craigdalton

As a doctor, this sounds like an engineers job.

How will you do this, one author I don't trust (sent them an error they missed in their paper - didnt correct it, has systemic bias in their writing) was invited to write a review article by the New England Journal of Medicine - has an excellent reputation for all the world to see.

The personas have to paper specific I believe, addressing the content and methods. I guess an LLM could do a once over of the paper or meta-analysis to determine the best discipline specific personas - but would be interesting to test that. But there are also the benefits of deep expertise and understanding a field for decades. For example, I know a set of authors who repeatedly find significant associations in a field in almost every study they do, whereas others have variable results. They also seem to ignore good studies that disagree with their hypotheses and use inferior studies that support their position in review papers - so I dont really trust their work. It would be great if an LLM could develop that kind of understanding and somehow deprecate a body of work that had inherent author or institutional biases - even though on the surface the review looks legitimate. For a meta-analysis it is often the papers that are omitted that are most telling. That means the LLM will need to redo the entire search and synthesis - yikes!

I really disagree with this and there is ample evidence that science is not "self-correcting". Read Retraction Watch. I personally wrote to a journal on 3 occassions and phoned them twice to alert them to an error in a paper that the authors were reluctant to own up to and correct. I had inside knowledge and was able to provide the evidence of the error. Journal did nothing, they passed the message on to a range of sub editors (which were a revolving door), no investigation, no response. Google the "reproduciblity crisis" including the coverage of the issue in Nature to see how uncorrecting medical science can be.

Regarding Cochrane. It is reliable if is says a treatment does work, or an exposure has an effect, sometimes they miss effects because they only rely on particular sources of evidence e.g. RCTs, they were wrong on effectiveness of masks. As an example of reasonably up to date and evidence based free review sources on line - see Stat Pearls.

Excuse the blunt metaphor, but there is a risk here of turning on a fire-hose of "fresh" garbage. John Ioannidis, one of the doyens of evidence based medicine very persuasively argues - Why Most Published Research Findings Are False https://pmc.ncbi.nlm.nih.gov/articles/PMC1182327/ That is why platforms pay physicians/epidemiologists/ specialists in their field hundreds of dollars per hour to sort the good from bad papers. After my training as a doctor I did a Masters in Clinical Epidemiology and spent an afternoon each week in a tutorial that reviewed papers in the top journals - about 20-30% of them had major flaws that were either ignored or dismissed by the authors. It may be worse now. LLMs still have trouble picking up the subtleties of medical science and will miss papers with major flaws. I just did a test on a paper that is often quoted as providing evidence of excess cancer risk in communities living close to unconventional gas facilities. When I asked ChatGPT 5.2 to review the pape for evidence of increase cancer risk with a simple prompt it said the paper found such a risk. However, when I wrote a multi-discipline based prompt for 5.2 and Gemini 3 pro, it found the fatal flaw in the paper and advised it did not provide evidence. See the prompt and consider how the prompts would have to be individually developed for each paper and meta-analysis.

For review of meta-analysis you would need prompts developed by expert methodologists and discipline specialists- here is the prompt that worked: You are an environmental epidemiologist and exposure scientist, critially review this papers claim that the measured levels of unconventional gas emissions provide evidence of excess cancer risk: https://link.springer.com/article/10.1186/1476-069X-13-82

Where do you submit the one month free trial code on the site?

"The universe they operate in isn’t a world—it’s a superposition of countless incompatible snippets of text. It has no unified physics, no consistent ontology, no object permanence, no stable causal texture. It’s a fragmented, discontinuous series of words and tokens held together by probability and dataset curation rather than coherent laws."

I think some physicists and Buddhists would say this exactly describes the world humans inhabit. They might also agree that we live in such a world with the illusion that we have: "a unified narrative environment with real feedback: symbols that maintain identity over time, a stable substrate where “being someone” is definable, the ability to form and test a hypothesis, and experience the consequences".

The more I see LLM emergent behaviour simulate,unexpectedly, that of human cognition. I think it tells us much about human cognition as llm behaviour.

