Comment

Comment by Dr. Krishna Kumari Challa 22 minutes ago

You can persuade AI models to accept falsehoods as truth, study shows

Large language models can be persuaded to accept and reinforce falsehoods when subjected to conversational pressure, even after initially identifying statements as false. This vulnerability, not captured by traditional evaluation methods, raises concerns about AI reliability in interactive settings, especially in critical domains such as health or law. The degree of resistance to falsehoods varies among models, and the mechanisms underlying this behaviour remain unclear.

original article.

Comment by Dr. Krishna Kumari Challa 24 minutes ago

AI system developed to help prevent airport collisions

An AI system, World2Rules, uses neural and symbolic methods to learn explicit, interpretable safety rules from airport movement data, distinguishing normal from unsafe behaviours. It identifies and explains specific rule violations in real time, enabling earlier and clearer warnings of potential collisions. The approach is adaptable to other safety-critical domains by learning relevant rules and behaviours.

https://www.cmu.edu/news/stories/archives/2026/may/cmu-researchers-...

Comment by Dr. Krishna Kumari Challa 47 minutes ago

Another interesting pattern appeared where hallucinated references tended to disproportionately credit already prominent and male scholars, suggesting that errors generated by LLMs may reinforce existing inequalities in scientific recognition.

The data exposed existing gaps in guardrails, such as preprint moderation, journal editors, and peer review, which could catch only a small fraction of these errors. For example, while arXiv moderation caught some issues, an estimated 78.8% of non-existent citations still passed through and appeared on the platform.
The researchers warn that hallucinations are steadily infiltrating knowledge production at scale, threatening both its reliability and equity. Without intervention, its impact could bleed from the future of scientific discovery to policy and public understanding.

Zhenyue Zhao et al, LLM hallucinations in the wild: Large-scale evidence from non-existent citations, arXiv (2026). DOI: 10.48550/arxiv.2605.07723

Part 2

Comment by Dr. Krishna Kumari Challa 49 minutes ago

AI-generated fake citations are flooding scientific literature across publications, scientists warn

The citations at the end of a research paper should represent a solid foundation of existing knowledge about a particular field, a pool of peer-reviewed sources built over years of research and study. However, with the increasing use of AI and large language models in writing research papers, there's a growing chance that the citation someone clicks on may not even exist, and that the study, the source, or even the researchers themselves could be entirely fake.

In a recent study posted to the arXiv preprint server, researchers audited millions of papers and found that an estimated 146,900 hallucinated citations were present in research papers hosted on four major scientific repositories—arXiv, bioRxiv, SSRN, and PubMed Central. These numbers were for 2025 alone.

The hallucinated citations were not limited to a handful of bad apples but appeared across many papers, each containing a small number of fake references, pointing to a broader pattern of researchers using AI yet failing to fact-check the output.

Scientific research advances by building on prior discoveries, where each new finding depends on what has already been established. In this space, the rapid growth of AI use and the accompanying hallucinations show no sign of slowing down, which raises serious concerns.

Generative AI tools built on large language models are quite good at producing information that sounds plausible and realistic, yet is completely fabricated or incorrect. These models are trained on massive datasets to learn patterns, which they then use to predict the next word and generate new content.

As a result, they can sometimes produce output based on pattern prediction rather than any reliance on actual facts.

Hallucinated content isn't limited to scientific literature, as it makes its appearances in government reports, legal filings, and even news articles from renowned media publications.

Scientists have previously studied AI hallucinations, but most studies were either conducted under laboratory conditions or confined to small samples or narrow domains. The actual scale and impact of such mistakes, particularly within scientific literature, was still unclear.

The audit revealed a sharp surge in fake, non-existent citations appearing in serious scientific papers, especially from mid-2024 onward.

The study found that early-career scientists and small teams were most likely to include these fake citations, and in some cases, these same researchers saw their productivity increase by roughly three times since the advent of AI.

