Wherein I Learn the Secrets of the Masons

The property management company that managed our rental had become intolerable, overbearing, and quite possibly corrupt. As they have a near-monopoly on Marin rentals, we made the possibly financially, certainly socially, unwise decision to try to find a house we could buy. As luck would have it, interest rates were at what everyone thought back then would be their peak, so prices were down and we managed to find something. The previous owners had been there for 50 years, so to say it needed some work would understating things a bit. Also lucky for us, we had no idea how much work the work would be, and so this summer I decided to turn the rocky hillside in front of the house into terraces that we could plant a garden in. I won’t go into all the details, but I would like to share how I uncovered the secrets of the masons while building a stone retaining wall on that hillside.

Continue reading →

Science Communication and Epistemics

There’s no escaping the fact that the world is becoming more complex, hard to understand, and science-y. There’s no escaping the fact that we all are deluged with more information that we can process, much of which is tailor-made by the algorithms we’ve trained to understand us so well, to compel us to pay attention to it. What is a person to do, caught between the increasing difficulty of understanding the world and the increasing difficulty of making space to even try?

Two examples:
Over the past year, stories have permeated my media bubble about the environmental impacts of AI, particularly the water they use to keep the chips from melting under the strain of all that inference. I’ve also heard about cultivated meat and how energy-intensive it is to produce. The question I had reading about these was first, “To what extent might this be true?” and “If so, so what?”. For the first one, just to set the tone, there were many, many articles. 50000 English-language news articles in the first half of 2026. Many of these repeated the claim that a single prompt required 50 mL of cooling water, and some exaggerated it to 500 mL. Google initially estimated 0.26 mL. Andy calculates it to be 1.36 mL.

From 2023 – 2025, there were hundreds of articles about the CO2 impact of lab-grown meat, most of them claiming the impact was 10x farmed beef. To what extent were these true? It’s hard to say. Was the inflated estimate due to an extrapolation of pilot experiments, when mass production would likely be much more efficient? That didn’t stop several states with big agricultural lobbies from banning it.

Where does that leave us? With an increasing need for skilled science communicators. With a need for people that understand it doesn’t matter what you say if no one hears it. With a need for people who understand how platforms work, what they promote, and how to cultivate a community of people, given these adversarial conditions, who want to understand the world as it is, not how they wish it were.

You’ve heard of vibecoding, now meet spitecoding

Thermoworks makes great hardware, but terrible software. Periodically, I would ask them why the temperature plots of data from their wireless thermometers have to look like this:

They would tell me that the app can’t get data from the temperature probe while the phone screen is off. It made it hard to estimate the remaining cooking time without leaving my phone screen open the whole time, but I’m not a software developer, so what could I do?

I would ask why plots sometimes look like this:

It’s hard to tell if the temperature reading is just noisy or if something’s going on, but I’m pretty sure the internal temperature of a brisket doesn’t jump up and down several degrees in a span of a minute or two. I would ask if it would be possible to get just a dot plot of the data, without a line connecting the points. They would persist to such an extent in failing to understand the request, across multiple support threads and representatives, that I suspected I was part of some postmodern “Who’s on first?” routine. Sometimes they would tell me to use the webpage of their cloud service instead of the app.

One rainy night, when I was smoking a brisket and annoyed by gaps in the plot, I thought of checking back in with them to see if they were planning any updates. I found the last thread, re-read it, and thought, “I hear Claude is pretty good at reverse engineering. I wonder if it can just grab the data over Bluetooth, I can make my own damn plot, and I can keep stuff out of someone else’s cloud, too?”

About 30 minutes later…

Thermoworks, if you’re reading this, you may want to ask your app developers to request a wakelock. I don’t know exactly what that is, or if that’s the solution, but it is in fact possible to grab data while the screen is off. Hope this helps someone.

Repo: https://github.com/williamgunn/localthermo-android

AI-generated systematic reviews – are they possible?

AI-assisted research is an exciting prospect – no more laborious keyword searches, scanning abstracts, and organizing PDFs – but do they live up to the promise? I explored this question using Elicit’s Systematic Review tool. I found a Cochrane Review on a topic and then asked Elicit the same question. I compared the two on retrieval of relevant studies, selection of studies to include, extraction of data, assessment of risk of bias, synthesis of evidence, and assessment of evidence quality. I found the report said, qualitatively and approximately, a similar thing for some primary and secondary outcomes as the Cochrane review. The Elicit report could be used by an expert to reach a similar high-level understanding as they would have gotten from the Cochrane review, but the most important between-study comparisons were not made within the report. The tool was not able to assess included studies for risk of bias, pool data across studies, or assess evidence quality. Differences were also found in study selection and detail reported for key findings.

Continue reading →

AI and Bioterrorism: Risks Explained

How AI can be used for bioterrorism

Humanity has climbed to the top of the food chain, but as the pandemic demonstrated, it’s a precarious perch. It’s not just directly pathogenic viruses that we need to worry about, either. Small molecules and proteins synthesized by bacteria also pose a risk to life, to crops and livestock, and to the environment. These risks would be realized by a terrorist through genetic engineering, the means by which new capabilities are introduced to an organism through changes in its DNA. This is a time-consuming and laborious process that can be vastly accelerated with AI and biological design tools. A terrorist might design a novel virus that causes human illness, one which kills livestock, makes soil infertile, or fouls the ocean.

How dangerous are current AI systems?
Continue reading →

Where’s the art in artificial intelligence?

“In the beginning was the Word, and the Word was with God, and the Word was God.” – John 1:1 KJV

When I think about creative works generated by generative AI, I think about golems. According to legend, Rabbi Yehudah Loew created a automaton from clay to protect the Jews of Prague by animating a soulless lump of clay with the Word of God. Like the Rabbi of folklore, when we write or produce a work of art, we like to think we’re doing something that imbues our work with soul. People say an image created by GenAI lacks the “creative spark”, but what does this really mean? What distinguishes an illustration by Sendak or DorĂ© or Geiger from a similar work created by prompting Midjourney?

Imagine Hugging Face puts on an event where they put a generated illustration next to a work from a famous illustrator and ask an art critic to point out differences. The art critic will get deep in the weeds and point out various differences, then they’ll ask another art critic, who’ll emphasize different distinctions, then another will identify still others as crucial and then Hugging Face will say, “Aha! The critics couldn’t reproducibly find differences and therefore there really are none!” Lots of breathless media headlines and dunking on critics will ensue, unless…
Continue reading →

Do we really want a global town square?

In 2005, Lawrence Summers, president of Harvard, former US Secretary of the Treasury and Chief Economist of the World Bank, gave a speech at NBER where he discussed some employment data. He said he was offering a positive, observational view and not a judgemental or normative view.

…the data will, I am confident, reveal that Catholics are substantially underrepresented in investment banking […], that white men are very substantially underrepresented in the National Basketball Association, and that Jews are very substantially underrepresented in farming and in agriculture. These are all phenomena in which one observes underrepresentation and I think it’s important to try to think systematically and clinically about the reasons for underrepresentation.

He thought that these prefatory statements, and that he was attempting to be a little provocative and not speaking on behalf of Harvard, would have opened up a thoughtful discussion of various factors leading to under-representation of women in tenured positions in science at top universities. He could not have misjudged his audience more poorly.1 Continue reading →