Anton Rodenhauser
Posts 5
Comments30
Maybe some of Richard Ngo's fiction writing? I like "Succession" best: (1) Succession - by Richard Ngo - Narrative Ark
Or some of his non-fiction. E.g. (1) Techno-humanism is techno-optimism for the 21st century (mindthefuture.info)
Simce you asked for more resources: Scott Alexander: "Why NOT slow AI progress" https://astralcodexten.substack.com/p/why-not-slow-ai-progress
Why am I not discouraged by poor outcomes in FMT studies.
FMTs look very promising, but there is by no means a clear scientific consensus on their efficacy. It’s an active area of research with many studies currently under way.
And while many scientists are very excited about FMTs, many studies actually yield very mixed results - with poor or no results being common. Anecdotal evidence on FMTs is also often not very convincing - with many recipients reporting no benefits.
However, I'm personally not discouraged by this, and indeed still excited about FMTs with good donors. The reason is that in my impression the vast majority of FMT studies and FMT self-experiments are flawed. Here's why I believe this:
1. Poor donor quality.
Typical donor criteria you see in most FMT studies are severely deficient. The focus seems to be on choosing merely not sick donors, when it really should be about choosing exceptionally healthy donors with fantastic microbiomes. Most studies donor criteria’ are along the lines of "no antibiotics in the past 3 months, 18-50 yrs old, no viruses, infections, parasites, etc. in stool & blood test". In short, if you are an overweight, depressed smoker who regularly eats fast food, you often still qualify as a donor in these studies. And even those studies who do actually try to find good donors that are actually healthy and fit, often don't go nearly far enough - not least just because it is just extremely hard to find exceptionally good donors.
Find me a well-done study that uses donor criteria as described here and takes into account the points below - and if that study still shows only underwhelming results, then I’ll change my mind on this.
2. Oxygen exposure.
Your colon is an anaerobic environment, i.e. very low in oxygen. Thus, most bacteria and even more of the bacteriophages (which might be especially important for FMT efficacy) in stool are anaerobes. However, the vast majority of FMT studies does various processing like filtering, blending, etc. - all of which heavily oxygenates the stool sample, killing anaerobes, and thus very likely reducing efficacy. No wonder these studies show underwhelming results! What you want to do is to immediately vacuum seal the donated stools and suck the air out, and then do all the processing in an oxygen-free environment.
3. No fresh stool, but frozen
Freezing likely kills many of the bacteria and bacteriophages in the stool - very likely greatly reducing its efficacy. Ideally you'd use fresh stool that has been kArticlesept in a de-oxygenated vacuum bag. This is hard to do on a bigger commercial scale, but totally possible if you have your own donor living in the same city like you for DIY FMTs.
4. Inadequate treatment protocols
For example:
a) Insufficient treatment length. Many studies only do a single infusion, but for many people/conditions you might need to do it daily for 2+ months (Eg: ASU autism study).
b) Colon-only procedures. The small intestine is very important, so completely ignoring it is a likely flaw for some conditions. When doing FMTs, I'd always combine both the rectal and oral route for FMTs.
c) FMTs done in isolation, instead of synergistically with other treatments. It might well be that for some diseases, even well done FMTs from great donors might not be sufficient on their own, but when combined synergistically with many other positive microbiome interventions, they turn out to be the key ingredient to push the needle over the edge. Yet most studies don't test for this.