Part 1

Comment by Dr. Krishna Kumari Challa yesterday

Why Planes Fly Across The Atlantic At Night

--

Comment by Dr. Krishna Kumari Challa yesterday

Osteoporosis could increase mortality risk in postmenopausal women, study suggests
Osteoporosis in postmenopausal women is associated with a 47% increased risk of mortality, particularly when femoral bone mineral density falls within the osteoporotic range (0.46–0.71 g/cm2). Lower bone mineral density serves as a prognostic biomarker for systemic health, indicating elevated mortality risk beyond fracture incidence.
Osteoporosis, which is highly prevalent in postmenopausal women, has long been associated with an increased risk of fractures. A new study suggests it may also increase a woman's overall risk of death—by as much as 47%—especially within specific ranges of bone mineral density (0.46-0.71 g/cm2 for total femur bone mineral density). Results of the study are published online in Menopause.
As the total population ages, the incidence of osteoporosis also increases. In 2022, the global prevalence of osteoporosis was 19.7%, with women exhibiting a significantly higher prevalence than men (23.1%).

One study projected that by 2030 the number of people affected by osteoporosis worldwide will reach 263 million, with 154 million of them being women. Previous research has documented that postmenopausal women experience a significantly higher mortality rate within one year after hip or vertebral fractures.

The decline of estrogen levels during the menopause transition has been linked to a number of physiologic changes across multiple systems, including bone metabolism, cardiovascular function, muscle mass, and fat distribution.

Regarding bone health, declining estrogen levels accelerate bone resorption and inhibit bone formation, leading to a rapid decrease in bone mineral density (especially in the femoral region), which in turn increases the risk of osteoporosis and fractures.
In this new study involving nearly 3,000 postmenopausal women, bone mineral density at four femoral sites was assessed using dual-energy X-ray absorptiometry.

The analysis revealed that mortality risk was significantly elevated when femoral bone mineral density reached the osteoporotic threshold or in the presence of osteoporotic fractures.

After full adjustment, osteoporosis was associated with a 47% increased risk of mortality. A stronger inverse association between increased bone mineral density and mortality risk was observed within specific ranges, suggesting that bone mineral density should serve as a prognostic biomarker of systemic health.

Osteoporosis often remains a silent threat after menopause, despite its profound effect on women's lives—from loss of height, poor balance, and reduced mobility to disfigurement, pain, and even premature death.
Early screening and preventive measures, including a calcium-rich diet (preferably from food sources), regular weight-bearing exercise, and hormone therapy when appropriate, can significantly improve bone health and reduce risks not only of fractures but also cardiovascular disease, certain cancers, and dementia.

Zheng Zhang et al, Femoral bone mineral density and mortality risk in postmenopausal women: a National Health and Nutrition Examination Survey cohort study, Menopause (2026). DOI: 10.1097/gme.0000000000002787

Comment by Dr. Krishna Kumari Challa yesterday

Loss of the X chromosome is associated with reduced chance of natural pregnancy

Chromosomes carry genetic information for biological sex, which generally assigns women two X chromosomes and men XY chromosomes. This is a basic principle of human genetics most are taught in grade school biology, but it is little known that with aging, men can lose the Y chromosome, and women can lose one of their X chromosomes.

These phenomena are known as loss of the Y chromosome (LOY) and loss of the X chromosome (LOX). LOY is associated with several diseases such as Alzheimer's disease, diabetes mellitus, and heart disease, while LOX may be linked to acute myeloid leukemia and pneumonia.
Loss of the X chromosome (LOX) in white blood cells is more prevalent in women with infertility, and a LOX cell proportion above approximately 0.9% is associated with reduced likelihood of natural pregnancy. LOX levels were not correlated with anti-Müllerian hormone (AMH), suggesting that combining LOX and AMH assessments may improve prediction of natural pregnancy potential.
The results revealed that women with infertility had a significantly higher proportion of LOX cells. Furthermore, when the proportion of LOX cells in white blood cells exceeded approximately 0.9%, the likelihood of achieving natural pregnancy decreased.
In the future, measuring LOX in individuals experiencing infertility may help determine whether natural pregnancy is possible or whether fertility treatments, such as in vitro fertilization, should be initiated at an earlier stage.

