#AIForscience
3/ All 5 exhibits (QFT v2.0, 3-Body/Kepler, 3D Turbulence, Black Holes & StatMech) use ONLY the Python 3 standard library—zero floats/sqrt in Part I—and verify in isolated clean-room mode:

/usr/bin/python3 -I

Unredacted red-team failure logs inside! #Physics #MathSky #AIforScience
November 16, 2026 at 5:07 PM
Our two presenters for the #AIforScience webinar will be:
Yuhan Wang - Gen AI Hub and the University of Manchester.
Dylan John - AIchemy and Imperial College London
@aifunmcr.bsky.social | @aichemyhub.bsky.social | @imperialcollegeldn.bsky.social
Join us on 1 December at 2pm.
What can AI tell us about diseases? Can it help discover new materials faster?
Find out at our next webinar: AI for Science
📆1 December at 14:00 GMT.
Hear from researchers applying generative AI to challenges in biology and chemistry.
Sign up👇
www.genai.ac.uk/events/webin...
#AIforScience
November 18, 2026 at 10:32 AM
Jeff Dean on starting small: a mission of 10 people in Palo Alto, focused exclusively on automating science and engineering, cuts through the "slight distractions" of a larger organization. Full talk: https://www.youtube.com/watch?v=0kC3xOZChdA #AASF #JeffDean #AIforScience
November 18, 2026 at 9:59 PM
🤖 Efficient Models: Tencent’s AI excels with fewer active parameters.
🧪 AI-for-Science: Models now propose molecules and experiments.
🤝 Global Safety: US-China coordinate on AI risk.
#AI2026 #EfficientAI #AIForscience #AISafety
View in Timelines
November 17, 2026 at 3:01 PM
🚀 The HydroAgent-Lab Paper Digest just got an upgrade!

We're making it easier to discover, understand & keep track of research at the intersection of LLMs × Hydrology.

🔗 Explore it here: hydroagentlab.com/paper-digest/#…#Hydrologyg#LLML#AIforSciencec#AIAgentst#HydroAgentLaba#Watere#paperdigestst
November 19, 2026 at 9:55 AM
With one element and no internal reference, the authors caution, charging can mimic an oxidation-state shift. Our suggestion: store your charge reference per spectrum.

Saddiq et al., arXiv 2603.05350 (2026)

One study, synthetic spectra, uniform shifts.

#MLSky #AIForScience #research
November 18, 2026 at 12:34 PM
Overnight vs months in the lab: Kosmos (Edison Scientific) runs cycles of parallel analysis and literature agents that share one world model. Collaborators put a 20-cycle run at ~6 months of their research; 79.4% of checked statements held up. Details at toolome.bio #AIforScience #AIagents
November 14, 2026 at 4:28 PM
Misses at 90, 85 and 80%: the authors tie them to desorption, which the fitted curve has no term for. Our suggestion: test every target you will run.

Navabi & Takoudis, J. Vac. Sci. Technol. A 44, 032412 (2026)

Figures from one study.

#MLSky #AIforScience #research
November 18, 2026 at 2:45 PM
Our suggestion: if measured (not padded) points show no low-q plateau, barring aggregation and interparticle effects, give the largest dimension as at least roughly pi over your lowest measured q.

Roberts et al., Digital Discovery 4, 1467 (2025)

One study.

#MLSky #AIforScience #Research
November 18, 2026 at 7:21 AM
In Athavale et al.'s microring study, the optimiser targeted field medians, not ring-to-ring spread. Our suggestion: re-measure your reference with the new samples, several times.

Athavale et al., ACS Photonics 12 (2025)

Figures from one study's preprint.

#MLSky #AIforScience #research
November 17, 2026 at 12:53 PM
In Giunto et al.'s SEM-EDS pipeline a person names the elements; the method expects particles of a few micrometres or more.

Suggestion (ours, not the authors'): give the discard fraction with any accuracy figure.

Nat. Commun. 17, 9735 (2026)

Figures from one study.

#MLSky #AIforScience
November 17, 2026 at 11:21 AM
7/x We would be glad to hear about collaborations or use cases. Questions are most welcome here, or on other channels.

