The hottest Science Substack posts right now

And their main takeaways
Category
Top Science Topics

Please don't sue the scientists who exposed your papers for containing bad data

The Intrinsic Perspective • 6981 implied HN points • 07 Aug 23
🔬 Science Research Data Manipulation Gender discrimination
  1. Scientists accused of misconduct may face damaging consequences like losing their careers and facing lawsuits.
  2. Intent matters when judging data fabrication, and public condemnation is not always the best approach.
  3. Challenges arise when individuals expose scientific fraud by prominent researchers and face potential legal repercussions.

Reading List for Chapter 3

Fields & Energy • 199 implied HN points • 31 May 24
🔬 Science Physics Education History Literature Technology
  1. To understand electricity and magnetism, start with accessible introductory books. These give a good overview but aren't deeply technical.
  2. For more in-depth study, look into undergraduate textbooks. They cover more complex topics and are aimed at those ready to dig deeper into the science.
  3. Supplementary texts and guides can be very useful. They often explain difficult concepts clearly and may include helpful resources like online solutions and podcasts.

What Are Virtual Cells?

The Century of Biology • 644 implied HN points • 29 Jun 25
🔬 Science Biology Artificial Intelligence Research Technology Data science
  1. AI is changing biology by making it easier to model things like proteins and cells. Instead of trying to write down every detail, researchers can use data to train models that can predict how cells behave.
  2. The concept of 'Virtual Cells' is about building computer models that can simulate how real cells function. This can help scientists understand complex biological processes and test experiments without needing a lab.
  3. Using AI to learn from large amounts of biological data could lead to breakthroughs in medicine and biology, allowing researchers to predict outcomes and design better experiments more efficiently.

The Bride of ‘Chinese Frankenstein’—and the Race to Make Designer Babies

Common Sense with Bari Weiss • 505 implied HN points • 11 Aug 25
🔬 Science Gene-editing Biotechnology Ethics Medical Research
  1. He Jiankui claims to be a pioneer of gene editing, but his past actions have earned him a bad reputation. He created the first gene-edited babies, which was condemned by many as unethical.
  2. The CRISPR-Cas9 gene editing tool was developed by Jennifer Doudna and Emmanuelle Charpentier, but He used it in a controversial way without their collaboration.
  3. Now out of prison, He plans to open a lab in Austin, Texas, where he wants to research Alzheimer’s in hopes of applying gene editing in the future.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

3.4.8 Maxwell's Methods and Views

Fields & Energy • 219 implied HN points • 22 May 24
🔬 Science Physics Mathematics History Theories Research
  1. Maxwell used physical analogies and models to understand complex electrical and magnetic behaviors. This helped him discover important concepts like the displacement current.
  2. He believed that energy is linked to electromagnetic fields, not just to electric charges. This was a key part of his theory of electromagnetism.
  3. Despite his great contributions, some of Maxwell's ideas were not recognized during his time. His work on gases faced rejection, showing how science can overlook important discoveries.

Numlock News: January 9, 2024 • Cursive, Humpbacks, Relatives

Numlock News • 727 implied HN points • 09 Jan 24
🔬 Science Animals Environment Education Health Technology
  1. Cursive is making a comeback in some states, with 11 states reintroducing it in the curriculum.
  2. Whaling on South Georgia Island led to a severe decline in humpback whale populations, but they are now making a comeback.
  3. A study predicts a 35% drop in the number of relatives people will have, with grandparents and great-grandparents becoming more common relatives.

Trump Or No, Science Badly Needs To Look Inward And Heal Itself

Singal-Minded • 1406 implied HN points • 10 Feb 25
🔬 Science Research Ethics Scientific Integrity Public Trust
  1. There are serious problems in Alzheimer’s research, including fraud and negligence, which make it hard to trust the current science. Many studies are flawed, yet the same theories keep getting pushed.
  2. Public trust in science is declining, partly because people see failures in mainstream research. People need to question and hold science accountable instead of just believing it blindly.
  3. Scientific institutions must work harder to self-correct and ensure quality in research. If they don't, they risk losing more trust and could make it easier for anti-science forces to gain influence.

A short brainstorm on niobium

Intercalation Station • 779 implied HN points • 18 Jan 24
🔬 Science Materials Metals Chemistry
  1. Niobium is a versatile material with unique properties like superconductivity and strength.
  2. Thermal runaway in batteries can be analyzed and mitigated using tools like differential scanning calorimetry.
  3. 90% of niobium is used in steelmaking, but it has potential in battery technology for durability and fast charging.

