I found the Science article refreshing (compared to the writing on AI existential risk). It's clearly-written, educational, non-hyperbolic, free of dubious metaphors, and grounded in mechanistic understanding and real-world knowledge.
And I appreciate, in particular, that the article is just as clear about what _isn't_ dangerous (synthesis of mirror nucleic acids and proteins) as what is (creating mirror bacteria). Plus, for someone interested in pursuing mirror bacteria in the long run, the countermeasure research proposed gives a concrete path to doing that safely.
Totally agree that while it makes sense to restrict mirrored cells, we could (and should) still use synthetic and cell free methods to still synthesize mirror biomolecules. A mirrored ribosome doesn’t have to exist only in a mirrored cell. That could allow us to access some of those mirrored peptide medicines and other potentially more robust mirror biomaterials, without the risk of uncontrolled spread.
Yeah, there are some useful things we can do with certain mirror proteins, and we can make some of them without significant risk, as long as we don't make anything approaching a full mirror proteome. The article and technical report go into more detail on that.
This article and the science article it references are very intresting and thought provoking.
And while the threat proposed is real and do support a moratorium on the research, I do not think our immune systems would be as defenseless as the authors make it seem.
It is one thing to test the immunogenicity of discrete proteins, another to test the same for whole cells. Its like trying to hide a wheel vs a car.
In particular several nonchiral metabolites or structural components mirror life is required to have are likely to elicit an immune response. The lipid bilayer comes to mind. It does have some chirality, but I don't think that will mater as much in the side thats facing any hypothetical immune system. Immune cells are going to look at a long structured wall of lipids and treat it like a lipid bilayer, whether it is mirrored or not. And then like with other pathogens, when immune cells ask it for its ID (MHC) and find its not there, that mirrored cell is going to be in trouble, mirrored or not.
Frankly this article has mainly succeeded in making me more curious to test mirrored life to see what would happen than scared. Even if my more rational self still agrees with the policy prescriptions
It could well be something like this - but it’s unclear which side would have the advantage or disadvantage between a right-handed pitcher that has never seen a left-handed batter, and a left-handed batter that has never seen a right-handed pitcher.
> Even under stringent contamination control measures, terrestrial microbes managed to colonize samples returned from Ryugu. “Transported to Earth in a hermetically sealed chamber, the sample was opened in nitrogen in a class 10,000 clean room to prevent contamination. Individual particles were picked with sterilized tools and stored under nitrogen in airtight containers. Before analysis, the sample underwent Nano-X-ray computed tomography and was embedded in an epoxy resin block for scanning electron microscopy. Rods and filaments of organic matter, interpreted as filamentous microorganisms, were observed on the sample's surface. Variations in size and morphology of these structures resembled known terrestrial microbes.” The researchers recommend further enhancing contamination control procedures for future missions. Life finds a way.
I realize you called for nothing to ever leave the lab, but strange things can happen in the fullness of time (e.g. a collision resulting in contaminated orbital debris which eventually falls to Earth...)
He should only be read as proposing hypothetical possibly sufficient containment mechanisms for a change in policy, not a definitive statement of what should be legally defined as sufficient!
I think it is very important to understand that the greatest gap in our understanding is ecosystems.
all life on earth exists as part of a tree of continuous linked and branching and merging chemical interactions which have been running for billions of years.
let that sink in - all life is different interacting branches of a single network of continuous chemical reactions.
so your dna that makes you human is a small fraction - less than a percent of the information content of your body… most of the life that is you isn’t human and we are just beginning to understand it.
and all those reactions are right handed.
so the idea that we would be able to synthesize a functioning network of left handed continuous chemical reactions which could be sustained and amplified in our right handed world … well it feels victorian doesn’t it? like how one would think before hilbert and godel and einstein and bohr and heisenberg and so many others. how we thought when we thought we were in control and could control and know everything.
to tap stuart kauffman - who is the father of how we should think of progress (adjacent possible)…. how is a left handed system in the adjacent possible?
i’ll just leave it at that and let everyone come to their own conclusions.
All life exists as part of a historical tree of chemical reactions. But not all life exists within a network of contemporary branches of that tree. There are some extremophiles that exist in bits of the world where barely any other life does, and mirror versions of those might be able to start a new evolutionary tree.