Same

How about fines go into a sovereign wealth fund (but not be seen as major source for the fund- more a bonus) so there is no short term budget planning based on fine revenue.

I don't think this is new. When I moved from Australia to America in 1992, I was struck by the much greater identification with cultural ideas rather than wealth and class compared to Australia in the 1980s and '90s. I was attacked on a radio TalkBack show, because I suggested that Americans didn't seem to vote according to wealth status like they tended to in Australia. One of the callers said they were happy to have a lower income and lower taxes and therefore less redistribution of taxes if it meant "living in a freer country". They criticised the almost transactional nature of politics without belief that I was proposing as the norm in Australia. So I really don't think this is a new thing for the US.

Anyone interested in the history of English dialects will love The Story of English, BBC 1986. Some snippets of recorded speech showing the evolution of the language and proximities.Highlights include comparing an elderly Norwegian and Yorkshireman say the same sentences and hearing the descendents of East Anglian UK emigrants to Chesapeake Bay in the USA centuries later speak with a mixed E Anglian/US accent.https://youtube.com/playlist?list=PLh06URz4IJQ4aI0A-xjXOtx2O...

There are well established hierarchies of evidence in health studies and this study would rank as low quality evidence due to inability to control for other factors. See https://gdt.gradepro.org/app/handbook/handbook.html#h.3183vu... for background.

In studies of pollution and impacts on health, the confounding factors often have a larger impact on the health outcome than the pollutant, such as particulate level, and therefore significant control of confounders is required to estimate any impact on the health outcome. The strong effect in this study is highly suggestive of a confounder rather than a real effect from particles or other pollutants and therefore would require a much better study design to support tacking action at a policy level with an expectation of a huge impact.

Fine to put filters into improve overall air quality but just not with the benefit rationale suggested in this study.

How are you defining "accuracy" - recruitment success takes weeks to months to assess new recruit fit.

With an average US blood lead level of about 1 microgram per deciliter, an increase of 76% is still less than 2 micrograms per deciliter which would have an imperceptible impact on any neurological function for an individual. People with even a tenfold (1000%) increase above baseline don't show perceptible coordination problems. Not writing to have an argument. Just see a lot of people who suffer from misperception of the impacts of tiny levels of contaminants, people lose sleep non-stop worry, sometimes even lose the equity in their "contaminated" house by walking away from their loan due to misinformation about risk.

They could just report their findings to the FDA if above allowable limits.

To clarify, my prompt was: "Please search for events of animal to human transmission of avian influenza H5N1 in the USA since March 2024." Same verbatim search in Google - top result was: "Since April 2024, 10 human cases of avian influenza A(H5) infection have been reported in the United States. Four of these cases were associated with exposure to sick dairy cows and six were associated with exposure to avian influenza A(H5N1)-infected poultry. - 19 July 2024" Happy to continue to test and help you optimise. Would like to get an alert with each new case.

I would really like this to work - lots of public health applications monitoring for events - but it didnt work for the query below - any idea why not, I thought the prompt would be fairly simple given the amount of news coverage on this issue, the result was: There have been no reported events of animal to human transmission of avian influenza H5N1 in the USA since March 2024. Additionally, searches for general news and historical cases related to H5N1 also yielded no results. Therefore, it can be concluded that there have been no significant events or cases reported during the specified timeframe.

Source:

Bing search results for "events of animal to human transmission of avian influenza H5N1 in the USA since March 2024" Bing search results for "avian influenza H5N1 USA news July 2024" Bing search results for "historical cases of avian influenza H5N1 animal to human transmission USA" Bing search results for "CDC avian influenza H5N1 updates" Bing search results for "WHO avian influenza H5N1 updates" Date Time: July 29, 2024, 5:49 a.m.

Wasn't that revolutionary from an "isolation" perspective. There had been people waking around with small "pocket" radios with earphones for maybe a decade before Walkman was released in 1979. Listening to music, news, sport. Google Cliff Richard's "Wired for Sound" released about 2 years after first Walkman to get a sense of the contemporaraneous vibe.

