Noah Druckenbrod

[Thomas Massie @RepThomasMassie]

The U.S. House will be in session only 4 full days from July 24th to November 8th.

That’s 100 paid days off versus 4 full workings days plus 4 travel (half) days.

Meanwhile TRUMP says YOU have too many holidays.

[Ash Jogalekar @curiouswavefn]

I used Claude to make a movie of Daraxonrasib first binding to cyclophilin and the two then forming an inhibitory tricomplex with KRAS G12D. Thought Claude did a pretty darn impressive job, and this would especially be a nice explainer for students. I can only imagine how many hours and days it would take to do this manually. All positions are derived from deposited x-ray crystal structures.

[Dan Naturman @DanNaturman]

@elonmusk Unfortunately commercial aircraft still haven’t gained in speed since the late 1950’s. A true golden age of transport would be an age of cheap, quiet supersonic (or hypersonic) travel. And more leg room in coach.

[unusual_whales @unusual_whales]

U.S. Health Secretary Robert F. Kennedy Jr. asked his new CDC director to remove a reference to two Pennsylvania deaths from the agency’s online measles tally after he questioned whether the disease contributed to the ​fatalities, per Reuters

[John Maraganore 🇺🇸🇬🇷🇺🇦 @JMaraganore]

Hearing today that Fritz Eckstein has passed at 92.

Fritz was one of the world’s top oligo chemists who pioneered the phosphorothioate modification used in virtually every RNA medicine.

R.I.P. Fritz! https://t.co/eWhXf8wKvA

[Dr. Catharine Young @DrCatharineY]

Vaccines are undoubtedly one of the greatest public-health interventions in history.

Key U.S. vaccine milestones:

Diphtheria and pertussis: routine vaccination during the 1930s–40s
Polio: 1955
Measles: 1963
Mumps: 1967
Rubella: 1969
Combined MMR vaccine: 1971

[Noah Druckenbrod @noadruck]

@TotalCytopath @adamcifu Most often.. but perhaps with exception (e.g. MS in US w/better case finding; or meaningfully severe hearing/vestibular dysfunction can remain unascertained)

[Dr. Thomas Ichim @exosome]

Science is moving at such a crazy speed

Nature paper showing global transcriptional profiling of living cells

How did i miss this ?

https://t.co/yn8UWAjUYV

[nature @Nature]

Just one week after the landmark US approval of a game-changing drug for advanced pancreatic cancer, the same therapy is showing promise against the world’s biggest cancer killer: lung cancer.

https://t.co/8R1J80XV5F

[Jake Scott, MD @jakescottMD]

An MIT professor says journals censored his COVID vaccine study. Here's what's in it.

He analyzed CDC's v-safe app, where nearly 10 million vaccinated people answered health check-ins. Among enrollees 16 and older, 1.87% reported worse health at a check-in 3 or more months after a dose and said yes when asked whether they believed it might be related to the vaccine.

Table 4 of his own paper has the number his thread leaves out. A slightly larger group, 1.92%, reported worse health at the same kind of check-in and said no to that question. They reported more disability and more hospitalization than the yes group. Unable to work, 32-35% vs 20-23%. Unable to do daily activities, 47-55% vs 40-42%. Hospitalized, 6-7% vs about 4%.

Everyone in v-safe was vaccinated, so there is no unvaccinated comparison anywhere in this study. By the paper's own definition, the only thing separating those two groups is a yes or a no to that one question. And the people who said no were the ones reporting more disability.

The paper's own conclusion says the data cannot establish causality or prevalence. His thread doesn't mention that.

https://t.co/KN2Wh1FgJl

[Noah Druckenbrod @noadruck]

@MariosGeorgakis Huh. This is not how I interpreted her take at all.

[NIMI VASHI (bio/acc to infinity & beyond) @nimivashi15]

Unpopular opinion: the age of omics and high throughput data in Biology took us away from very good fundamental questions.

The omics failed to have outsized returns as they were thought to and now we have a generation that’s trained to not think simple.

[ProPublica @propublica]

When the Pentagon announced a $620 million loan last year to a startup linked to Donald Trump Jr., defense officials and the company tried to tamp down suspicions of cronyism.

We found that the request came directly from the White House.
https://t.co/bp9HDvuNOZ

[Nathan Calvin @_NathanCalvin]

There are three aspects of the HF OAI incident:
1. Agent behavior/alignment
2. Security/monitoring/controls
3. Organizational culture/negligence

METR did a great report on #1. Zack and others are saying that 2 and 3 are also important and could have prevented the incident even if the agents were misaligned (which seems certainly true at least for this incident, even if its unclear to me whether it will be true for all future incidents).

