Scott McGrath

U.S. Solar Installations Fell in 2025 as Trump Attacked Clean Energy

U.S. Solar Installations Fell in 2025 as Trump Attacked Clean Energy.
More solar energy was added to U.S. grids than any other technology, but the amount installed fell by 14 percent, according to a new report.

Integrating temporally localized notes into sepsis models boosted predictive performance to an AUC of 0.84, significantly outperforming traditional models using only structured numerical data (AUC 0.77).
#MedSky #MedAI

Identifying and timing patient outcomes in clinician notes using large language models Key challenges in leveraging unstructured clinician notes for predictive models include identifying and timing patient outcomes. To address these challenges we applied large language models (LLMs) to identify and temporally localize patient outcomes in clinician notes, and evaluated whether this contextual data enhances predictive modeling for conditions like sepsis.

LLMs can reliably identify and time patient outcomes in unstructured clinician notes. Meta-Llama-3.1-8B hit the best balance of precision and coverage, while the biomedical-focused BioMistral 7B tended toward over-prediction. Contextual reasoning beats exact term matching.
#MedSky #MedAI

How to Talk to Someone Experiencing 'AI Psychosis' Mental health experts say identifying when someone is in need of help is the first step — and approaching them with careful compassion is the hardest, most essential part that follows.

How to Talk to Someone Experiencing 'AI Psychosis'

www.404media.co/ai-psychosis...

5 takeaways from Nvidia CEO Jensen Huang's rare insider blog post on AI Huang made the case that AI is essential as the internet and electricity.

Nvidia’s Jensen Huang argues we’re still in the early stages of a long AI buildout.

He lays out what he considers his "five-layer cake" framework for AI:

Energy → chips → infrastructure → models → apps.

Effects of daily multivitamin–multimineral and cocoa extract supplementation on epigenetic aging clocks in the COSMOS randomized clinical trial - Nature Medicine In a prespecified ancillary analysis of the COSMOS randomized trial, supplementation with daily multivitamins, but not with cocoa extract, over the course of 2 years decreased biological aging, as…

#MedSky Direct link to the study: www.nature.com/articles/s41...

Daily multivitamin slows signs of biological ageing The supplement’s anti-ageing effect was greater in people who were already biologically older than their years.

Taking a daily multivitamin for two years slowed biological aging by about four months in older adults, according to new data in Nature Medicine. What’s interesting is the effect was strongest in those already showing accelerated aging.
#MedSky

OpenAI and Google Workers File Amicus Brief in Support of Anthropic Against the US Government Google DeepMind chief scientist Jeff Dean is among the AI researchers and engineers rushing to Anthropic's defense.

OpenAI and Google Workers File Amicus Brief in Support of Anthropic Against the US Government

www.wired.com/story/openai...

Glad to be at the @himssgc.bsky.social preconference focused on ##datamodernization in #publichealth. #interoperability #partnerships

How Rising Home Insurance Costs Are Linked to Your Credit History In many states, homeowners with “excellent” credit scores are charged less for home insurance than those with worse scores, even if they live in a disaster-prone area.

How Rising Home Insurance Costs Are Linked to Credit Scores

www.nytimes.com/interactive/...

A novel digital twin strategy to examine the implications of randomized clinical trials for real-world populations - npj Digital Medicine npj Digital Medicine - A novel digital twin strategy to examine the implications of randomized clinical trials for real-world populations

Why do two trials for the same treatment give opposite results? It’s often the population, not the drug. Researchers built RCT-Twin-GAN to map trial data onto real-world traits. It predicted outcomes in SPRINT and ACCORD by shifting baseline characteristics.
#MedSky

Who’s a Better Writer: A.I. or Humans? Take Our Quiz. (Gift Article) A.I. chatbots contain the sum of all human knowledge. That can make them pretty good writers.

Take a quiz to see whether you prefer human or A.I. written passages.

MAJOR warning required here that you might hate your results.

www.nytimes.com/interactive/...

Career effects of preprints get mixed reviews from biomedical researchers Junior researchers are more likely to embrace preprints; grant reviewers and hiring committees express doubts

🧪 While biomedical preprints spiked during COVID, the field is adopting them slower than physics or economics. A new survey of 1,800 researchers finds that only 16% believe preprints actually help with hiring or grants, as most evaluators still default to the "vetted" status of traditional journals.

