The hottest Research Substack posts right now

And their main takeaways
Category
Top Health Politics Topics

Import AI 370: 213 AI safety challenges; everything becomes a game; Tesla's big cluster

Import AI • 439 implied HN points • 29 Apr 24
🕹 Technology Research
  1. Chinese researchers introduced MMT-Bench, a benchmark for assessing visual reasoning in language models with diverse tasks and scenarios.
  2. Researchers developed a system to turn 2D photos into 3D gameworlds, showing AI's capability to transform real-world imagery into interactive experiences.
  3. A consortium of researchers addressed 213 AI safety challenges across 18 areas, emphasizing the urgent need for solutions to ensure the reliability and safety of language models.

We need to industrialize virtual synthetic repurposing trials

Trevor Klee’s Newsletter • 970 implied HN points • 10 Jul 25
🏥 Health Politics Research
  1. Virtual synthetic repurposing trials use existing healthcare data to see how already available drugs can help treat various diseases. This method can lead to important insights without needing traditional trials.
  2. Currently, these trials are done by small teams and can be slow and hard to replicate. There’s a call for a more organized approach that uses technology to speed up the process and improve access to data.
  3. By setting up teams focused on software, data cleaning, and navigating regulations, we could create a system that shares results openly. This would allow more researchers to explore and build on findings.

Bioelectric Signalling and Cellular Communication

Solve Cancer in 365 days • 39 implied HN points • 28 Aug 24
🔬 Science Research
  1. Bioelectricity is how cells communicate using tiny electrical signals. Think of it as the way cells send messages to each other in the body.
  2. Cells have gate-like structures called ion channels that control the movement of charged particles, creating bioelectric signals. These signals help in many processes such as healing and development.
  3. Manipulating bioelectric signals could potentially change how cells behave, which might lead to new ways to treat diseases like cancer by changing unhealthy cells back to normal ones.

Disease X - Turns Out Death Is Profitable

AND Magazine • 1061 implied HN points • 15 Jan 24
🏥 Health & Wellness Research
  1. Lab leaks and mishaps involving dangerous viruses are increasing worldwide post COVID-19 pandemic.
  2. Scientists are creating highly lethal viruses through experiments, raising concerns about potential future pandemics.
  3. There is a focus on preparing for 'Disease X' which may lead to profit-making opportunities and increased restrictions.

All Bayes Everything

Holodoxa • 239 implied HN points • 14 Jun 24
🔬 Science Research
  1. Bayes' Theorem is a powerful concept in probability theory that helps update beliefs based on new evidence, highlighting the importance of combining prior knowledge and new data.
  2. Bayesian methods can offer valuable improvements to scientific research practices by emphasizing uncertainty, effect magnitude, and probability distributions over traditional p-values and null hypothesis testing.
  3. The concept of the brain functioning as a prediction machine aligns with Bayesian principles, suggesting that the brain uses prior knowledge and new sensory inputs to make predictions and construct conscious experiences.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Natural Hydrogen: Something or Nothing?

Doomberg • 6944 implied HN points • 03 Feb 24
🔬 Science Research
  1. Hydrogen could transform from just an energy carrier to a fuel.
  2. Discoveries of natural hydrogen deposits could revolutionize energy production.
  3. Challenges in using hydrogen as fuel could be addressed with new natural resources.

A new generation of AIs: Claude 3.7 and Grok 3

One Useful Thing • 1968 implied HN points • 24 Feb 25
🕹 Technology Research
  1. New AI models like Claude 3.7 and Grok 3 are much smarter and can handle complex tasks better than before. They can even do coding through simple conversations, which makes them feel more like partners for ideas.
  2. These AIs are trained using a lot of computing power, which helps them improve quickly. The more power they use, the smarter they get, which means they’re constantly evolving to perform better.
  3. As AI becomes more capable, organizations need to rethink how they use it. Instead of just automating simple tasks, they should explore new possibilities and ways AI can enhance their work and decision-making.

Kimi

ChinaTalk • 815 implied HN points • 18 Jul 25
🕹 Technology Research
  1. Moonshot AI recently released Kimi K2, a powerful open-source language model that focuses on long context, allowing it to analyze large texts effectively.
  2. The Kimi K2 model learned a lot from its competitors, especially DeepSeek, and showcases the strength of open-source culture in driving innovation in AI.
  3. Moonshot aims to create user-friendly AI that feels engaging and human-like, shifting from traditional chatbots to interactive experiences that meet user needs.

The High Tragedy of the Pandemic

Glenn Loury • 2856 implied HN points • 04 Jul 23
📰 News Research
  1. The pandemic may have originated from a lab leak due to gain-of-function research.
  2. Fauci and others believed they were doing good by conducting this research, but it had unintended consequences.
  3. Cooperation and transparency from both Chinese and American authorities could have potentially prevented the pandemic from reaching a global scale.