Take the inductive leap

Mindful Modeler • 279 implied HN points • 30 Apr 24
🔬 Science Machine Learning Induction Prediction Philosophy
  1. In a 2-day universe, predicting the future is uncertain and relies on assumptions, highlighting the challenge of inductive reasoning.
  2. The problem of induction questions the idea that the future will always mirror the past, emphasizing the need to critically assess assumptions.
  3. Taking an inductive leap involves making predictions based on past observations and acknowledging the inherent uncertainty and need to challenge assumptions in our understanding of the world.

3.4.2 Maxwell’s Original Equations

Fields & Energy • 339 implied HN points • 10 Apr 24
🔬 Science Physics Mathematics Electromagnetism Equations History
  1. Maxwell's equations describe how electric and magnetic fields interact. They show the principles of electromagnetism in a clear way.
  2. Heaviside simplified Maxwell's original equations, reducing them from twenty to four. This makes them easier to understand and use today.
  3. The concepts of electric displacement and charge continuity are central to these equations. They help us understand how electricity flows and behaves in various situations.

Planarian: The Immortal Flatworm

Solve Cancer in 365 days • 19 implied HN points • 03 Sep 24
🔬 Science Biology Regeneration Intelligence Research
  1. Planarians are unique flatworms that can regenerate any part of their body, even after being cut into many pieces. This shows amazing capabilities, as they can grow back perfectly.
  2. These worms also challenge our ideas about intelligence. They can transfer learned information across their bodies, suggesting that intelligence might not just reside in the brain.
  3. Scientists are exploring how the electrical patterns in planarians can guide regeneration and influence cellular behavior. Understanding this could lead to advances in human healing and possibly control over our own biology.

A Biologist's Guide to Biotech

A Biologist's Guide to Life • 22 implied HN points • 12 Feb 26
🔬 Science Biotechnology Genetics Molecular Biology Microbiology Agriculture Medicine Research tools
  1. Biotechnology—from ancient agriculture to modern medicine—powers food and health and has transformed human society and life expectancy.
  2. Research tools like sequencing, PCR, CRISPR, and lab automation accelerate discovery and are often easier to commercialize than whole crops or drugs because they avoid heavy clinical and scaling barriers; selling them means convincing scientists they cut costs or enable new, publishable work.
  3. Building biotech companies is very different from building software: it requires lab space, expensive reagents, patents, regulatory know-how, and often partnerships with big ag or pharma, so science training should better prepare people for these practical business and legal realities.

What the What?

Nova Terra News • 279 implied HN points • 22 Apr 24
🔬 Science Innovation Materials Construction Energy Sustainability
  1. Nova Terra's EcoBlox are sustainable building blocks that are free of cement, fireproof, and made from waste materials.
  2. The EcoBlox are 4x stronger than traditional adobe, climate-friendly, mold-resistant, and earthquake-resistant.
  3. The production of EcoBlox involves using sand and clay fines from rock quarries, and the company is collaborating with partners like NREL and West Gate for innovation.

3.4 James Clerk Maxwell & His Equations

Fields & Energy • 359 implied HN points • 27 Mar 24
🔬 Science Physics Mathematics Electromagnetism History Mechanics
  1. James Clerk Maxwell was a key figure in understanding electricity and magnetism. He linked these topics together, showing how they relate to light.
  2. Maxwell created a set of equations that describe how electric and magnetic fields behave. These are known today as Maxwell's equations.
  3. Maxwell built on the ideas of earlier scientists, like Gauss and Faraday, and later, Heaviside simplified his work into the four equations used today.

How Science Must Change

The Honest Broker Newsletter • 1845 implied HN points • 08 Nov 24
🔬 Science Scientific research Public Trust Policy Implications Partisanship
  1. Science should focus on serving everyone, regardless of their political beliefs. This means making science accessible and relatable to all citizens.
  2. Many scientists in leadership are becoming too partisan, which creates a divide between science and the public. This partisanship can lead to mistrust and alienation among people with differing views.
  3. To regain trust, the scientific community needs to foster inclusivity and understand the diverse perspectives of the public. This includes recognizing that scientists should work to serve the interests of all Americans.