#OpenSource #AIforScience #NeuroAI
November 11, 2026 at 4:23 PM
🤖 Free master prompt: paste into ChatGPT / Claude / Gemini → your AI becomes a PubMed paper-chat assistant.

Grounded strictly in the abstract you paste. Grad students, postdocs, PIs welcome. Unlimited use in your own subscription.

https://sciencepubmed.net/en/lab/toolkit/

#AIforScience #PhDchat
November 15, 2026 at 7:53 PM
Claude read transcriptions of Kurt Gödel's 1,009 notebook pages in days and surfaced 663 theorems. An AI ocean model now forecasts straight from satellites and floats. #AIForScience #KurtGodel https://sharedsapience.com/century-report/the-century-report-november-14-2026/
November 14, 2026 at 9:12 PM
3/ Every script uses ONLY the Python standard library (zero sqrt, zero fft, zero floats in the Q-engine) and runs in isolated clean-room mode:

/usr/bin/python3 -I

Full prompts and unredacted red-team failure logs from Cycles 1-3 are in the repo. Try to break it! #Physics #MathSky #AIforScience
November 13, 2026 at 2:40 PM
Science's Last Exam is now open. I wrote one of its hydrology tasks. It tests whether an AI agent can do research on its own, given a goal, real data and real tools.

We still need tasks, especially in Earth sciences. Deadline Dec 31, 2026.
sciences-last-exam.org/contribute
#AIforScience
Contribute | Science’s Last Exam
Science's Last Exam is an open research initiative led by AItonomy, built with PhD students and postdoctoral researchers from 34 institutions.
sciences-last-exam.org
November 18, 2026 at 5:37 PM
In a 2026 olive-oil study, 15 to 20 pixels of a band the authors call noise sorted grades with over 80% accuracy; guessing the larger class scored 60–63%. The authors advise a no-signal test; so do we.

Michelucci and Venturini, Analyst (2026)

Figures from one study.

#MLSky #AIforScience
November 17, 2026 at 2:46 PM
🔬 AI proved the Aaronson-Ambainis quantum conjecture, open since 2008, in 3 days using a non-public Anthropic model. ✅ Lean verified every step. 🧠 AI-driven science now outpaces human comprehension. That is the singularity. #AIForScience #TheHallwayTrack
November 9, 2026 at 4:33 PM
Per the authors, denoising preserves separability; its gain: more readable cluster spectra. One sample; no accuracy spread given. Our suggestion, as the study did: test raw fast data too.

Martin-Calle et al., arXiv 2605.18511 (2026)

Preprint, not yet peer reviewed.

#MLSky #AIforScience
November 9, 2026 at 2:52 PM
In a 2025 AI-tuned microscopy study, people set the search ranges from experience; no quantitative expert comparison is reported. Our suggestion: decide what "satisfactory" means before the run.

Liu et al., ACS Nano 19, 19659 (2025)

Figures from one study.

#MLSky #AIforScience
November 9, 2026 at 12:17 PM
The authors tie under-counting to touching particles merging into one mask; they also report missed small particles. Our suggestion: check the count against a hand count.

Zhong et al., Nanomaterials 15, 1886 (2025)

Figures from one study.

#MLSky #AIforScience #research
November 9, 2026 at 10:15 AM
Anthropic destekli araştırmacılar bilimsel modellerin inşasını ve yorumlanmasını otomatikleştiren yapay zeka araçları geliştiriyor

www.airehber.com.tr/timeline/ant...
#airehber #Anthropic #AIforScience #YapayZeka #BilimselAraştırma
Anthropic destekli araştırmacılar bilimsel modellerin inşasını ve yorumlanmasını otomatikleştiren yapay zeka araçları geliştiriyor - Yapay Zeka Haberleri
ELLIS Institute Tübingen bünyesinde çalışmalarını sürdüren Anthropic "AI for Science" hibesi sahibi araştırmacılar Jakob Macke, Daniel Gedon ve Stefan Wahl,...
www.airehber.com.tr
November 8, 2026 at 1:31 AM
An AI research system on an unreleased Anthropic model proved an 18-year-old quantum conjecture by day three, checked by computer. METR says Anthropic's pace has doubled. #QuantumComputing #AIForScience https://sharedsapience.com/century-report/the-century-report-november-7-2026/
November 7, 2026 at 7:49 PM