Most Impactful Generative AI Papers of 2023

AI Supremacy • 1022 implied HN points • 06 Jan 24
🕹 Technology Research
  1. The post discusses the most impactful Generative AI papers of 2023 from various institutions like Meta, Stanford, and Microsoft.
  2. The selection criteria for these papers includes both objective metrics like citations and GitHub stars, as well as subjective influence across different areas.
  3. The year 2023 saw significant advancements in Generative AI research, with papers covering topics like large language models, multimodal capabilities, and fine-tuning methods.

Reader Question – What are Weight-Related Health Conditions?

Weight and Healthcare • 898 implied HN points • 24 Jan 24
🏥 Health & Wellness Research
  1. Weight-related health conditions are health issues that are often labeled as related to weight just because higher-weight individuals have them more frequently.
  2. The labeling of health conditions as 'weight-related' overlooks other factors like weight stigma, weight cycling, and healthcare disparities that could actually be the main contributors.
  3. The concept of 'weight-related conditions' is used to push for risky weight loss interventions and profits, but it doesn't necessarily support the health of higher-weight individuals.

The End of Search, The Beginning of Research

One Useful Thing • 2047 implied HN points • 03 Feb 25
🕹 Technology Research
  1. New AI Reasoners can think better and solve tougher problems by producing thinking steps before answering. This makes them more effective than earlier chatbots.
  2. AI agents are being developed to autonomously pursue goals, but they currently face limitations when tackling complex tasks. They show promise with narrow, task-specific applications.
  3. OpenAI's Deep Research represents how specialized AI can work like a human researcher by engaging deeply with academic topics, paving the way for significant advancements in research efficiency.

Zepbound/Mounjaro (Tirzepatide) for Weight Loss - Part 1

Weight and Healthcare • 818 implied HN points • 03 Feb 24
🏥 Health & Wellness Research
  1. Tirzepatide, known as Zepbound for weight loss, was initially a Type 2 diabetes medication by Eli Lilly that got FDA approval for weight loss due to its side effect of weight loss. It works by interfering with normal digestion processes and hunger signals.
  2. There are dosing differences between the diabetes and weight loss applications of Tirzepatide. For weight loss, patients are encouraged to continuously increase the dosage to the highest tolerable level, potentially facing side effects that impact their quality of life.
  3. The side effects of Tirzepatide/Zepbound can range from mild discomfort to serious conditions like thyroid C-Cell tumors, gastrointestinal diseases, and pancreatic issues. Patients need to be aware of potential risks, like drug interactions and impact on pregnancy.

Building a phosphorescence detector

lcamtuf’s thing • 2652 implied HN points • 27 Nov 24
🔬 Science Research
  1. There isn't much information available on natural phosphorescence, so creating a detector can help explore its presence in everyday items.
  2. The device design prioritizes speed and sensitivity to detect phosphorescence briefly brought out by UV light, using specific electronics like photodiodes and amplifiers.
  3. Experimenting with common household materials showed that some items, like powdered milk and natural ruby, exhibit weak phosphorescence, although many other items didn't show any notable glow.

13% of British athletes admit to doping, new study finds

Honest Sport • 99 implied HN points • 24 Jul 24
🎾 Sports Research
  1. A study found that 13% of British athletes admitted to using banned substances. This suggests that doping may be more common in sports than people think.
  2. UKAD, the organization in charge of anti-doping in the UK, needs more responses from athletes to better understand the full scale of the problem.
  3. The study reveals a big gap between how many athletes might be doping and how few get caught. This raises concerns about the effectiveness of current anti-doping efforts.

The American DeepSeek Project

Democratizing Automation • 839 implied HN points • 04 Jul 25
🕹 Technology Research
  1. The U.S. is losing its edge in AI to China, where there's more open-source innovation and a larger number of AI researchers. This is changing the landscape of AI research worldwide.
  2. There's a plan to build a fully open-source AI model in America that matches current top models. This aims to reclaim leadership in AI technologies and ensure that the AI ecosystem remains accessible and accountable.
  3. To succeed in this initiative, the community needs support and collaboration, emphasizing the importance of shared goals and new habits in developing AI models that anyone can trust and use.

Long COVID moonshot

Logging the World • 877 implied HN points • 21 Jan 24
🏥 Health & Wellness Research
  1. Perceptions of Long COVID statistics may be misleading when not considering factors like vaccination and immunity
  2. The severity of COVID infections might be linked to the likelihood of developing Long COVID
  3. Long COVID advocacy could benefit from focusing on providing treatment and research for those suffering long-term effects

SV40 Enhancer qPCR

Nepetalactone Newsletter • 1572 implied HN points • 17 Jul 23
🔬 Science Research
  1. A new qPCR assay is designed to detect SV40 enhancer in the Pfizer vaccine.
  2. The assay is more sensitive than other assays and can help track tissues with SV40 elements.
  3. This qPCR tool could be useful in screening clinical samples for PCR positivity and prioritize for further sequencing.

