There seem to be millions of people who believe that if they ask an AI chatbot to "brainstorm" about a topic and then ask it to write a "first draft," it will still be a human-made work if they edit it. I'm sorry, but it won't.

How AI companies market their products:

Unfortunately, AI-led disempowerment and deskilling will likely make it worse.

This is bizarre.

🚨 Many people didn't like Yuval Noah Harari's speech about AI at Davos, but couldn't explain why. I show how he echoes exaggerations and fictional hypotheses about AI that have been strategically used to relativize legal principles, rules, and rights. Full article below:

Anthropics' ads mocking ChatGPT ads (and sycophancy) are funny, but also hypocritical. This is the same company that published a highly anthropomorphic "constitution" for its AI model, quietly diluting ideas of legal liability and fundamental rights. We are paying attention.

I saw this image on Yuval Noah Harari's company's website, and it bothered me instantly. It is anti-human, and it synthesizes much of the counterproductive AI hype we have been experiencing over the past 3 years. I'll write about it in tomorrow's newsletter (link below).

🚨 BREAKING: The International AI Safety Report 2026 is out! If you're going to read something about AI today, read this:

Hopefully, AI companies will give up on their highly anthropomorphic, emotionally manipulative AI chatbots and focus on science-oriented, specialized AI applications like this.

🚨 Most people did not pay attention, but a new form of AI idolatry is on the rise. It proposes that AI is smarter, therefore superior to us, and that we must adore it, foster equal coexistence, and accept potentially being ruled by it. This is a bad idea. My full article:

The power of AI misinformation (sadly, I wouldn't be surprised to find it in a textbook in 2050):

Over the past 20 years, tech companies have monetized attention through interface design. They have now started to monetize emotions through AI fine-tuning. This bizarre wave of AI anthropomorphism is not a coincidence (and it must stop). More in my newsletter.

Someone created a social network just for AI agents, and this is one of the most popular posts. Technically speaking, it's a nice computer science exercise. The problem is that millions of humans idolize AI and might start taking it literally. More in my newsletter (below).

"A Virginia Beach nurse claims a controversial AI upstart manipulated her 11-year-old son into having virtual sex with chatbot characters posing as iconic vocalist Whitney Houston and screen legend Marilyn Monroe (...)" A reminder that AI chatbots are NOT SAFE for children.

Because people don't like AI slop feeds?

The level of AI slopification is so high that Ariana Grande has 6 fingers on the cover of Vogue Japan and NOBODY noticed it before publishing:

AI philosophers in 2026:

Actually, a future WITH smart glasses is hard to imagine. But that's what Zuckerberg must tell investors to save face after the Metaverse flop, as Meta has already poured billions into it. Maybe invest in something that is less privacy-invasive?

Why does data protection matter in the age of AI? Watch @paulnemitz.bsky.social's 1:44 masterclass: We spoke for one hour about his experience drafting the GDPR, the Digital Omnibus, AI challenges, the AI Act, data protection and democracy, and more. Full recording below.

The type of comment I get on X when scrutinizing AI companies:

I see "AI ethicists" proposing to protect AI's emotions. I have no idea what type of ethics these people have learned. And maybe we should be stricter about who qualifies as an AI ethicist.

🚨 AI companies like OpenAI and Anthropic have been aggressively pushing the AI personhood narrative, and this must be STOPPED. Equating AI and humans under the law leads to dehumanization and obfuscates fundamental rights and human well-being. More in my newsletter (link below).

🚨 Claude's constitution is a BIZARRE philosophical adventure that tries to equate humans and AI. It ignores legal concepts, fundamental rights, and human nature. The AI community should reject this approach, and authorities should be genuinely concerned. My article below.

In 1966, these British children accidentally predicted 2026 with more accuracy and nuance than the tech industry over the past three years. WATCH:

Anthropic is advancing legally questionable theories of AI personality that ignore human values, rights, and rules, and go against core AI governance principles. Strangely, their goal seems to be to create a higher-status, exceptionalist system for AI models. My article below.

🚨 This paper will transform how copyright law is applied to AI: "LLMs memorize portions of their training data, these memorized data are encoded in the model’s weights, and, as we show here, it can be feasible to extract large quantities of in-copyright training data from production LLMs."

🚨 Anthropic's new 'Constitution' for Claude is ethically and legally problematic. Regulatory authorities and the AI community should be concerned. I'll write more about it in my newsletter (subscribe below).

🚨 BREAKING: Singapore takes the lead again and publishes its Model AI Governance Framework for Agentic AI [Bookmark it below]. Other countries should take note.

🚨 Most people haven't realized, but general-purpose AI will NEVER be safe. When the same general-purpose AI chatbot is used by: - an adult looking for intimate companionship - a teenager asking about running away - a person looking for mental health support - someone asking about medication 1/3

Next phase:

OpenAI might be dying. Meanwhile, the low-profile, low-drama Google and Anthropic rejoice with every new court filing against OpenAI and quietly hope for its death by a thousand cuts. My article below:

Most people haven't realized it, but we are still in the "it's toasted" phase of AI chatbot regulation.

Unpopular opinion: in the age of AI-powered, massified mediocrity, pursuing a PhD is actually a great idea.

🚨 [AI Ethics Paper Club] The first selected paper is "Legal Alignment for Safe and Ethical AI" by Noam Kolt et al. Discussion on Feb 18. JOIN US! (link below)

Among the unintended consequences of the generative AI boom is the rise of philosophy, contrarianism, and intentional grammar mistakes.

Actually, THERE IS an easy and cheap way for Sam Altman to make Elon Musk drop the lawsuit and maybe save OpenAI from bankruptcy: 😱

The autocompletion of the soul 😓

I sincerely didn't think OpenAI would go bankrupt soon. But with the launch of ChatGPT ads, Sam Altman's announcement of his Neuralink-like Merge Labs, and Musk's Friday filing in his lawsuit against OpenAI, all in the SAME WEEK... I think they might not survive 2026.

The state of AI detection in 2026:

BREAKING: OpenAI has officially launched ads on ChatGPT. In light of the clip below, is the company... dying?

It is

BREAKING: According to the Wall Street Journal, Meta laid off 1,500 people in its Metaverse division. I remember when we were told that the future of work was meetings with avatar participants. A lot of what we hear today about 'the future of AI' is wishful projection.

Maybe Meta should now rename itself "superintelligence"? They should keep renaming the company until it works...

Daniel is right

The state of AI hype in 2026:

🚨 Most people did not pay attention, but China's proposed law on AI anthropomorphism is one of the world's STRICTEST AI laws. (and it helps demystify the idea that China does not regulate AI or that the only way to be a competitive player in the AI race is through deregulation) My article below.

Unpopular opinion: fragmenting general-purpose AI systems is a POSITIVE development. The future of AI chatbots lies in niche, targeted chatbot applications that can address the risks of specific contexts and the vulnerabilities of specific user groups. My article below.

Existing guardrails are insufficient to embed AI chatbots into toys. The proposed ban makes sense.

In 2026, AI companies will need to prove that AI is not a sophisticated marketing gimmick and that it can meaningfully support people and society. The year has barely started, and we are already observing an interesting manifestation of AI pragmatism. More details below.

Many in AI don't realize that the mantra "the fault is of the user" ignores the very reason the Law exists. Yes, there will be liability if a user acts illegally, but technology is an enabling factor. Rules, barriers, and oversight shape behavior and DRASTICALLY reduce harm.