I think the expected cost of climate change is far lower than “wiping out all multicellular life.” It might be, say, a trillion-dollar problem. (Contrast with world GDP of ~$100T.) Far easier to adapt to different agricultural patterns and build some levees against sea-level rise, than to fight an invasive bacteria with immune evasion. Also we can capture carbon out of the sky, we can put SiO2 into the stratosphere to reflect extra sunlight, etc. Lots of ways to deal.
On the other side of the equation, climate change activists sometimes call for significantly curtailing energy use—an enormous cost. Whereas here we're talking about a relatively narrow line of biological research with no known major benefits.
So I think the cost-benefit calculation comes out quite differently in the two cases. (Although I still think we should invest in solutions to prevent, mitigate, or adapt to climate change—as long as they don't require reducing energy use.)
Jason, I was using incalculable as "impossible to calculate." You seem to be using it as "so large as to be impossible to calculate the magnitude (but we know the magnitude is larger than climate change) Help me out here.
I don't think the problem with the climate change argument is that the magnitude just isn't large enough to be a concern. It's more along the lines of arbitrarily asserting global disaster from climate change and then using that to inform what we should do today. That's just really poor epistemologically.
I'm interesting in the sources and credibility of them as most of what you describe today is something chemists and biochemists learn early. I have a ton of questions and a lot of skepticism. TO acknowledge that the future state is so far off yet so disastrous simultaneously is problematic to me and uncharacteristic of a technology optimist such as yourself. You haven't stated anything groundbreakingly new understanding in your article that would flip my understanding of the situation.
In a way I would ask "how is this not like the climate catasrophizing that Alex Epstein pokes fun at every day?" I don't think he pokes fun at something that is a debate about magnitudes, but rather an epistemologically bad way of thinking about the issue.
I meant incalculable as in “priceless”, as in how do you put a value on human existence vs. extinction?
Re epistemics: It's not inherently arbitrary to talk about future threats, or to attempt to weight the cost vs. benefit of certain courses of action. We have to assess risks based on what we know.
• Sometimes the science is extremely well-understood, like orbital mechanics: if an asteroid were heading towards us, we would have a pretty good idea of whether it was going to hit us or not.
• Sometimes the domain is poorly understood and the risks are largely speculative, such as with AI. (Which isn't a reason not to think about the risks, or an excuse for not assessing them, just a reality that we don't know much about them and can't say much for sure.)
• Domains such as climate change or biology I think are in between: there's a lot of science to help understand them, but they're also pretty complex systems that are hard to predict. So we have to make estimates.
If you still don't agree, then I'm wondering what exactly you think is bad epistemology here.
Jason, I see how you are parsing these fields and thinking about it. One aspect is that the "speculative" nature isn't just a property of the field, but also of the hypothesis you're working with in the thought experiment. That is how far away from known science is the particular scenario you've posited.
In that sense, you could have a hypothesis in a "speculative" field which is not that far afield of known science which could be plausible. In another scenario, you could have a hypothesis that is very far away from what is currently known in an "in between" or "well understood" field and it does not deserve consideration just because the field is better understood.
The other aspect is whether its a field we have much control over. Asteroids hitting us are possible, but not much to do about them, and bio-engineering or AI are man-made so choices of what to do are important.
What I always find interesting about "catastrophic scenarios" in fields which are primarily human oriented is that they follow a particular pattern. That is sort of ignoring this distinction of how far away from known science you are in characterizing your hypothesis, and then put together a narrative, using elements of known science which seems *plausible,* and forecasts a disaster. The disaster triggers emotional concern, which is usually brought back to the present to prescribe action. This is something I face every day, and the number of relatively smart people who understand aspects of the science and yet still trot out, "but global disaster" argumentation on things like climate change. I face an industry who has all but accepted the scenario and is actively changing their business plans to address it, but will likely end in a big reset as the bubble builds. When you scratch the surface of these folks, it always ends in this "but global disaster" sort of thinking.
In a sense it conflates what is plausible, with possible with probable. And blurs the distinction between the arbitrary and the remote. Even if you wrap this in "long-way-off", "low probability" language it's necessarily valid to continue to work with it in the same way.