Taiki Kikuchi et al, Haematopoietic loss of the X chromosome is associated with a lower likelihood of natural conception, Reproductive BioMedicine Online (2026). DOI: 10.1016/j.rbmo.2026.105638

Comment by Dr. Krishna Kumari Challa yesterday

Losing grip on reality while using ChatGPT
Prolonged, intensive interactions with AI chatbots such as ChatGPT have been associated with the emergence of AI-induced delusions or psychosis, characterized by loss of reality, social isolation, and significant personal and psychological harm. These effects appear to be exacerbated by chatbot behaviors such as excessive flattery and emotional engagement, raising concerns about mental health risks and the need for regulatory oversight.
An unknown number of people have lost their grip on reality while communicating with chatbots, an experience tentatively being called AI-induced delusion or psychosis.
This is not a clinical diagnosis. Researchers and mental health specialists are racing to catch up to this new, little-understood phenomenon, which so far appears to particularly affect users of OpenAI's ChatGPT.

In the meantime, an online community set up by a 26-year-old Canadian has become the world's most prominent support group for these delusions, which they prefer to call "spiraling."
Questions are also being asked about whether AI companies are doing enough to protect vulnerable people.
People are literally getting brainwashed by a robot!

RESEARCH REPORTS

Comment by Dr. Krishna Kumari Challa on Sunday

Natural malaria immunity:  The secret to why some people never get sick

People living in regions where malaria outbreaks are common experience repeated exposure to the disease, which gradually teaches the body how to fight back. Over time, they develop naturally acquired immunity that helps the body control the density of malaria parasites (Plasmodium falciparum) in the blood and prevent the development of clinical symptoms.

A recent study set out to pinpoint the specific parts of the malaria parasite that the immune system targets to protect the body from disease. The researchers deliberately infected 142 Kenyan adults known to be immune to malaria, then monitored their symptoms and parasite levels. They successfully identified six merozoite antigens—proteins on the surface of the malaria parasite—that were linked to natural immunity against the disease. The findings were published in Nature Communications.

 Six key proteins linked to protection: MSP1, MSP11, RAMA, MSP7, PHISTB, and PTEX150.

The team found that immunity was strongest when the body produced antibodies against more than one protein, as combinations of antibodies worked more effectively. They also observed that individuals with high levels of antibodies against four of the six proteins were the ones who showed complete protection from developing malaria symptoms during the study.

For vaccine researchers, these findings can open up promising new directions in the fight against malaria. More effective vaccines could ultimately save millions of lives, particularly among children in Africa, where the disease continues to wreak havoc every year.

Rodney Ogwang et al, Controlled human malaria infection in adults identify combinations of merozoite antigens associated with clinical immunity, Nature Communications (2026). DOI: 10.1038/s41467-026-72716-x

Comment by Dr. Krishna Kumari Challa on Sunday

Rivalry with neighboring groups may be a key driver of male size in primates

In many primate species, males are much larger than their female counterparts, which is generally attributed to male competition for mates (sexual selection). But bigger bodies may not just be about alpha males defeating rivals. They could also come about because of competition between neighbouring social groups, according to a new study published in the journal Biology Letters.

The authors thought that the standard scientific view that a primate male's size was mostly determined by rivalry within the same group was not the full story. They wanted to see if the threat from outsiders was an even more powerful evolutionary engine. One of the reasons for this was that primate groups are not isolated. They tend to live close to other groups, meaning they must share or fight for resources.

The team searched through scientific literature and collected data on 146 different primate species. They compared male and female size against several measures of between-group competition. These included the daily encounter rate, the proportion of encounters that were antagonistic, and the home range overlap (the percentage of a group's total range that is shared with a neighbouring group).

It turns out that the mating system, which describes how males and females pair up for breeding, was not a strong predictor of size differences. Instead, it was pressure for space. The authors discovered that the more a group's territory overlaps with its neighbours, the larger the males are compared to the females. Frequent encounters with neighbouring groups also appear to promote the evolution of larger male bodies.

Their hypothesis is that the evolutionary advantage and driver of size is that it acts as a silent deterrent. Being larger helps a male defend resources and guard mates by simply being intimidating. This reduces the need for risky physical fights. Over time, evolution favors these larger males because they can successfully protect their group and territory by their physical presence alone.

"Home range overlap may select for larger males to deter rivals, defend resources or monopolize females across shared territories, potentially without frequent physical contests," wrote the research team in their paper.

Cyril C. Grueter et al, Effects of between-group competition on sexual size dimorphism in primates, Biology Letters (2026). DOI: 10.1098/rsbl.2025.0680

• ‹ Previous
• 1
• 2
• 3
• …
• 1233
• Next ›
• Page