Why do we do astrophysics?

2026/02/10 by David W. Hogg · 37 voices
#astro-ph.IM #physics.hist-ph

paper · pdf

Abstract

At time of writing, large language models (LLMs) are beginning to obtain the ability to design, execute, write up, and referee scientific projects on the data-science side of astrophysics. What implications does this have for our profession? In this white paper, I list - and argue for - a set of facts or "points of agreement" about what astrophysics is, or should be; these include considerations of novelty, people-centrism, trust, and (the lack of) clinical value. I then list and discuss every possible benefit that astrophysics can be seen as bringing to us, and to science, and to universities, and to the world; these include considerations of love, weaponry, and personal (and personnel) development. I conclude with a discussion of two possible (extreme and bad) policy recommendations related to the use of LLMs in astrophysics, dubbed "let-them-cook" and "ban-and-punish." I argue strongly against both of these; it is not going to be easy to develop or adopt good moderate policies.

Citations

• Quantifying large language model usage in scientific papers
• Sloan Digital Sky Survey-V: Pioneering Panoptic Spectroscopy
• The Astropy Project: Building an Open-science Project and Status of the v2.0 Core Package^*
• Sloan Digital Sky Survey IV: Mapping the Milky Way, Nearby Galaxies, and the Distant Universe
• emcee : The MCMC Hammer
• ASTROMETRY.NET: BLIND ASTROMETRIC CALIBRATION OF ARBITRARY ASTRONOMICAL IMAGES
• I.—COMPUTING MACHINERY AND INTELLIGENCE

