so here. joanna-bryson.blogspot.com/2024/06/what...
August 14, 2025 at 7:07 PM
“Geoff Hinton is a super genius … However, he is a genius of machine learning, not governance.”
Geoffrey Hinton is more than a household name in AI.
He knows his words shape public debate. And that seems to be his #1 mission now.
Proposing to embed “maternal instincts” in AI is anything but helpful.
fortune.com/2025/08/14/g...
He knows his words shape public debate. And that seems to be his #1 mission now.
Proposing to embed “maternal instincts” in AI is anything but helpful.
fortune.com/2025/08/14/g...
'Godfather of AI' says tech companies should imbue AI models with 'maternal instincts' to counter the technology’s goal to 'get more control'
Nobel laureate Geoffrey Hinton has previously likened AI to a “cute tiger cub” that is harmless now, but could easily turn on you.
fortune.com
August 14, 2025 at 6:36 PM
Geoffrey Hinton is more than a household name in AI.
He knows his words shape public debate. And that seems to be his #1 mission now.
Proposing to embed “maternal instincts” in AI is anything but helpful.
fortune.com/2025/08/14/g...
He knows his words shape public debate. And that seems to be his #1 mission now.
Proposing to embed “maternal instincts” in AI is anything but helpful.
fortune.com/2025/08/14/g...
To add to the confusion, I asked Perplexity's chatbot to explain the rationale behind Perplexity's offer to buy Chrome:
August 13, 2025 at 10:44 AM
To add to the confusion, I asked Perplexity's chatbot to explain the rationale behind Perplexity's offer to buy Chrome:
“the unsolicited bid is a long shot, since Perplexity itself is valued at an estimated $18 billion. Jesse Dwyer, a spokesman for the company, told The New York Times that outside investors had agreed to back a potential deal.” www.nytimes.com/2025/08/12/t...
A.I. Start-Up Perplexity Offers to Buy Google’s Chrome Browser for $34.5 Billion
www.nytimes.com
August 13, 2025 at 9:53 AM
“the unsolicited bid is a long shot, since Perplexity itself is valued at an estimated $18 billion. Jesse Dwyer, a spokesman for the company, told The New York Times that outside investors had agreed to back a potential deal.” www.nytimes.com/2025/08/12/t...
Congratulations, Petunia! 🏆
Petunia, a 2-year-old hairless English-French bulldog mix, won the World’s Ugliest Dog Contest on Friday. www.nytimes.com/2025/08/09/u...
August 10, 2025 at 8:24 AM
Congratulations, Petunia! 🏆
Some way to go, GPT-5.
August 9, 2025 at 10:35 AM
Some way to go, GPT-5.
The US state of Illinois has just banned AI systems from “providing mental health and therapeutic decision-making.” The wording is vague, but many interpret it as a ban on using AI tools like ChatGPT for therapy.
And I’m genuinely torn on how to feel about this.
And I’m genuinely torn on how to feel about this.
idfpr.illinois.gov
August 8, 2025 at 10:33 AM
The US state of Illinois has just banned AI systems from “providing mental health and therapeutic decision-making.” The wording is vague, but many interpret it as a ban on using AI tools like ChatGPT for therapy.
And I’m genuinely torn on how to feel about this.
And I’m genuinely torn on how to feel about this.
Reposted by Janosch Delcker
This matches with my understanding of where LLMs are likely to impact science quickest – bringing techniques that are well-understood to one set of experts, and applying them to questions raised by a different set of experts www.quantamagazine.org/ai-comes-up-...
AI Comes Up with Bizarre Physics Experiments. But They Work. | Quanta Magazine
Artificial intelligence software is designing novel experimental protocols that improve upon the work of human physicists, although the humans are still “doing a lot of baby-sitting.”
www.quantamagazine.org
July 22, 2025 at 8:57 AM
This matches with my understanding of where LLMs are likely to impact science quickest – bringing techniques that are well-understood to one set of experts, and applying them to questions raised by a different set of experts www.quantamagazine.org/ai-comes-up-...