In that sense, I don't really react to debates of the impact of climate change vs. the impact of this scenario as being comparable until the scenarios have passed some sort of validation. Climate change for me is classified into a scenario that is hardly worth discussing as a disaster, not because of it's relative magnitude but because of the flawed perspective of it. And I think in a post climate change world where the cultlure has responded to the "but global disaster" flawed thinking, I would be even more skeptical of another group of scientists trying it.
Now, evaluating a scenario takes expertise. I think a higher standard than can I construct a plausible scenario is to then integrate it back into all the other aspects of what the science tells us, which requires expertise. Which is why I asked what is the essential thing that has changed in the recent present which somehow makes this a concern. There's nothing so far in the article above that trips that trigger for me. It is a plausible scenario, but from my perspective not well integrated which leaves me with a hundred questions before I'd even start getting worried.
All is to say, scenarios, and the way they are constructed matter, and until someone understands them well, they should be highly skeptical of them, and especially of those peddling far off disaster scenarios as reasons to stop action today. I know you didn't do this, and not accusing you of thinking this way, but I'm suspicious of the people you cite at the core of this who seem to be doing that. But open to what they know that is new that might change my mind.
I'm certainly not an expert, in the sense that I have a PhD, but I've done professional work in both biotech and chiral chemistry having worked for units that produced such products for the pharma field, and at scale (which is a unique perspective)
How do you feel about warnings regarding the mounting national debt?
Some of those are also disaster scenarios. They're in a complex domain, economics, which can't be predicted precisely. Some people have been warning about them for a long time, continually predicting disaster that never quite arrives. If your epistemic rule is “you don't have to worry about speculative future disasters,” then you would hand-wave away concerns about the debt.
I think we have to judge each case on its epistemic merits. The scientific case for risk from mirror life makes sense to me. There are a lot of unknowns but there is a lot of deep scientific expertise being applied here.
And we always have to make decisions in the face of uncertainty. We can't apply a rule “never act on anything distant or low-probability”, or else we would never buy insurance, or get cancer screening. Instead we have to weigh costs and benefits. Stopping a relatively narrow line of research with no major applications is low-cost. Slowing economic growth worldwide is very high cost. So the epistemic bar should be very different for those actions.
I don't see why this is a threat. A mirror bacteria wouldn't be able to metabolize ordinary organic compounds, and in the competition between the mirror biosphere and the ordinary biosphere, the ordinary biosphere has a several billion year head start.
Re metabolism, this is addressed in the paragraph beginning “In the body, mirror bacteria could feed on achiral molecules…”
The competition point is relevant but not dispositive IMO. Also remember that mirror life would be engineered, which may make up for the natural head start.
Did you read the paragraph? Glyercol can be a source of carbon, and with the right genes, mirror bacteria could even use D-glucose. E. coli have been shown to replicate in growth media with only achiral nutrients.
"On the cost side, future technologies could improve biocontainment. Imagine, for instance, a space-based bio lab, fully roboticized, in orbit around the Sun. Suppose this lab can accept shipments of materials, but nothing ever leaves it. Such a lab, separated by more than 1 AU from Earth, might provide sufficient protection for very dangerous experiments."
Jeez, this sounds like a bad script for an old SciFi movie - scientist warns of dangers, warnings ignored, experiment unleashes disaster, scientist saves the world...
Really? With all the existential threats currently available you are encouraging us to waste time on imagined ones? Focus people, focus...
I’ll be the techno-optimist that doesn’t worry too much about this.
It would be odd if mirror bacteria haven’t spontaneously evolved in billions of years. Seems more likely that they pop into existence every now and then but can’t survive. This is evidence that we probably have little to worry about.
Weren’t there people who didn’t want us splitting the atom in case it caused a chain reaction that destroyed our atmosphere and doomed us all? If we had heeded their warning we’d have given up on an incredibly valuable and important technology, and we may have had a much bloodier second half of the 20th century than we had.
But even if there is some big risk to developing this, much like concerns over ASI, the idea that we can control China or Russia or whoever else and do anything other than slow them down some from developing a technology is foolhardy. You have to assume they’re researching it and putting far more resources into it than you realize, so best to at least be developing the science here in the free world so we have the ability to develop countermeasures.