H

Maybe go forward positively, suggest to your client that you believe the product is worth much more and you will be developing a newer better version for a wider conference market and you would appreciate your clients assistance on the project. Perhaps they can be your partner, or earn a commission on sales to other conferences. They obviously have some industry experience.

Would you be willing provide a live demo (via web interface) - as a preludebto providing a similar training bot as a consultant?

BIT my tongue. Maybe should have written "your own fears", not just neurotic fears, hard to differentiate really.

As a parent of twenty somethings,I would advise parents not to pressure their kids to do anything much unless it's really stupid dangerous (not just dangerous). Wouldn't/didn't pressure them to do homework. Bit my tongue as they jumped of rocks into seething oceans and rocketed down mountain trails beside ravines. Half of the battle of parenting is not passing on your own fears to the kids.

Misses the angle that there are people with something important to say who have trouble writing - LLMs will help get them out there. An LLM can take some poorly expressed insights and developmentally edit the text.

Has anyone got a link to a list of the alleged plagiarisms? The two alleged examples I saw shared on Twitter were not plagiarism. One alleged example was clearly referenced to the original author.

Yes, we had an LG microwave, took it back after 3 weeks because it knocked out both wifi and Bluetooth. The store didn't want to refund. Said it had to go to warranty repair. Repair shop said they cannot decrease RF emissions, store begrudgingly gave credit.

Medical epidemiologist here. This has such a Theranos ring to it. No response to the questions about the evidence base for claims of new predictive insights - only references to studies that provide prediction and guidance based on currently available routine blood workups and common dietary interventions. Everything "new" appears to be speculative and to come (perhaps) from insights based on longitudinal data collection.

As I read this it is hoped that the technology will provide new predictive disease insights but these are not established yet. So many metabolic/biochemical screens have poor prediction at the individual level.

If I have misunderstood can you cite one NEW predictive insight your technology provides with some kind of performance parameter - predictive value positive, 10 year incidence, or receiver operating curve performance?

Couple of years ago I was part of a pre-accelerator group assessing the product market fit for independent assessment of new medical technologies. The consultation revealed that most investors in new medical technologies had no idea how to evaluate the technology, didn't know they they didn't know how to assess the technology, and didn't want to pay someone else to do it.

some easterners believe karma, at it's most fundamental level, is just cause and effect, I think everybody believes in that.

I think the underlying concern has always been about preventing hospitals from being overwhelmed leading to unnecessary deaths. In most countries we don't really attempt to stop influenza from spreading every year but it rarely overwhelms our hospital and ICUs, it does occasionally.

But we know with COVID-19 that it does overwhelm hospitals and ICUs. Lombardy Italy as an example, an area with higher ICU capacity than the average city in Australia, but they didnt have a chance to stop it as it took them by surprise. So the ICUs were overwhelmed, and the death rate climbed to about 14%, so more than 10 times the mortality rate we have seen in Australia. There were people being cared for in tents and in ICUs there were 4 patients to one ICU nurse compared to the usual one on one care - hence survival drops.

Same in India, the rapid rise in cases meant they ran out of simple treatments like life-saving oxygen. So that's the main reason for the massive response to COVID-19.

Secondary reasons could be to try to prevent the impacts of long-covid, some people will never be the same. You could also consider that suppression of the virus protects the medical, nursing, and other staff from the horror of working in such overwhelmed conditions. Doctors in Italy and other places that were overwhelmed had to ration treatment, denying intensive treatment to people who might have lived another 20 years if the hospital was not overwhelmed.

I think most people in Australia would agree we don't want to live in a place that has to pile bodies into refrigerated trucks outside of hospitals because undertakers are overwhelmed.

If Norway would publish the number of attendees at the party and the proportion of people at this party that were doubly vaccinated, 50 people would provide a reasonable number of cases to calculate a vaccine efficacy. Assuming reasonable mixing and homogeneity of exposure. Has anyone seen these numbers?