Reposted by Janosch Delcker
Scoop from @eileenguo.bsky.social - huge amounts of personal information are embedded in a major data set used to train AI models www.technologyreview.com/2025/07/18/1...
A major AI training data set contains millions of examples of personal data
Personally identifiable information has been found in DataComp CommonPool, one of the largest open-source data sets used to train image generation models.
www.technologyreview.com
July 18, 2025 at 5:21 PM
Scoop from @eileenguo.bsky.social - huge amounts of personal information are embedded in a major data set used to train AI models www.technologyreview.com/2025/07/18/1...
“Many members of Mr. Wang’s team reported to Meta’s headquarters in Menlo Park, Calif., last week for the first time. The group is working in an office space siloed from the rest of the company and next to Mr. Zuckerberg.”
www.nytimes.com/2025/07/14/t...
www.nytimes.com/2025/07/14/t...
Meta’s New Superintelligence Lab Is Discussing Major A.I. Strategy Changes
www.nytimes.com
July 15, 2025 at 8:00 PM
“Many members of Mr. Wang’s team reported to Meta’s headquarters in Menlo Park, Calif., last week for the first time. The group is working in an office space siloed from the rest of the company and next to Mr. Zuckerberg.”
www.nytimes.com/2025/07/14/t...
www.nytimes.com/2025/07/14/t...
Okay, we—as in humanity—have inadvertently been broadcasting our existence to potential alien civilizations for decades.
gizmodo.com/aliens-can-d...
gizmodo.com/aliens-can-d...
Aliens Can Detect Earth’s Airports From 200 Light-Years Away
If intelligent life exists beyond Earth, these beings may already know where we live.
gizmodo.com
July 15, 2025 at 7:51 PM
Okay, we—as in humanity—have inadvertently been broadcasting our existence to potential alien civilizations for decades.
gizmodo.com/aliens-can-d...
gizmodo.com/aliens-can-d...
There's a simple yet effective trick for showing other insiders that you (and not AI) are still writing your LinkedIn posts.
And it involves my high school teacher, Ms. Curtis--who will probably never know.
www.linkedin.com/pulse/secret...
And it involves my high school teacher, Ms. Curtis--who will probably never know.
www.linkedin.com/pulse/secret...
The Secret Signal That Says: I Wrote This LinkedIn Post, Not AI
There's a subtle trick to signal that you’re still writing your own LinkedIn posts rather than using AI. Only insiders will pick up on it.
www.linkedin.com
July 12, 2025 at 12:55 PM
There's a simple yet effective trick for showing other insiders that you (and not AI) are still writing your LinkedIn posts.
And it involves my high school teacher, Ms. Curtis--who will probably never know.
www.linkedin.com/pulse/secret...
And it involves my high school teacher, Ms. Curtis--who will probably never know.
www.linkedin.com/pulse/secret...
Reposted by Janosch Delcker
I got permission to share this, and I'm extremely grateful for that.
The Onion got this letter from one of our subscribers in Alaska. She works with dementia patients and decided to leave a copy in the car for each one.
This email made my year. Read it and you'll see what I mean. People are good.
The Onion got this letter from one of our subscribers in Alaska. She works with dementia patients and decided to leave a copy in the car for each one.
This email made my year. Read it and you'll see what I mean. People are good.
July 1, 2025 at 6:46 PM
I got permission to share this, and I'm extremely grateful for that.
The Onion got this letter from one of our subscribers in Alaska. She works with dementia patients and decided to leave a copy in the car for each one.
This email made my year. Read it and you'll see what I mean. People are good.
The Onion got this letter from one of our subscribers in Alaska. She works with dementia patients and decided to leave a copy in the car for each one.
This email made my year. Read it and you'll see what I mean. People are good.