IMO at least we should be studying and perhaps creating mirror-bacteria hungry bacteriophages.
Re “Seems more likely that they pop into existence every now and then”—this is a good point, and it occurred to me too when learning about this, so I asked Kevin Esvelt about it. One possibility is that abiogenesis is actually extremely rare, and they *don't* pop into existence every now and then. Another possibility is that any cells that form from abiogensis are very weak and can't compete with evolved life. But mirror life would be engineered, possibly as a mirror of an already-evolved bacterial species, or perhaps even genetically enhanced. So we can't count on this.
Re the chain reaction that would destroy the atmosphere: this was a real concern, and some people took it seriously enough to do a computation and assess whether it would happen. They concluded that it wouldn't, and they were right—but they also got a related computation wrong, leading to a larger-than-expected explosion in a bomb test that killed bystanders. We're just lucky that they got that one wrong, and not the atmosphere one. (Scott Alexander summarized that story in his review of The Precipice: https://slatestarcodex.com/2020/04/01/book-review-the-precipice/)
I don't think we can *control* China or Russia, but I do think that if there were a technology which could run wild and destroy almost all life on Earth, then even they would want to be very careful with it. They're not insane, and they don't want to die or destroy their countries.
And re countermeasures, we can develop them without creating mirror bacteria themselves. Yes, bacteriophages might be an example?
Freaky. Thanks for sharing, and agree with your take. If a path is all risk and no reward, then pursuit of that path is silly. I would be interested in a a pro-exploration argument of what there is to gain from studying mirror molecules.
There are some uses for mirror biomolecules that are mentioned in the article/report. But I don't think there's a lot further to gain from creating full mirror cells.
As a techno-optimist, and based on the little I know of this, I agree with your take.
I found the Science article refreshing (compared to the writing on AI existential risk). It's clearly-written, educational, non-hyperbolic, free of dubious metaphors, and grounded in mechanistic understanding and real-world knowledge.
And I appreciate, in particular, that the article is just as clear about what _isn't_ dangerous (synthesis of mirror nucleic acids and proteins) as what is (creating mirror bacteria). Plus, for someone interested in pursuing mirror bacteria in the long run, the countermeasure research proposed gives a concrete path to doing that safely.
Totally agree that while it makes sense to restrict mirrored cells, we could (and should) still use synthetic and cell free methods to still synthesize mirror biomolecules. A mirrored ribosome doesn’t have to exist only in a mirrored cell. That could allow us to access some of those mirrored peptide medicines and other potentially more robust mirror biomaterials, without the risk of uncontrolled spread.
Yeah, there are some useful things we can do with certain mirror proteins, and we can make some of them without significant risk, as long as we don't make anything approaching a full mirror proteome. The article and technical report go into more detail on that.
This article and the science article it references are very intresting and thought provoking.
And while the threat proposed is real and do support a moratorium on the research, I do not think our immune systems would be as defenseless as the authors make it seem.
It is one thing to test the immunogenicity of discrete proteins, another to test the same for whole cells. Its like trying to hide a wheel vs a car.
In particular several nonchiral metabolites or structural components mirror life is required to have are likely to elicit an immune response. The lipid bilayer comes to mind. It does have some chirality, but I don't think that will mater as much in the side thats facing any hypothetical immune system. Immune cells are going to look at a long structured wall of lipids and treat it like a lipid bilayer, whether it is mirrored or not. And then like with other pathogens, when immune cells ask it for its ID (MHC) and find its not there, that mirrored cell is going to be in trouble, mirrored or not.
Frankly this article has mainly succeeded in making me more curious to test mirrored life to see what would happen than scared. Even if my more rational self still agrees with the policy prescriptions
I was hoping to hear that it would be more like right-handed batters finding it easier to face down left-handed pitchers. 😁
Ha, good one, wish I'd thought of that!
It could well be something like this - but it’s unclear which side would have the advantage or disadvantage between a right-handed pitcher that has never seen a left-handed batter, and a left-handed batter that has never seen a right-handed pitcher.
The argument is that the advantage is with the bacteria, since they can invade the bodies of multicellular organisms and replicate inside them.