Discussions

• David Hogg has written a white paper on doing astrophysics in the age of LLMs. It looks to be thought-provoking. My initial reaction is that either LLMs will destroy the field or they will force a rec [bsky, 137 points, 15 comments]
• This white paper about using LLMs in astronomy is making the rounds. It's very interesting, and while I can make some nitpicks, overall the point of the paper is to create a fundamental basis for tack [bsky, 76 points, 3 comments]
• Fun argument, from a paper on the meaning of astronomy in a world of powerful AI. arxiv.org/pdf/2602.10181 [bsky, 74 points, 12 comments]
• "Every person is a human being, whose personal development is more important than our short-term scientific accomplishments" - and more bangers in this from David Hogg (spoiler: it's about AI!) arxiv. [bsky, 61 points, 2 comments]
• Let me recommend grabbing a cup of tea & reading this really thoughtful piece from the inimitable @hogg.bsky.social on moderating AI/LLM use in astrophysics. 🔭 "the real question we face is not ... h [bsky, 46 points, 2 comments]
• Astronomers: highly recommend this thoughtful opinion piece by @hogg.bsky.social on how to think about our field in light of the development of large language models. whether you agree with him or not [bsky, 40 points, 2 comments]
• Kudos to @hogg.bsky.social for his new paper on arXiv that is stimulating a lot of discussion around the value of astrophysics and also the discourse around #LLMs. I think even non astro people will f [bsky, 14 points, 0 comments]
• This morning I'm reading a very cool white paper by David W. Hogg at NYU, a big proponent of using machine learning methods on space data, and he's tackling basically nothing less than the SOUL of AST [bsky, 7 points, 4 comments]
• 'I have heard it said... that an LLM can do some task “better than a graduate student.” That language makes me uncomfortable... Are graduate students in our groups and our laboratories and our univers [bsky, 5 points, 0 comments]
• Thanks Andrew. If you haven't already read it, you may find @hogg.bsky.social's article "Why do we do astrophysics?" interesting: arxiv.org/abs/2602.10181 as it informed my ideas about the relationshi [bsky, 4 points, 0 comments]
• This seems relevant to @hogg.bsky.social 's "Why do we do astrophysics" (arxiv.org/abs/2602.10181). My worry is that while the reasons we do astrophysics might not change, the job conditions might. [bsky, 4 points, 0 comments]
• Why do we do astrophysics? a nice review and thoughts by D Hogg arxiv.org/abs/2602.10181 [bsky, 4 points, 0 comments]
• Why do we do astrophysics? [hn, 3 points, 0 comments]
• cf. arxiv.org/abs/2602.10181 #esr #astrophysics #AI [bsky, 3 points, 0 comments]
• ghyle wenn im white paper über astrophysik und LLMs ACAB steht arxiv.org/pdf/2602.10181 [bsky, 3 points, 0 comments]
• I enjoyed reading this thoughtful position piece by @hogg.bsky.social on "why we do astrophysics" - a lot of these points can naturally be expanded to particle physics and any fundamental science arxi [bsky, 3 points, 0 comments]
• Why do we do astrophysics? [hn, 3 points, 0 comments]
• Why do we do astrophysics? [hn, 3 points, 0 comments]
• ざっくり読んだだけだけど、興味深かった。宇宙物理以外にもLLM時代の研究のあり方とか、そもそも研究するってどういうこと、みたいなところの話がなんか刺さるというか、現状の私のアクティビティを考えると心が痛いなあという感想 "Why do we do astrophysics?" arxiv.org/abs/2602.10181 [bsky, 2 points, 1 comments]
• Using them to replace the hard work of thinking, wrestling with ideas, and the practice of astronomy that leads to real mastery is a mistake. Hogg's paper on this is similarly insightful and gets to t [bsky, 2 points, 2 comments]
• blogging to arxiv is a massive flex https://arxiv.org/pdf/2602.10181 [bsky, 2 points, 3 comments]
• doi.org/10.48550/arX... [bsky, 2 points, 1 comments]
• I'm losing my mind over here ... please go read this white paper if you haven't yet (even if acrobat would like to save you the mental effort): arxiv.org/abs/2602.10181 [bsky, 2 points, 0 comments]
• i quit astrophysics 12 years ago, but one thing that will get me to read an astro-ph posting to the arxiv is "Author: @hogg.bsky.social" and "white paper". "Why do we do astrophysics?" does not disapp [bsky, 2 points, 1 comments]
• Going to assign this paper by @hogg.bsky.social to my Computational Physics students, who have expressed that they don't feel well-informed about AI/LLMs or the issues surrounding them. arxiv.org/pdf/ [bsky, 2 points, 0 comments]
• arxiv.org/abs/2602.10181 [bsky, 1 points, 0 comments]
• David Hogg wrote a thing (did not read yet, myself) arxiv.org/abs/2602.10181 [bsky, 1 points, 1 comments]
• arxiv.org/abs/2602.10181 Interesting white paper. [bsky, 1 points, 0 comments]
• tbh I think the argument in this is persuasive - we need to think about entirely new norms & processes or else this will make us all just have to be cops and we don't want to do that [bsky, 1 points, 0 comments]
• The issues raised should concern all of us in the sciences. A must-read. “…LLMs are beginning to obtain the ability to design, execute, write up, and referee scientific projects on the data-science si [bsky, 1 points, 0 comments]
• I’ve only read the introduction so far, but it hints at something important: astrophysics is collaborative; always has been. He gives the example of data scientists who took an astro class, but also i [bsky, 0 points, 1 comments]
• Nice! Did you see the similarly-veined essay from David Hogg, not maths but astronomy, a while ago? [bsky, 0 points, 0 comments]
• “Anyone working in #astrophysics is someone who wants to do astrophysics, not someone who wants to learn the answers,” argues David Hogg. If #AI tools can do much of the data science and mathematics t [bsky, 0 points, 0 comments]
• I work in astronomy and very little of that is verifiable/provable... so I'm a bit more sensitive to the failure modes of AI for science. And that doesn't even get into the motivations of *why* I migh [bsky, 0 points, 1 comments]
• arxiv.org/abs/2602.10181 [bsky, 0 points, 0 comments]
• Yet another thinkpiece about LLMs. Worth the read. "The real threat is a slow, comfortable drift toward not understanding what you're doing." arxiv.org/abs/2602.10181 [bsky, 0 points, 0 comments]
• Link: arxiv.org/abs/2602.10181 [bsky, 0 points, 0 comments]

Related

• What do astrophysics and the world's oldest profession have in common?
• What is the Role of Large Language Models in the Evolution of Astronomy Research?
• Realizing the potential of astrostatistics and astroinformatics
• Can LLMs Produce Original Astronomy Research in a Semester? A Graduate Class Experiment
• The Crisis in Astrophysics and Planetary Science: How Commercial Space and Program Design Principles will let us Escape