Science and Serendipity

Dr’s Newsletter • 746 implied HN points • 11 Feb 24
🔬 Science Research
  1. PhD research can be challenging but can lead to unexpected and serendipitous discoveries
  2. The stability of certain compounds is crucial in preventing toxicity, as observed in experiments with aluminium and silicic acid in fish
  3. Understanding the chemistry behind experimental outcomes is key, as seen in the success of a PhD research project despite initial doubts

The Second Gemini

Don't Worry About the Vase • 2374 implied HN points • 17 Dec 24
🕹 Technology Research
  1. Google's Gemini Flash 2.0 is faster and smarter than previous versions, making it a strong tool for those who want quick assistance and information.
  2. Deep Research is a new feature where users can get detailed reports based on multiple websites; it's useful but still needs improvement in accuracy and relevance.
  3. Projects like Astra and Mariner are experimental tools that aim to enhance user experience by providing real-time assistance and better interaction through voice and web browsing.

Open Thread 382

Astral Codex Ten • 1170 implied HN points • 19 May 25
🕹 Technology Research
  1. There are meetups happening this week in Oxford, Shanghai, and Austin where people can connect and share ideas.
  2. A few fellowships are available for those interested in AI safety and reasoning, with opportunities to work and collaborate in the Bay Area.
  3. Grants are being offered for projects that explore how AI can support open inquiry, encouraging creative and innovative approaches.

Suffer the little children...

Break Free with Karen Hunt • 746 implied HN points • 09 Feb 24
🕹 Technology Research
  1. Tech companies are using data from babies to teach AI common sense abilities.
  2. China and the United States are developing AI that mimics human behaviors, like intuitive physics and common sense.
  3. Artificial children like Tong Tong, created by AI scientists, raise ethical questions about the future of AI and its impact on humanity.

MIT and the non-existent I-129 problem

Gordian Knot News • 256 implied HN points • 09 Nov 25
🔬 Science Research
  1. The author feels disappointed in MIT's response to criticisms of their nuclear studies, specifically regarding a controversial analysis by a faculty member. They believe the school once encouraged rigorous debate and fact-based reasoning.
  2. Iodine-129, a radioactive element, is discussed as posing minimal health risk, but the author argues that the recent study on it fails to address its real impact on thyroid cancer risk. They highlight that the dangers may not be as significant as some suggest.
  3. The author criticizes the approach to radioactive waste management, suggesting that the focus should be on dilution rather than concentration. They believe this would significantly reduce health risks from contamination.

Nutrient Targets

Harnessing the Power of Nutrients • 1058 implied HN points • 12 Dec 23
🏥 Health & Wellness Research
  1. Customized nutrient targets for protein, ethanol, vitamins, and minerals suggested by the research may differ from official recommendations.
  2. The post is specifically for paid subscribers, indicating exclusive content access.
  3. The author, Chris Masterjohn, PhD, discusses the importance of personalized recommendations for nutrient intake.

Doesn't FDA Approval Mean Weight Loss Drugs Are Safe? - Reader Question

Weight and Healthcare • 758 implied HN points • 31 Jan 24
🏥 Health & Wellness Research
  1. FDA approval of a drug does not mean it is completely without risks; it means the benefits are deemed worth the risks.
  2. The FDA assesses the risks and benefits of a drug based on the research provided by the pharmaceutical company before granting approval.
  3. When it comes to weight loss drugs, the risk/benefit analysis by the FDA may be skewed due to questionable research around weight and health, leading to a lower safety bar for approval.

My Research on Gender Dysphoria Was Censored. But I Won’t Be.

Common Sense with Bari Weiss • 8778 implied HN points • 10 Jul 23
🔬 Science Research
  1. The retraction of a scientific article on Rapid Onset Gender Dysphoria sparked controversy and highlighted the influence of activist movements in academia.
  2. The authors faced accusations of violating editorial policies around consent, which they refuted by claiming to have followed ethical research practices.
  3. Despite the retraction, the authors plan to continue studying the controversial topic of ROGD, emphasizing the importance of scientific inquiry and resisting censorship.

Zepbound/Mounjaro Tirzeptatide for Weight Loss Part 2

Weight and Healthcare • 718 implied HN points • 07 Feb 24
🏥 Health & Wellness Research
  1. The concept of 'ob*esity' as a disease and the idea that it is progressive and relapsing are constructs of the weight loss industry and not scientifically supported.
  2. The study promoting Zepbound/Tirzepatide for weight loss has ties to the pharmaceutical industry, raising questions about potential bias and conflicts of interest among the authors.
  3. The lead author of the study, Louis J. Aronne, has a history of promoting weight loss methods without substantial research backing, raising concerns about the credibility of research surrounding weight loss drugs.