I clearly feel that Opus 4.6 is worse at using Swedish than 4.5. (Not the conclusions when working in Swedish, just in using the language correctly.) Anyone else with similar/corresponding experience? #opus #anthropic

Watermarks are useless. Digital signatures are great. So obvious when you think about it. Read more here: falkai.substack.com/p/why-waterm...

My second blog post in a series about how AI affects knowledge, knowledge work and education. This one focuses on the necessity of using scenarios, and how to select them. 12 scenarios are consolidated into 3, for analysis in upcoming posts. falkai.substack.com/p/using-scen...

What just happened at the METR chart for AI ability to complete long tasks? Claude Opus 4.5 bumped from under five hours (50 percent pass) to 5:20. Also, GPT-5.1-codex-max was dropped. metr.org/blog/2025-03... #METR #opus #ai

Peter Sparber, a former architect of Big Tobacco’s successful war against regulation, says that the big AI companies are using the same playbook as Big Tobacco did. And that it works. aipodcastpicks.substack.com/p/why-ai-gov...

I just saw an example of Claude Code working with models not from Anthropic, by setting up a proxy server routing calls to elsewhere. Pretty smart. Anyone else seen something like that?

I can't help thinking that Claude Cowork in some sense is a step in the wrong for Anthropic. It's good, but a strength of Anthropic is their straight shot at coding and ignoring most of productification. If Claude Cowork becomes the same success as Claude Code, that might change. #ClaudeCowork

YES. More pages with this feature, please.

Statements like "The AI Act is killing AI innovation" made me frustrated enough to summarize what the AI Act actually says. It's reasonable, and not nearly as harsh as people think. Having a single regulation is something the US should envy. falkai.substack.com/p/the-eu-ai-...

One of a thousand ideas from the mind of Anders Bjarby, who is possibly the biggest AI explorer in Sweden. (Yes, the Lovable crew included.) www.youtube.com/watch?v=Y2cR... #badhairday #randomidea #justdoit #experimentandlearn

Great interview with Yoshua Bengio. Personal, factual and human. How should we balance the risks and benefits of AI, and how can we keep development at a sustainable speed? aipodcastpicks.substack.com/p/yoshua-ben...

Maybe we can use LLMs for solving parts of the coordination problem. An LLM can be a near-perfect trusted mediator, that don't leak information beyond what is specified on beforehand (such as "no deal is possible" or "there is common ground regarding X"). falkai.substack.com/p/llms-as-tr...

What does AI literacty actually mean? Here's a concrete answer. youtu.be/2PvMfhKfYdI

With the capable AI models we have today, I would argue that sycophancy now is a bigger problem than hallucinations. This is a short and clear video explaining this phenomenon, how you can get better at spotting it, and decrease the risk of being affected by it. youtu.be/nvbq39yVYRk?...

While we shouldn't take the results of GPT-5.2 on GDPval at face value, we really must start considering what continued AI progress means for our society. And we need to do that yesterday. falkai.substack.com/p/when-intel...

Here's a pitch for my new channel Graspable.AI, with short videos about AI for teachers world wide. I'd love to hear what you think. youtu.be/iHw-I3btmUA

Watch Amanda Askell chew through heavy philosophical questions as if they were breakfast cereal in this interview. I’m especially fond of how the entire recording radiates an ’80s vibe – including the image ratio made for TVs that aren't flat. aipodcastpicks.substack.com/p/amanda-ask...

Updates to my analysis on what a collapse of investments in AI might mean, with a fourth scenario: Microsoft, Apple, Amazon or Nvidia buys struggling AI startups. This is, I think, the most likely outcome, and Nvidia could be the real winner. More details here: falkai.substack.com/p/what-would...

I just published an analysis of what an AI investment crash would mean for technical development. The short version: Google, with its deep pockets, would likely end up quite alone in the AI race on the American side of the Atlantic. China continues regardless. falkai.substack.com/p/what-would...

California just introduced AI safety legislation that is, very briefly summarized, self-regulation with some edge to it. This podcast episode explains SB 53 in a very accessible way. aipodcastpicks.substack.com/p/a-quick-gu...

This is the best analysis I've seen/heard, of whether there's a financial AI bubble going on right now. Short story: Some warning signs, but not a bubble. aipodcastpicks.substack.com/p/great-anal...

Proud and humbled by appearing in The Cognitive Revolution – one of the best podcasts on AI I know. www.cognitiverevolution.ai/education-in... Thanks to @nathanlabenz.bsky.social for great discussions!