As a side note: even locating a research lab in outer space might not be enough to keep mirror microbes safely contained. From https://orbitalindex.com/archive/2024-12-11-Issue-297/:
> Even under stringent contamination control measures, terrestrial microbes managed to colonize samples returned from Ryugu. “Transported to Earth in a hermetically sealed chamber, the sample was opened in nitrogen in a class 10,000 clean room to prevent contamination. Individual particles were picked with sterilized tools and stored under nitrogen in airtight containers. Before analysis, the sample underwent Nano-X-ray computed tomography and was embedded in an epoxy resin block for scanning electron microscopy. Rods and filaments of organic matter, interpreted as filamentous microorganisms, were observed on the sample's surface. Variations in size and morphology of these structures resembled known terrestrial microbes.” The researchers recommend further enhancing contamination control procedures for future missions. Life finds a way.
(Original source: https://phys.org/news/2024-11-ryugu-asteroid-sample-rapidly-colonized.html#google_vignette)
I realize you called for nothing to ever leave the lab, but strange things can happen in the fullness of time (e.g. a collision resulting in contaminated orbital debris which eventually falls to Earth...)
He should only be read as proposing hypothetical possibly sufficient containment mechanisms for a change in policy, not a definitive statement of what should be legally defined as sufficient!
I think it is very important to understand that the greatest gap in our understanding is ecosystems.
all life on earth exists as part of a tree of continuous linked and branching and merging chemical interactions which have been running for billions of years.
let that sink in - all life is different interacting branches of a single network of continuous chemical reactions.
so your dna that makes you human is a small fraction - less than a percent of the information content of your body… most of the life that is you isn’t human and we are just beginning to understand it.
and all those reactions are right handed.
so the idea that we would be able to synthesize a functioning network of left handed continuous chemical reactions which could be sustained and amplified in our right handed world … well it feels victorian doesn’t it? like how one would think before hilbert and godel and einstein and bohr and heisenberg and so many others. how we thought when we thought we were in control and could control and know everything.
to tap stuart kauffman - who is the father of how we should think of progress (adjacent possible)…. how is a left handed system in the adjacent possible?
i’ll just leave it at that and let everyone come to their own conclusions.
All life exists as part of a historical tree of chemical reactions. But not all life exists within a network of contemporary branches of that tree. There are some extremophiles that exist in bits of the world where barely any other life does, and mirror versions of those might be able to start a new evolutionary tree.
“This is a simple cost-benefit calculation. On the cost side, the threat is plausible, and the potential damage incalculable.”
Isn’t this the same logic as is levied for climate change? Incalculable damage is arbitrary by its definition.
I think the expected cost of climate change is far lower than “wiping out all multicellular life.” It might be, say, a trillion-dollar problem. (Contrast with world GDP of ~$100T.) Far easier to adapt to different agricultural patterns and build some levees against sea-level rise, than to fight an invasive bacteria with immune evasion. Also we can capture carbon out of the sky, we can put SiO2 into the stratosphere to reflect extra sunlight, etc. Lots of ways to deal.
On the other side of the equation, climate change activists sometimes call for significantly curtailing energy use—an enormous cost. Whereas here we're talking about a relatively narrow line of biological research with no known major benefits.
So I think the cost-benefit calculation comes out quite differently in the two cases. (Although I still think we should invest in solutions to prevent, mitigate, or adapt to climate change—as long as they don't require reducing energy use.)
Jason, I was using incalculable as "impossible to calculate." You seem to be using it as "so large as to be impossible to calculate the magnitude (but we know the magnitude is larger than climate change) Help me out here.
I don't think the problem with the climate change argument is that the magnitude just isn't large enough to be a concern. It's more along the lines of arbitrarily asserting global disaster from climate change and then using that to inform what we should do today. That's just really poor epistemologically.
I'm interesting in the sources and credibility of them as most of what you describe today is something chemists and biochemists learn early. I have a ton of questions and a lot of skepticism. TO acknowledge that the future state is so far off yet so disastrous simultaneously is problematic to me and uncharacteristic of a technology optimist such as yourself. You haven't stated anything groundbreakingly new understanding in your article that would flip my understanding of the situation.
In a way I would ask "how is this not like the climate catasrophizing that Alex Epstein pokes fun at every day?" I don't think he pokes fun at something that is a debate about magnitudes, but rather an epistemologically bad way of thinking about the issue.