"Books have sometimes been licensed for only a couple hundred dollars each, and that a publisher that asks too much may be turned down, only for tech companies to take their material anyway."
www.theatlantic.com/technology/a...
www.theatlantic.com/technology/a...
The End of Publishing as We Know It
Inside Silicon Valley’s assault on the media
www.theatlantic.com
July 2, 2025 at 7:11 AM
"Books have sometimes been licensed for only a couple hundred dollars each, and that a publisher that asks too much may be turned down, only for tech companies to take their material anyway."
www.theatlantic.com/technology/a...
www.theatlantic.com/technology/a...
“Meta has tapped Alexandr Wang, 28, the founder and chief executive of the A.I. start-up Scale AI, to join the new lab…and has been in talks to invest billions of dollars in his company as part of a deal that would also bring other Scale AI employees to the company.”’ www.nytimes.com/2025/06/10/t...
Meta Is Creating a New A.I. Lab to Pursue ‘Superintelligence’
www.nytimes.com
June 10, 2025 at 10:53 AM
“Meta has tapped Alexandr Wang, 28, the founder and chief executive of the A.I. start-up Scale AI, to join the new lab…and has been in talks to invest billions of dollars in his company as part of a deal that would also bring other Scale AI employees to the company.”’ www.nytimes.com/2025/06/10/t...
For all our advances, we've barely scratched the surface of what's really going on out there in space.
www.cbsnews.com/amp/news/mas...
www.cbsnews.com/amp/news/mas...
Massive planet discovered orbiting tiny star, leaving scientists stumped
It had not been thought possible that such tiny, weak stars could provide the conditions needed to form and host huge planets.
www.cbsnews.com
June 5, 2025 at 10:25 AM
For all our advances, we've barely scratched the surface of what's really going on out there in space.
www.cbsnews.com/amp/news/mas...
www.cbsnews.com/amp/news/mas...
Meta announces partnership with defense contractor Anduril Industries to develop military technology, including an AI-powered helmet.
Given the basic AI research conducted within Meta over the last years, this expansion into defense makes a lot of sense.
www.washingtonpost.com/technology/2...
Given the basic AI research conducted within Meta over the last years, this expansion into defense makes a lot of sense.
www.washingtonpost.com/technology/2...
Meta is working on a high-tech helmet for the U.S. military
Social network company Meta is partnering with weapons-maker Anduril to develop military technology for the Pentagon using AI and augmented reality.
www.washingtonpost.com
May 30, 2025 at 10:52 AM
Meta announces partnership with defense contractor Anduril Industries to develop military technology, including an AI-powered helmet.
Given the basic AI research conducted within Meta over the last years, this expansion into defense makes a lot of sense.
www.washingtonpost.com/technology/2...
Given the basic AI research conducted within Meta over the last years, this expansion into defense makes a lot of sense.
www.washingtonpost.com/technology/2...
“The push has put a key Palantir product called Foundry into at least four federal agencies, including D.H.S. and the Health and Human Services Department.”
www.nytimes.com/2025/05/30/t...
www.nytimes.com/2025/05/30/t...
Trump Taps Palantir to Compile Data on Americans
www.nytimes.com
May 30, 2025 at 9:48 AM
“The push has put a key Palantir product called Foundry into at least four federal agencies, including D.H.S. and the Health and Human Services Department.”
www.nytimes.com/2025/05/30/t...
www.nytimes.com/2025/05/30/t...
Major disruptions often occur when technologies from different fields converge.
Worth keeping an eye on this emerging convergence of scientific computing and AI:
www.nytimes.com/2025/05/29/t...
Worth keeping an eye on this emerging convergence of scientific computing and AI:
www.nytimes.com/2025/05/29/t...
May 30, 2025 at 9:14 AM
Major disruptions often occur when technologies from different fields converge.
Worth keeping an eye on this emerging convergence of scientific computing and AI:
www.nytimes.com/2025/05/29/t...
Worth keeping an eye on this emerging convergence of scientific computing and AI:
www.nytimes.com/2025/05/29/t...