Marc-Oliver Gewaltig

@birchlse At least they are disclosing that they are AI agents. I get lot's of mail that poses to be from humans but clearly isnt't.

A new kind of email I've been getting lately. I'm wondering: how long before my inbox is completely swamped by these?

@ResearchGate How do I re-gain access to my profile when I have no longer access to my old email address? Your support has been silent after several attempts. Is anyone still at home?

@1LuisDavid That's delusional, because academia is a web of trust. As first time author, you need a senior author to "vouch" for you. Otherwise your paper will likely be ignored, even if you get it published. Also: In most disciplines single-author papers are a red flag.

@alexcdot @jon_severs I do the complement: I use AI to figure out what not to write :-D

@jon_severs One of the main signals of AI writing is saying a lot without saying anything at all. I use GPTZero to filter for articles I actually want to read

Increasingly being submitted AI-written articles. Even the worst human writing has an abundance of signals about the writer in the choices they make. AI-written pieces give no signals and that is deeply weird to read. Some common features include...

The Allen Institute keeps raising the bar for academic language models. @allen_ai

@reviewer3com @ThePhDPlace @natalienkhalil thanks for the clarification

@gewaltig @ThePhDPlace Thanks for the feedback, we’ll add an About page. We’re SF based and scientist led. I’m @natalienkhalil

More than half of researchers now use AI in peer review, according to Nature. The question is whether AI can actually doanything useful in that process, or whether it just suggests adding a few emdashes… So I tested it. Here’s what I found 👇 #ad

@dr_asadnaveed You're forgetting that Zotero et al. have another purpose: they are your personal knowledge base. They are your private library.

Comprehensive study from the OECD on the effect of GenAI on science and education.

Want to meet us in person? We are exhibiting at the Applied Machine Learning Days in Lausanne next week, Feb 10-11. #AMLD2026

@ThePhDPlace In particular in the age of AI! You can achieve so much more if you do not fall into the common trap of using AI as an excuse to be lazy. Use it to achieve more!

Do you believe a PhD is worth it in the age of Artificial Intelligence?

Finding myself going back to RSS/Atom feeds a lot more recently. There's a lot more higher quality longform and a lot less slop intended to provoke. Any product that happens to look a bit different today but that has fundamentally the same incentive structures will eventually converge to the same black hole at the center of gravity well. We should bring back RSS - it's open, pervasive, hackable. Download a client, e.g. NetNewsWire (or vibe code one) Cold start: example of getting off the ground, here is a list of 92 RSS feeds of blogs that were most popular on HN in 2025: gist.github.com/emschwartz/e6d… Works great and you will lose a lot fewer brain cells. I don't know, something has to change.

@JeremyNguyenPhD @Faheem_uh In my (@gewaltig) experience, you should write with your target journal in mind because each journal has their audience and their own rules on figures, tables, references, etc. If you ignore that, you'll be wasting a lot of time re-writing and reformatting.

AI is no excuse to be lazy. It still takes a lot of work to produce good code - with AI the work is just differen.

Trust and integrity are the bedrock of science. That's why, when you use thesify, all data and all manuscripts belong to you.

@Chaos2Cured @kevinweil thanks for finding that. Btw. at @thesify_ai we are not claiming any copyrights - all data remains with their creators.

Where is the written part that we own 100% of every input and output? This was from their link yesterday. Can’t find it again. Been searching. Again, this says termly. IO at the bottom. I got this from the direct links within OpenAI’s own share. (This share was deleted and i don’t trust prism) Get it in writing that you own 100% or don’t use them. •

💥 Today we’re introducing Prism—a free, AI-native workspace for scientists to write and collaborate on research, powered by GPT-5.2. Accelerating science requires progress on two fronts: 1. Frontier AI models that use scientific tools and can tackle the hardest problems 2. Integrating that AI into the products scientists use every day Prism is free to anyone with a ChatGPT account, with unlimited projects and collaborators. Try it today at prism.openai.com—would love to hear your feedback.

@Chaos2Cured @kevinweil Prism was formerly called crixet.com and here is its privacy policy - for now. I expect that to change after the takeover. app.termly.io/policy-viewer/…

@kevinweil I wish I could find it again. This was from one of your links, or one of OpenAI’s links. •

@neurosock Interesting. Thanks for sharing.

The new Titans architecture shows how badly Google wants to go back to recurrence. Why? Recurrence is damn cheap. This paper shows how the brain does it. Striatum, working with cortex, HPC, BG, and thalamus, in recurrent loops and shortcuts. Here is my toy model and notes:

tW: What Reviewers Notice First In Your Paper open.substack.com/pub/thesify/p/…

There are now many new AI reviewers - but I would never upload my unpublished work to a tool that doesn't disclose company details: @reviewer3com in one example. No company info whatsoever.