I meant incalculable as in “priceless”, as in how do you put a value on human existence vs. extinction?
Re epistemics: It's not inherently arbitrary to talk about future threats, or to attempt to weight the cost vs. benefit of certain courses of action. We have to assess risks based on what we know.
• Sometimes the science is extremely well-understood, like orbital mechanics: if an asteroid were heading towards us, we would have a pretty good idea of whether it was going to hit us or not.
• Sometimes the domain is poorly understood and the risks are largely speculative, such as with AI. (Which isn't a reason not to think about the risks, or an excuse for not assessing them, just a reality that we don't know much about them and can't say much for sure.)
• Domains such as climate change or biology I think are in between: there's a lot of science to help understand them, but they're also pretty complex systems that are hard to predict. So we have to make estimates.
If you still don't agree, then I'm wondering what exactly you think is bad epistemology here.
Jason, I see how you are parsing these fields and thinking about it. One aspect is that the "speculative" nature isn't just a property of the field, but also of the hypothesis you're working with in the thought experiment. That is how far away from known science is the particular scenario you've posited.
In that sense, you could have a hypothesis in a "speculative" field which is not that far afield of known science which could be plausible. In another scenario, you could have a hypothesis that is very far away from what is currently known in an "in between" or "well understood" field and it does not deserve consideration just because the field is better understood.
The other aspect is whether its a field we have much control over. Asteroids hitting us are possible, but not much to do about them, and bio-engineering or AI are man-made so choices of what to do are important.
What I always find interesting about "catastrophic scenarios" in fields which are primarily human oriented is that they follow a particular pattern. That is sort of ignoring this distinction of how far away from known science you are in characterizing your hypothesis, and then put together a narrative, using elements of known science which seems *plausible,* and forecasts a disaster. The disaster triggers emotional concern, which is usually brought back to the present to prescribe action. This is something I face every day, and the number of relatively smart people who understand aspects of the science and yet still trot out, "but global disaster" argumentation on things like climate change. I face an industry who has all but accepted the scenario and is actively changing their business plans to address it, but will likely end in a big reset as the bubble builds. When you scratch the surface of these folks, it always ends in this "but global disaster" sort of thinking.
In a sense it conflates what is plausible, with possible with probable. And blurs the distinction between the arbitrary and the remote. Even if you wrap this in "long-way-off", "low probability" language it's necessarily valid to continue to work with it in the same way.
In that sense, I don't really react to debates of the impact of climate change vs. the impact of this scenario as being comparable until the scenarios have passed some sort of validation. Climate change for me is classified into a scenario that is hardly worth discussing as a disaster, not because of it's relative magnitude but because of the flawed perspective of it. And I think in a post climate change world where the cultlure has responded to the "but global disaster" flawed thinking, I would be even more skeptical of another group of scientists trying it.
Now, evaluating a scenario takes expertise. I think a higher standard than can I construct a plausible scenario is to then integrate it back into all the other aspects of what the science tells us, which requires expertise. Which is why I asked what is the essential thing that has changed in the recent present which somehow makes this a concern. There's nothing so far in the article above that trips that trigger for me. It is a plausible scenario, but from my perspective not well integrated which leaves me with a hundred questions before I'd even start getting worried.
All is to say, scenarios, and the way they are constructed matter, and until someone understands them well, they should be highly skeptical of them, and especially of those peddling far off disaster scenarios as reasons to stop action today. I know you didn't do this, and not accusing you of thinking this way, but I'm suspicious of the people you cite at the core of this who seem to be doing that. But open to what they know that is new that might change my mind.
I'm certainly not an expert, in the sense that I have a PhD, but I've done professional work in both biotech and chiral chemistry having worked for units that produced such products for the pharma field, and at scale (which is a unique perspective)
How do you feel about warnings regarding the mounting national debt?
Some of those are also disaster scenarios. They're in a complex domain, economics, which can't be predicted precisely. Some people have been warning about them for a long time, continually predicting disaster that never quite arrives. If your epistemic rule is “you don't have to worry about speculative future disasters,” then you would hand-wave away concerns about the debt.
I think we have to judge each case on its epistemic merits. The scientific case for risk from mirror life makes sense to me. There are a lot of unknowns but there is a lot of deep scientific expertise being applied here.
