Maarten J. Bijlsma

Does he make bold statements for populist reasons (appearance of boldness), or is he actually bold and his words indicate what actions will soon follow?

OpenAI says it has evidence China’s DeepSeek used its model to train competitor White House AI tsar David Sacks raises possibility of alleged intellectual property theft

OpenAI says it has found evidence that Chinese artificial intelligence start-up DeepSeek used the US company’s proprietary models to train its own open-source competitor, as concerns grow over a potential breach of intellectual property. The San-Francisco-based ChatGPT maker told the Financial Times it had seen some evidence of ‘distillation’, a technique used by developers to obtain better performance on smaller models by using outputs from larger, more capable models. This allows them to achieve similar results on specific tasks at a much lower cost. OpenAI declined to comment further on details of its evidence. Its terms of service state users cannot “copy” any of its services or “use output to develop models that compete with OpenAI”.

I'm so sorry I can't stop laughing. OpenAI, the company built on stealing literally the entire internet, is crying because DeepSeek may have trained on the outputs from ChatGPT. They're crying their eyes out. What a bunch of hypocritical little babies. Cry more, freaks.

www.ft.com/content/a0df...

Unsolicited advice to authors:
In your responses to reviewers, the best practice is to respond point by point, using the same numbering and even including the reviewers' comments verbatim directly before each response.
Do not remove the numbering, jumble the order, and paraphrase the comments!

EUROPE IS DOING FINE. Look at these trade figures from Hélène Rey (London Business School and CEPR) which she presented in her prize talk at the 3rd Kiel-CEPR Conference on Geoeconomics this past autumn. In an equalising world, CN = US < EU

Video of talk is here www.ifw-kiel.de/institute/ev...

People, just stop doing Frontiers special issues. Find a nice diamond open access journal, and self-publish and collaboratively peer review the set of papers you all want to write. Do not give Frontiers money. They add no value. If anything, they hurt your reputation. MDPI is of course even worse.

Very nice view, but please be aware thay if anyone ever wanted to dox you this'll be quite helpful to them.

Four year postdoc position at the Oslo Centre for Biostatistics and Epidemiology (OCBE) (272144) | University of Oslo Job title: Four year postdoc position at the Oslo Centre for Biostatistics and Epidemiology (OCBE) (272144), Employer: University of Oslo, Deadline: Friday, January 31, 2025

We're still seeking candidates interested in causal inference and/or survival analysis for a postdoc position in Oslo. Deadline January 31.

Please apply or forward this to anyone who might be interested!

www.jobbnorge.no/en/available...

I don't want to fucking scan a qr code to order food. give me an actual physical menu or im leaving

Good to hear this is being pursued. If Europe wants strategic autonomy it needs to use its own infrastructure.

I wonder if people like Mark Zuckerberg and Jeff Bezos use AI themselves or if they know its dangers and steer clear, considering it only for their underlings.

Kind of like a drug dealer: you don't get high on your own supply.

My letter in today's Guardian. Just asking, like. But I think we know the answer.

I wish cars would just do what I expect and follow traffic patterns. No, I'm not going to turn left across the street in front of you while you are unexpectedly stopped and waving for me to bike across and I can see a line of cars building up behind you getting grumpier and grumpier.

A conversation on treatment effects The trial statistician and the clinical investigator took a step back to admire their creation.

Two angels discussing the basics of treatment effects.

(ICYMI)

statsepi.substack.com/p/a-conversa...

HPV vaccine stops 90% of cervical cancer cases England began vaccinating teenage girls in 2008 and results show it is paying off.

Vaccines are amazing.

www.bbc.com/news/article...

Just saw Heretic. Was about 2/3rds of a movie in terms of storyline. Not much character development. The acting was great, though.

You shouldn't install glass windows in your house because someone can just break the glass.

there's no possible or even imaginable non-corrupt use case here

Russische Sabotage in Europa: Wie Russland einen hybriden Krieg in Europa führt Russland bekämpft europäische Staaten seit Langem mit Desinformation, inzwischen aber auch mit Sabotage und Anschlägen. Wann und wo es Attacken gab

Extremely useful list by @zeit.de on (suspected) cases of Russian sabotage, espionage and more in Germany and Europe. It’s a LONG list!

www.zeit.de/politik/ausl...

Sometimes Siri would be inadvertently activated, a whistleblower told The Guardian, when an Apple Watch was raised and speech was detected. The only clue that users seemingly had of Siri's alleged spying was eerily accurate targeted ads that appeared after they had just been talking about specific items like Air Jordans or brands like Olive Garden, Reuters noted.

You all said I was being paranoid and it was just coincidence.

arstechnica.com/tech-policy/...

Een oud basisschool docent van me heette Dolf. Vlak na WOII geboren. Gegeven ook zijn didaktische stijl hadden wij als kinderen zo onze verdenkingen.

Researcher: "We let the data speak for itself."

Earlier that day:

One of the top journals in my field, a really dark day... @elsevierconnect.bsky.social is a cancer on academic publishing!

Harvard Professors Leaving for Private Sector Research | Harvard Magazine Three distinguished scientists on leaving academia to advance biomedical research

“A donor wanted to give $2 million to my lab, but only wanted to allow a maximum 10 percent to go to indirect cost recovery. And the school declined to receive the $2 million because they wanted at minimum, I believe it was 20 percent.”

www.harvardmagazine.com/2025/01/harv...

a countdown clock with the number 10 in the center ALT: a countdown clock with the number 10 in the center

Let us start 2025 in a positive mood: here are 10 methods things researchers can worry *less* about in 2025

Yes AND I personally would like those academics to also engage with the ways in which LLMs are "plagiarism machines" [1], and to reflect on their attitudes towards plagiarism in scholarly work, the classroom, and the academy generally.

[1] www.chronicle.com/article/chat...

*sighs*

Percentage change should *always* be expressed on a log scale, because it’s multiplicative.

A doubling and a halving should be given equal space.

I did a whole thread on this at the time x.com/jburnmurdoch...

Probably not helpful for this current paper though; if applied psychologists are like demographers, they generally don't see the point of formalizing counterfactuals and it'll just confuse things.

Ah, those type of papers are helpful / necessary. While I'm generally in favour of avoiding APC, I know you are an expert on causal inference. Perhaps you'll find this paper of interest since as far as I know it's the only paper describing APC with counterfactual notation doi.org/10.1007/s135...

What's the gist of your talk? A general caution or a substantive interest of some sort?

For anyone who reads this thread and was as confused as I was; this is a post about absence of evidence not being evidence of absence.