While the recent 3–7% budget cut for US research is better than the 33% reduction originally feared, money isn't the only thing at stake.
We’re seeing a massive shift at the NIH toward top-down control that risks gutting the decentralized model that made US science a superpower.
#AcademicSky

Opinion | AI Detection Pushed My Students to Use AI The perverse incentives of our attempts to combat cheating.

Here is the article referenced from in the Techdirt piece from The Chronicle of Higher Education about AI Detection and college writing:
#AcademicSky

Things don't occur in a vacuum. Threatening students with failure or expulsion is going to terrify them, particularly when the "AI detection" methods are flawed.

This is a really good look at some of the unintended consequences of harsh AI penalties from @masnick.com

#AcademicSky

The uncomfortable truth about hybrid vehicles Plug-in hybrid owners rarely actually plug in their vehicles, practically negating the climate advantages of the technology. 

This story is wild to me.
We leased our PHEV last summer, and it is never NOT plugged in. PHEVs tend to be pushed to the top of the trim line, so to be paying top dollar for something you aren't using blows my mind.

It was the last day of ski lessons, so why not go for the 360!

ML: For You feed

ML: Blend feed

Almost the same.

It is the same whitelist, same RegEx, just different algorithms pointing at it.

I'm curious to see what people think.

Governing real-world health data as a public utility The utility model offers a framework for ethical stewardship, patient empowerment, and distributed innovation

Governing healthcare data of the people, by the people and for the people -- a call to action!

www.science.org/doi/10.1126/...

Tagging for #MedSky

A Word to the Wise: Don’t Trust A.I. to File Your Taxes

This shouldn't be surprising, but do resist the temptation: LLMs are not ready to file your taxes for you.

Factors for Patient Trust and Acceptance of Medical Artificial Intelligence This survey study examines the associations of patient trust in and choice of medical scenarios involving artificial intelligence with receiving information on governance mechanisms, clinician…

What actually makes patients trust an AI diagnosis?
A survey of 3k adults asked people to choose between hypothetical AI diagnosis scenarios and pick a "visit" from the options presented. AI performance accuracy boosted visit preference by 32.5%, while a clinician in the loop added 18.4%.
#MedSky

RFK Jr.’s anti-vaccine policies are "unreviewable," DOJ lawyer tells judge The startling claim came amid a lawsuit from the American Academy of Pediatrics.

A DOJ lawyer just told a federal judge that Health Secretary RFK Jr.’s vaccine policies are "unreviewable". This claim of limitless authority comes as the AAP and medical groups sue to block unilateral changes to the CDC's childhood vaccine schedule.
#MedSky 🛟

ChatGPT Health performance in a structured test of triage recommendations - Nature Medicine A stress test of ChatGPT Health triage revealed missed high-risk emergencies and inconsistent activation of suicide-crisis safeguards, raising safety concerns for consumer-scale deployment.

Direct link to the Nature study looking at ChatGPT Health: www.nature.com/articles/s41...

The era of Doctor AI is already here Artificial intelligence is opening access to medical information in an entirely new way.

40M people ask ChatGPT health questions daily.
While OpenAI promotes that GPT-5 hits 99% triage accuracy, a Nature study found older models under-triaged 50% of emergencies. Convenience is winning, but pushing forward before these systems are validated is irresponsible.
#MedSky

Pricing models for diagnostic AI based on qualitative insights from healthcare decision makers - npj Digital Medicine npj Digital Medicine - Pricing models for diagnostic AI based on qualitative insights from healthcare decision makers

Pricing for diagnostic AI is often a black box of technical metrics that don't fit hospital budgets. A new study (n=17) found a 76% resistance to pure usage models. Turns out, leaders want "clinically legible" units, like per patient or per case, to make AI both predictable and scalable.
#MedSky

NIH reneges on recognizing union for early career researchers The biomedical research agency says trainees in its labs are not “employees”

Nearly 5,000 early-career researchers at the NIH just saw their union recognition pulled. The agency now argues these fellows are trainees, not employees, despite a ratified contract from 2024. It’s a massive shift that puts raises and health insurance for 5,000 scientists in limbo.

AWS launches a new AI agent platform specifically for healthcare | TechCrunch AWS is launching Amazon Connect Health, an AI agent platform that will help with patient scheduling, documentation, and patient verification.

AWS launches a new AI agent platform specifically for healthcare

techcrunch.com/2026/03/05/a...
#MedSky

991.1 C2