And we always have to make decisions in the face of uncertainty. We can't apply a rule “never act on anything distant or low-probability”, or else we would never buy insurance, or get cancer screening. Instead we have to weigh costs and benefits. Stopping a relatively narrow line of research with no major applications is low-cost. Slowing economic growth worldwide is very high cost. So the epistemic bar should be very different for those actions.
I don't see why this is a threat. A mirror bacteria wouldn't be able to metabolize ordinary organic compounds, and in the competition between the mirror biosphere and the ordinary biosphere, the ordinary biosphere has a several billion year head start.
Re metabolism, this is addressed in the paragraph beginning “In the body, mirror bacteria could feed on achiral molecules…”
The competition point is relevant but not dispositive IMO. Also remember that mirror life would be engineered, which may make up for the natural head start.
Achiral molecules doesn't include sugars.
Did you read the paragraph? Glyercol can be a source of carbon, and with the right genes, mirror bacteria could even use D-glucose. E. coli have been shown to replicate in growth media with only achiral nutrients.
"On the cost side, future technologies could improve biocontainment. Imagine, for instance, a space-based bio lab, fully roboticized, in orbit around the Sun. Suppose this lab can accept shipments of materials, but nothing ever leaves it. Such a lab, separated by more than 1 AU from Earth, might provide sufficient protection for very dangerous experiments."
Jeez, this sounds like a bad script for an old SciFi movie - scientist warns of dangers, warnings ignored, experiment unleashes disaster, scientist saves the world...
Really? With all the existential threats currently available you are encouraging us to waste time on imagined ones? Focus people, focus...
I’ll be the techno-optimist that doesn’t worry too much about this.
It would be odd if mirror bacteria haven’t spontaneously evolved in billions of years. Seems more likely that they pop into existence every now and then but can’t survive. This is evidence that we probably have little to worry about.
Weren’t there people who didn’t want us splitting the atom in case it caused a chain reaction that destroyed our atmosphere and doomed us all? If we had heeded their warning we’d have given up on an incredibly valuable and important technology, and we may have had a much bloodier second half of the 20th century than we had.
But even if there is some big risk to developing this, much like concerns over ASI, the idea that we can control China or Russia or whoever else and do anything other than slow them down some from developing a technology is foolhardy. You have to assume they’re researching it and putting far more resources into it than you realize, so best to at least be developing the science here in the free world so we have the ability to develop countermeasures.
IMO at least we should be studying and perhaps creating mirror-bacteria hungry bacteriophages.
Re “Seems more likely that they pop into existence every now and then”—this is a good point, and it occurred to me too when learning about this, so I asked Kevin Esvelt about it. One possibility is that abiogenesis is actually extremely rare, and they *don't* pop into existence every now and then. Another possibility is that any cells that form from abiogensis are very weak and can't compete with evolved life. But mirror life would be engineered, possibly as a mirror of an already-evolved bacterial species, or perhaps even genetically enhanced. So we can't count on this.
Re the chain reaction that would destroy the atmosphere: this was a real concern, and some people took it seriously enough to do a computation and assess whether it would happen. They concluded that it wouldn't, and they were right—but they also got a related computation wrong, leading to a larger-than-expected explosion in a bomb test that killed bystanders. We're just lucky that they got that one wrong, and not the atmosphere one. (Scott Alexander summarized that story in his review of The Precipice: https://slatestarcodex.com/2020/04/01/book-review-the-precipice/)
I don't think we can *control* China or Russia, but I do think that if there were a technology which could run wild and destroy almost all life on Earth, then even they would want to be very careful with it. They're not insane, and they don't want to die or destroy their countries.
And re countermeasures, we can develop them without creating mirror bacteria themselves. Yes, bacteriophages might be an example?
Freaky. Thanks for sharing, and agree with your take. If a path is all risk and no reward, then pursuit of that path is silly. I would be interested in a a pro-exploration argument of what there is to gain from studying mirror molecules.
There are some uses for mirror biomolecules that are mentioned in the article/report. But I don't think there's a lot further to gain from creating full mirror cells.
Thought you were describing what’s happening on the Internet, initially. Nature, like the Internet, will route around this issue.