Lantern Bioworks says they have a cure for tooth decay. Their product is a genetically modified bacterium which infects your mouth, outcompetes all the tooth-decay-causing bacteria, and doesn’t cause tooth decay itself. If it works, it could make cavities a thing of the past (you should still brush for backup and cosmetic reasons).
Is this true?
The options in this market are based on Scott Alexander's latest post about Lantern Bioworks, in which he says:
My real opinion, as precisely as I can express it, is:
Advance of approximately the same magnitude as fluoride: 5%
Good on balance, comparable to other beneficial dental treatments: 35%
Doesn’t work in its current form, but could easily be modified into something that does: 10%
Doesn’t work at all and never will: 50%
Causes minor side effects for some people, same scale as Tylenol: 30%
Causes medium side effects, same scale as tricyclics: 5%
Causes disastrous side effects, same scale as thalidomide: <1%
These possibilities are all independent options in this market. If I am personally convinced that an option is true/untrue and am certain enough that I would make a real money bet on it at extreme odds, that option resolves YES/NO here accordingly.
So this market depends on my judgement, but notably I am a longtime reader of Scott Alexander and I trust him quite a bit. So if Scott writes a post saying that that these things are true or untrue, I am very likely to believe him. If you can convince Scott, or Zvi for that matter, you have likely convinced me.
I am very skeptical by default of anything claiming to be as amazing as Lumina, but I think that even a small chance that it is that good makes it well worth pursuing. I have previously traded in some Lantern Bioworks markets, but after making this market I intend to buy no more shares in those markets. I will not buy any shares in this market.
I will try to check in on the latest evidence regularly, and if it seems like we may never know the truth of an option I may decide to resolve it N/A.
I am prepared to wait several years for good data if necessary, but my hope is that some of the more extreme scenarios here will be resolvable sooner than that. The resolution date will be moved out as necessary.
Related questions
Mom: “We have lantern bioworks at home.”
The Lantern Bioworks at home:
Why didn't you make a regular multiple-choice market (not independent) for the first 4, which are not really independent? (Scott had the probabilities add up to 100 anyway, so it seems like he was going for only one being true).
You could then make a second market about side effects, and include the lower 3, with the option of "No significant side effects" filling the last slot.
@ShadowyZephyr I considered that, but I wanted all the options in one place. Also, with loans going away I think unlinked markets where we have to do our own arbitrage might now be better in many cases where we want to be able to have early resolutions.
If it makes hangovers '2x worse' in 10% of people who take it, would that count as a tylenol side effect? a medium side effect?
(i don't think it makes hangovers worse)
@jacksonpolack Hmmm I'd want to hear some other opinions, that sounds worse than tylonel but not as bad as as tricyclics? My first instinct is that this hypothetical sounds more medium than mild, though.
How much modification is necessary?
I'm pretty confident there's some oral genetically engineered organism that would mostly prevent cavities. The question is just the distance between here and there and whether or not "work" requires reducing 50% of cavities or preventing all cavities.
@jacksonpolack I wouldn't say the bar for "work at all" is as high as a 50% reduction in cavities, it's in contrast to the "comparable to other beneficial dental treatments" which seems to encompass everything beneficial but less beneficial than fluoride, and without terrible side effects.
@jacksonpolack I took his phrasing there as referring to the more common side effects of taking Tylenol as recommended. For example, Drugwatch says:
Tylenol, also known as acetaminophen and paracetamol, is generally safe at recommended doses. Nausea is the most common mild side effect, affecting approximately 34% of users. Vomiting is the second most common side effect after nausea, impacting 15% of studied users. Approximately 5% of people in clinical trials reported vomiting.
I suppose I am also curious what happens if you bathe in Lumina and drink it until you're full, but that would be a separate question.
@Joshua are the first 4 and last 3 options each mutually exclusive with each other?
The first 4 seem definitely so?
The last 3 seem like they could overlap in one direction?
@benshindel Yes, the first 4 seem mutually exclusive to me. The last 3 shouldn't be thought of as exclusive, particularly because I wouldn't wait to resolve them all together. If I'm certain enough to resolve one of them but not the others, I'd resolve that one first by itself.
@Joshua If something has a moderate side effect, does that resolve both “mild” and “moderate” options as YES? Or does it need multiple side effects of which at least one is ”mild” and at least one is “moderate”? What if the side effects seem like they’re AT LEAST “mild” so you resolve that and then turn out to be more serious?
@benshindel I think in most cases when the more extreme options about side effects resolve yes, the less extreme options would also resolve yes. As currently worded that isn't necessarily true, but maybe I should edit the options to say "at least mild side effects" and "at least medium side effects?
@Joshua I support the change to "at least" just to make it less confusing to resolve, in case of scenarios where, say, there are "medium" side effects for many that present as "mild" side effects for most... or something like this
@robm Scott's? I don't know that calibration is exactly the right thing to talk about, but I trust his judgment a lot, yes.