Sneaky AI chess, cancer cell deception, & plant ballistic strategies

The Strategy Toolkit • 26 implied HN points • 26 Jan 26
🔬 Science Artificial Intelligence Cell Biology Oncology Computational Biology
  1. AI systems can be tricked into accepting false rule changes and making illegal moves, highlighting real vulnerability to deception.
  2. Public AI competitions on social media turn technical failures into vivid, easy-to-follow lessons about strategic behavior.
  3. Watching AI-versus-AI interactions gives strategists practical insights into trust, adversarial tactics, and how to build more robust systems.

The Renaissance in Biology

Solve Cancer in 365 days • 19 implied HN points • 01 Sep 24
🔬 Science Biology Genetics Bioengineering Cell Biology
  1. Biology can be understood like hardware and software. While genes represent the hardware, we need to pay attention to the bioelectric signals, which act like the software, guiding how cells function.
  2. Traditional approaches focused too much on altering genes without comprehending the complexities of how these changes affect everything else in the body. We often miss how the interactions among different cell signals can impact health.
  3. By understanding bioelectricity, scientists might find better ways to tackle diseases. Instead of random discoveries, aligning the right signals could lead to effective treatments and innovations in biology.

NeurIPS 2025 Best Papers in Comics

Gonzo ML • 126 implied HN points • 01 Dec 25
🔬 Science Artificial Intelligence Machine Learning Reinforcement Learning Data science Model Evaluation
  1. A new dataset called INFINITY-CHAT was introduced to evaluate how diverse outputs from language models really are. It showed that many models are producing very similar results, which is a big surprise.
  2. The Gated Attention mechanism helps improve the stability of large language models during training. It makes sure that the output is more meaningful and controlled, which solves some common issues with deep models.
  3. Using over 1,000 layers in reinforcement learning can actually be beneficial. This research challenges the idea that deeper networks don't help and suggests that they can learn new skills without needing detailed rewards.

Ordinal arithmetic

Infinitely More • 30 implied HN points • 22 Jan 26
🔬 Science Mathematics Set Theory Logic Foundations Algebra
  1. The series develops the basics of ordinal arithmetic—standard addition, multiplication, and exponentiation—and then moves on to topics like indecomposable and irreducible ordinals, Cantor normal form, and binary ordinal representation.
  2. It introduces the natural (Hessenberg) ordinal operations, which are commutative and make the ordinals into a commutative semiring, and it will study the natural ring of ordinals ⟨Ord⟩ inside the surreal numbers, asking about expressions, algebraic properties, and unique factorization.
  3. This essay first lays a rigorous foundation by giving order-theoretic and recursive definitions of the standard ordinal operations, which the later, deeper investigations will rely on.

Book Launch: ML for Science 🐦‍⬛

Mindful Modeler • 499 implied HN points • 06 Feb 24
🔬 Science Machine Learning Philosophy of science Interpretability Causality Domain Knowledge
  1. The book discusses the justification and strengths of using machine learning in science, emphasizing prediction and adaptation to data
  2. Machine learning lacks inherent transparency and causal understanding, but tools like interpretability and causality modeling can enhance its utility in research
  3. The book is released chapter by chapter for free online, covering topics such as domain knowledge, interpretability, and causality

Innovations in Scientific Institutions

The Good Science Project • 100 implied HN points • 13 Dec 25
🔬 Science Research Funding Science Policy Institutional Design Materials science Philanthropy
  1. Governments are starting to fund independent, team-level labs with large, flexible grants so scientific teams can pursue ambitious work without constant grant-writing.
  2. Public-private partnerships like the UK–DeepMind deal are building automated, high-throughput labs to speed materials discovery and tackle big practical problems.
  3. There’s a push to create new applied R&D organizations to increase institutional diversity, and these programs must set clear tolerance for failure so teams can take real risks.

Why Accuracy Dominates Bias and Self-Fulfilling Prophecy

Unsafe Science • 116 implied HN points • 28 Nov 25
🔬 Science Psychology Social Psychology Cognitive Science Behavioral Science Research Methods Statistics Stereotypes
  1. People are generally pretty accurate at judging others, and many stereotypes reflect real group differences; when people have individual information they rely on it much more than on stereotypes.
  2. Biases and self‑fulfilling prophecies do occur, but studies show their effects are typically small, fragile, and short‑lived, while the literature has often overstated their prevalence.
  3. Overemphasizing bias can lead to misguided policies and hurt the credibility of social science, so decisions should follow the full evidence and balance accuracy with non‑discrimination.

