Janosch Delcker

Ist „mind-reading“ AI das nächste große Ding? Wie lange sind unsere Gedanken noch frei?

Der Futurist und Journalist @janoschdelcker.bsky.social zeigte auf der re:publica Hamburg x Reeperbahn Festival, wie diese Technologie schon bald unser Leben durchdringen könnte.

▶️ youtu.be/as5KWHNgllc?...

The Authoritarian Stack How Tech Billionaires Are Building a Post-Democratic America — And Why Europe Is Next

«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

OpenAI Restructures to Become a More Traditional For-Profit Company

“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...

Vielen Dank für das Feedback und das Teilen!

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-...

Foto von Janosch Delcker mit einem Schriftzug mit seinem Namen. Darunter sind die republica Hamburg und Reeperbahnfestival Logos zu sehen.

Das Visual zeigt den Titel der Session "Mind-reading AI: Wie neue Technologien unser Denken entschlüsseln – und wir trotzdem die Kontrolle behalten" Darüber ist das Logo der re:publica Hamburg zu sehen. Darunter ist der Ort (Schmidt Theater) und das Datum (18.-19.9.2025) sowie das Logo der Förderpartner*innen zu sehen.

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...

SpaceX Gets Billions From the Government. It Gives Little to Nothing Back in Taxes.

“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...

“Geoff Hinton is a super genius … However, he is a genius of machine learning, not governance.”

Well-said!

'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.

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...

Answer written by Perplexity.AI chatbot: "Perplexity’s move is both opportunistic—capitalizing on legal pressure for divestiture—and a bold statement about the future of AI, where browser control translates into data dominance and product leverage. Even if unlikely to be immediately accepted, the bid reframes the browser market as the new battleground for AI-driven innovation, investment, and control. In short: Perplexity is publicly jockeying for a pivotal role in the post-antitrust, AI-centric internet landscape, while promising stability for users, continuity for advertisers, and competition for regulators. Whether Google actually sells Chrome or not, this bid signals a seismic shift in where the next phase of AI and web competition will play out."

To add to the confusion, I asked Perplexity's chatbot to explain the rationale behind Perplexity's offer to buy Chrome:

A.I. Start-Up Perplexity Offers to Buy Google’s Chrome Browser for $34.5 Billion

“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! 🏆

Some way to go, GPT-5.

Genuinely curious what you think:

Will this ban safeguard users as intended — or will it unintentionally harm those who have no access to mental health care?

From my (non-lawyer) understanding, this law now removes that option for people in Illinois.

Early research suggests that conversations with an AI chatbot *can* bring real mental health benefits – from supporting emotional well-being, to helping address specific conditions, to encouraging positive behavior change.

BUT here’s the reality: in many parts of the world, access to a qualified human mental health professional is a privilege enjoyed by only a fortunate few.

AI systems are not trained therapists, and they should never replace them. The intention – to protect vulnerable people from harmful or misleading advice by AI chatbots – is right and important.

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.

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.”

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-...

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.

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...

Meta’s New Superintelligence Lab Is Discussing Major A.I. Strategy Changes

“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...

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.

Okay, we—as in humanity—have inadvertently been broadcasting our existence to potential alien civilizations for decades.

gizmodo.com/aliens-can-d...

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.

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...

I work as an elder and hospice caregiver, which I am oddly passionate about. I can love on them shamelessly and no one complains about my codependence. It's a win-win situation. Many of my caregiving colleagues complain about the repetitive questions and, sometimes reactions, of the elderly, especially when those patients happen to read the newspaper, especially with the current downward spiral of our country — as if these people haven't lived through enough horror... I love my dementia peeps, but sometimes wonder if there's a Guinness Book World Record for how many times an hour a dementia patient can repeat the same question — it's got to be in the hundreds. At least with small children, they ask different questions. Dementia patients will get stuck on one short question and ask it until you can interrupt their train(carousel) of thought and successfully redirect their attention. That carousel is pretty manic sometimes. A couple days ago, I picked up my mail from the post office and drove over to the nursing home to take (let's call her "Miss Daisy") Miss Daisy out for a drive. Before we took off on our road trip, to the end of The Road and back, in our landlocked little town (Juneau, Alaska), I set my copy of The Onion down in front of her. Over our two-hour excursion, tiny Miss Daisy read that front page at least a dozen times and each time she would snicker, giggle, and guffaw, then put it down on the dash board and, a minute later, discover it anew. I think it was the best afternoon of my life. We don't often hear them laugh and when they do, it's the sweetest thing you've ever heard. ... Thank you, thank you, thank you, for the really important work all of you do. You make the world a better place.

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 End of Publishing as We Know It Inside Silicon Valley’s assault on the media

"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...

Meta Is Creating a New A.I. Lab to Pursue ‘Superintelligence’

“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...

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.

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...