This interview with Estonia's Minister of Education is a must-listen for anyone working with strategic perspectives on education. The episode made me feel that maybe I should move to Estonia and start working full-time with AI and education again. aipodcastpicks.substack.com/p/estonias-m...

New videos about AI and education: Using AI for ideation. * Creating variations of an example * Brainstorming * Getting feedback on ideas * Exploring ideas in collaboration with AI * Creating artificial student panels/user panels * Creating bespoke software graspableai.substack.com/p/using-ai-f...

I revisited a blogpost I wrote summer 2024, just before I left my job with the goal of contributing to reducing societal-scale AI risks. It explains why I decided to leave a great job, without knowing what I should do. Still relevant, more than a year later. falkai.substack.com/p/what-drive...

Insight about AI, number 157: It is much easier to expand what you can do yourself using AI, than replacing what someone else is doing. Do you agree? What are potential implications?

Two videos for anyone working with software development (coding in particular) and haven't yet tried Claude Code. aipodcastpicks.substack.com/p/what-is-cl...

This podcast episode could actually be the best I've heard when it comes to AI and education. Multiple perspectives, and almost constant focus on the really deep impacts. Also quite fun to listen to. aipodcastpicks.substack.com/p/great-coll...

Superintelligent AI may be godlike, but not gods are created equal. What may superintelligent AI find difficult, and what type of problems will melt away? falkai.substack.com/p/the-hard-p... #ASI #AI #not-so-distant-future

This is the first time I hear of ChatGPT leading a person to suicide. I am truly appalled by some of the quotes recited in the episode. I've heard examples of Character AI bots doing this, but frankly didn't think normal chatbots could have this effect. aipodcastpicks.substack.com/p/a-disturbi...

250 words on what a 1.5 billion years old organism could teach us about how to navigate a potential future with superintelligent AI. falkai.substack.com/p/the-mitoch...

I stumbled into this video from Anthropic. Great resource for anyone who wants to understand more about the inside of large language models, and what "mechanistic interpretability" is. aipodcastpicks.substack.com/p/introducti...

I just published three more AI podcast picks. This one is worth listening to twice. How close are we to AGI? What happens after? And what does AGI mean for society, and for you and me as individuals? The episode doesn't give answers, but perspectives. aipodcastpicks.substack.com/p/benjamin-t...

The AI 20127 is well worth reading in full, but this 34 minute video is a summary that is *much* more accessible. www.youtube.com/watch?v=5KVD... (Worth knowing: Since writing AI 2027, the authors have pushed their timelines back by roughly a year. As is noted in the video.)

Great episode for anyone who wants to get a deeper understanding of the Trump administration's view of AI. aipodcastpicks.substack.com/p/what-does-...

Kudos to @nathanlabenz.bsky.social for delivering so many interesting podcast episodes on AI. I just reposted the best episodes I listened to last month, and three of the eight selected episodes were from The Cognitive Revolution. aipodcastpicks.substack.com

The best summary of what GPT-5 is and means I've seen so far. pca.st/zetp4njw

Recommended listening for anyone who wants to get an overview of China + AI policy. substack.com/@falkai/note...

Listening to the recently released episode of Last Week in AI. The episode doesn't particularly stand out, but the podcast as a whole is one of my top ways of staying up to date. Highly recommended. lastweekin.ai

Micro transactions by AI agents could be a really good thing. falkai.substack.com/p/how-ai-age... (Thanks to @nathanlabenz.bsky.social who in a recent episode of The Cognitive Revolution made me aware of the x402 protocol.)

Podcast pick! (On AI, as always.) falkai.substack.com/p/news-on-ai...

Solid Grok 4 overview linked below. My take, at high level: • Among the best LLMs now, but too early to call it *the* best • “Exponential scaling” might be working again after GPT-4.5/Llama 4 disappointments • Makes GPT-5 launch more likely soon pca.st/episode/2e6a...

Podcast pick for anyone who wants to keep up with what’s going on with AI regulation in the US. falkai.substack.com/p/news-on-ai...

"Early Tuesday morning, the United States Senate voted 99-1 to pass an amendment to the budget bill removing the proposed 10-year moratorium on the enforcement of state laws on artificial intelligence." www.techpolicy.press/us-senate-dr...

It took some time, but I finally found a way to think about "AI agents" that actually works. Maybe it works for you, too. falkai.substack.com/p/i-finally-...

Yay! My book "AI för nybörjare" is now available also in Norwegian! www.bokklubben.no/kunstig-inte...

Another AI podcast pick! Great explanation of reasoning models, test time scaling, why the AI scaling laws are inversely impressive, and what reasoning models mean for AI safety. falkai.substack.com/p/reasoning-...