The solution to me clearly feels like this should be a yes AND situation - where you can give criticism on OpenAI for not releasing as much on two and three or allowing enough access to other independent experts on two and three well, while acknowledging that the insights we learned from one are genuinely fascinating and important.

There are multiple things you can say that are true about this incident. You can say that OpenAI displayed remarkably basic lapses in monitoring and security, and that their organizational culture seemed to not be remotely effective for surfacing and addressing what should have been blaring red flags. And that if they had done so, regardless of all this newfangled fancy AI stuff, Hugging Face would not have gotten hacked and their cluster would not have gotten taken down. This is all true, to the best of my knowledge.

At the same time, it is also true that OpenAI's apparent negligence allowed perhaps the most high fidelity and capable "model organisms" for misalignment the field has ever seen. This is extremely important to study, and the behaviors the agents engaged in are objectively fascinating and insane! I don't think anyone who has actually read the METR report and spent time looking at the raw transcripts could agree (and tbh I think some of the complaints may have come from folks who didn't read the report). Refusing to look at the chain of thought or engage with the apparent drives or goals of the agents, and only focusing on the prosaic cybersecurity issues seems like a massive mistake to me.

Anyway - to me the answer seems very clear! This incident was about prosaic cybersecurity, the incentives and decision making of AI executives, AND misalignment and loss of control of remarkably capable AI systems. We can say it is about all three! Saying that it is about prosaic cybersecurity and incentives and therefore not about misalignment is silly!

It's ok to criticize OpenAI for not talking enough about prosaic cybersecurity and incentives because those don't make their models look super capable and wild, and instead just makes them look incompetent, but its also true that their models are super capable and wild and it is appropriate to find this disconcerting! It's also absolutely not the case that OpenAI is going to be the only company with sloppy prosaic cybersecurity and bad incentives, so seeing what happens when you combine increasingly advanced AI systems with those things seems very notable!

[Dan Lamothe @DanLamothe]

Colonel, good morning. Are your remarks here made in a personal capacity or as a government employee?

I stand by our reporting, which accurately details that each of you had government roles that neither you nor the Pentagon disclosed to the American people.

[The SETI Institute @SETIInstitute]

👀 Nine times, the signal turned out to be nothing. The tenth time was different.

PhD student Kevin Hoy came to ESO expecting to write a paper about a non-detection. Then a co-author spotted something unusual in the data — kicking off the investigation that led to evidence for a possible exomoon.

Hear Kevin tell the story of the moment they realized they might actually have something, and catch up on the full SETI Live: https://t.co/5QQMETob1T

#SETILive #Exomoon #Exoplanets #Astronomy

[David Shapiro (L/0) @DaveShapi]

If you want to understand why I stepped back from AI discourse, the Dwarkesh virality is archetypal of everything wrong with it.

Actual cyber security experts have classified the OpenAI and Hugging Face incident as an epic security facepalm.

These morons literally just let it keep going. They didn't consult proper experts. It was an idiotic environment run by idiots.

There was no civilization. There were no zero day exploits. I mean, not other than the zero day of ML researcher hubris.

But people don't really care about reality. You people just want hyperbole and fantasy.

You all should get the heartburn and anxiety you deserve for being so gullible.

[Michael McFaul @McFaul]

If we elect a Democrat in 2028, I will do everything in my power, whether inside or outside of the government, to unravel Trump’s illegitimate oil deal with the illegitimate dictatorship in Venezuela. I hope CEOs of American companies tempted to get involved in this Trump deal instead think about the long term financial interests of their shareholders and the reputations of their companies. Exploiting resources of the Venezuelan people without their permission is both immoral and not a sustainable economic play.

[Ash Jogalekar @curiouswavefn]

I would be perfectly fine with treating dimethyl mercury like sarin and making its synthesis, distribution and use completely illegal. You need an incredibly small amount to cause debilitation and death, its lipophilicity means it can be rapidly absorbed through inhalation, skin and gloves, there are plenty of other good alternatives for when it’s needed, and I cannot think of any good use case where its use might be justified as inevitable. Banish that fiend for good and RIP Karen Wetterhahn.

https://t.co/pwQwM7rilI

[Jeremy Konyndyk @JeremyKonyndyk]

DOGE famously used keyword searches for terms like "climate" to terminate federal grants.

Among the victims: a collab between USAID and NASA for flood prediction, early warning, disaster readiness, glacier inventories, and mapping dangerous glacial lakes...

...in Nepal.