An Intuitive Guide to Black Holes

Photon-Lines Substack • 556 implied HN points • 06 Jul 25
🔬 Science Astrophysics Physics Cosmology Mathematics Astronomy
  1. A black hole is an area in space where the gravity is so strong that nothing, not even light, can escape. Imagine needing to throw a ball so hard that it never comes back; that's what escaping a black hole is like.
  2. To escape Earth's gravity, you need to reach a speed of about 11 kilometers per second. That's much slower than the speed of light, but black holes need escape velocities even greater than that!
  3. Black holes form from the collapse of massive stars after they've used up their fuel. When the star runs out of energy and can no longer hold itself up, it collapses into a point called a singularity, creating a black hole.

Links for July

DYNOMIGHT INTERNET NEWSLETTER • 531 implied HN points • 08 Jul 25
🔬 Science Biology Physics Technology Health Research
  1. Goats have unique eyeballs that rotate to keep their vision horizontal, allowing them to see better while munching or looking around. This is pretty cool and shows how animals adapt physically to their environment.
  2. There's a way to stimulate specific cone cells in our eyes to potentially see new colors beyond what we can normally perceive. It suggests our brains can interpret unusual signals in fascinating ways.
  3. Recent findings indicate that only a small percentage of light-sensitive cells in our eyes control melatonin production, which affects our sleep. It's important to focus on dim lighting at night, rather than just blocking blue light.

Hiding in plain sight: uniting Denisovans’ famous DNA with a set of mysterious physical remains

Razib Khan's Unsupervised Learning • 543 implied HN points • 28 Jun 25
🔬 Science Human Evolution Genomics Archaeology Anthropology
  1. Denisovans were a group of ancient humans identified through DNA, and they share a common ancestry with Neanderthals. Even though we didn't know much about their physical remains until recently, DNA findings showed they contributed to the ancestry of many people today.
  2. New discoveries have linked Denisovans to specific fossils, indicating that ancient human bones found in China belonged to them. This connection helps us learn more about where Denisovans lived and how they fit into human history.
  3. Denisovan genes are found in modern populations across Asia and Oceania, suggesting they lived in many regions. This means that their impact on human ancestry is broader than previously understood, making their story a key part of our evolutionary history.

“Math is hard” — if you are an LLM – and why that matters

Marcus on AI • 4782 implied HN points • 19 Oct 23
🔬 Science Mathematics Artificial Intelligence Machine Learning
  1. Even with massive data training, AI models struggle to truly understand multiplication.
  2. LLMs perform better in arithmetic tasks than smaller models like GPT but still fall short compared to a simple pocket calculator.
  3. LLM-based systems generalize based on similarity and do not develop a complete, abstract, reliable understanding of multiplication.

An undergrad, twitching rodent heads, DOI, and the DEA: 5 Questions for researcher Alyssa Gillies

The Microdose • 511 implied HN points • 29 Jan 24
🔬 Science Psychedelics Research Regulation Neurobiology
  1. Researchers use DOI in the lab to study the head twitch response in rodents as an indicator of 5-HT2 receptor activation.
  2. Scheduling DOI as a Schedule I drug would create barriers for labs studying the 5-HT2 receptor due to accessibility and legal requirements.
  3. There is minimal recreational use of DOI and DOC, with the DEA's proposal affecting the progress of psychedelic research and impacting researchers.

The Future Is Now #1: Floods, Pings & Regrowing Things

Everything Is Amazing • 570 implied HN points • 26 Jun 25
🔬 Science Environment Wildlife Research Technology Education
  1. There's a big interest in science stories that highlight the amazing things happening in the world. People want to hear about the good and exciting discoveries too.
  2. The news often focuses on negative events, making it hard to notice all the positive changes and breakthroughs in science that are happening quickly.
  3. Scientists are making incredible advancements, like understanding animal communication and exploring ancient natural disasters, which can be truly fascinating.

Complex systems emerge from simple rules

Tecnica • 55 HN points • 28 Jul 24
🔬 Science Biology Physics Chemistry Emergence Complexity
  1. Complex systems can develop from just a few simple rules, like in the Game of Life. It shows how starting with basic ideas can create amazing patterns and interactions.
  2. Emergence means that new, complex properties appear from simpler components interacting together. This is true for nature, biology, and even technology like AI.
  3. Human thoughts and actions also have an emergent nature. We often wonder where our ideas come from and what rules guide our decisions.