The intention seems legit, despite the concern. Quote from the article:
The pathogen to be modified must pose a serious health threat, and the work must produce knowledge — such as a vaccine — that would benefit humans. Finally, there must be no safer way to do the research.
Also, it's regulated and overseen by a review board:
> "Now, a government panel will require that researchers show that their studies in this area are scientifically sound and that they will be done in a high-security lab."
It does sound like there are concerns. I'm glad that people are keeping their eyes on it. I found this quote interesting as well, regarding publishing the results of such research:
> “Physicists long ago learned to distinguish between what can be publicly available and what’s classified,” he added, referring to nuclear weapons research. “We want to keep some of this stuff on a need-to-know basis.”
The problem with the US keeping their eyes on it is that you can't trust them. All you need is a few lobbyists and rules and regulations seem to be just words. Hell, Net Neutrality had plenty of people "keeping their eyes on it" too.
If non-reputable agencies wanted to do illegal research, there is literally nothing that would stop them. Look at NSA spying, MKUltra, and the failed nerve gas test that killed hundreds of sheep.
The US still has the largest arsenal of nerve gas in the world, much of it which is still in the queue to be destroyed, much of it still leaking.
Here is the original documentary I watched. There is more information on the individual sites on more recent news and magazine sites, often referenced on wikipedia:
There was a good NBC dateline documentary about it years ago. The biggest stockpiles were being dismantled slowly, but they were indeed leaking, and it was a gargantuan project to neutralize the weapons and containers full of VX. I'm not sure where we're at now. I'll look it up later when I'm home.
I'm not sure why this is being downvoted, other than that it isn't precise. I'm general you can't trust regulators or legislators at the federal level.
The regulatory agencies responsible for ensuring compliance with safety and health standards in the USA have a rich history of being cozy with industry (see the fallout after the gulf oil spill, for example). These agencies are overseen by political appointees who are not reliable, and may even oppose the mission of the agency (see the current EPA). Creating an approval channel creates a path to false- positive legality that I don't think should exist. There's a substantive distinction between keeping the whole enterprise strictly verboten and creating an exploitable legal grey area.
I'd prefer if drilling for oil in the gulf were illegal, not legal with regulatory oversite, for example. I don't think a pessimistic outlook on the reliability of federal oversight is unwarranted, given my (admittedly dim) view of how the process has worked in the past.
To be more constructive: I'm up for substantive suggestions for ensuring more transparent and reliable oversite than exists today.
If you "cannot trust regulators or legislators at the federal level" how can you trust the laws? Who writes them, and who enforces them, if not legislators and regulators, respectively?
Legislators tend to not write the laws anymore. They just take the copy the lobby groups have prepared in advance and submit it as a bill. Congressmen don't even bother to read the laws they vote for.
Right but they're not saying they don't trust regulators and legislators as a concept, they're saying they don't trust them when they're basically writing what lobbyists tell them to. It's right they don't trust the law, but their criticism isn't of law and regulation. It's of lobbyists.
You're right that that can be an issue, and one raised in the article. I intended that particular phrase (which admittedly could have been written better) to mean that others outside of the government are watching the process, such as those commenting in the article. Definitely room for improvement, but I don't think this is something that's being done lightly. This whole thing concerns me, but like with many such topics, it's often more complicated than what is summed up in the headline.
I have no serious concerns with the US' good intentions, nor their ability to properly manage such a program.
What I fear, however, is that this action undermines the Biological Weapons Convention[0].
The convention explicitly allows for research such as this, yes. Nonetheless, it effectively worked as a moratorium on such research, with the US leading by example.
This program creates a situation much like nuclear research, where civil and military use cases exist on a continuum, with the first often being used to cover the second. Note how biological agents, where the US had the moral high ground, have entirely disappeared. Meanwhile, nuclear and chemical agents, which can be produced under the cover of closely related civil industries, are still very much threatening (and killing) people today.
Getting a bit off topic here, but I've always interpreted that aphorism differently, and apparently the history of the saying agrees with me.
I've always interpreted it as being a corollary to "actions speak louder than words". Simply saying something, or having the intention to do something does not mean it will actually get done.
I just assumed it meant both. On the off topic though, I've always liked: "Words without actions lack credibility; Actions without words lack clarity." - John Stott.
A negative result is also considered knowledge: rejecting or disproving the null hypothesis is usually valuable. Tom Scott just did a video on it https://www.youtube.com/watch?v=KOJnOCdBr4s
So a bioweapon might not satisfy this criteria. Besides it probably violates other laws, and is thus still illegal.
This ban that's being lifted stems from a few incidents, but really the driving force was Ron Fouchier's [0] gain-of-function study where his lab made a few mutations that enabled transmission of the bird flu (H5N1) virus between ferrets. All they had to do was make five mutations and it was suddenly more dangerous (here's the paper [1]). Ferrets may sound random, but they can become infected with human flu virus and show similar pathology to humans - they're a model for humans.
I remember hearing Fouchier speak in Washington at a biodefense conference right after the study came out. He was (is?) pretty much a celebrity among scientists, but the policy folks had a different opinion...
Pretty sure these things will escape from the lab at some point. I've personally handled some B. anthracis (causes anthrax) that was supposedly non-viable only to find out later that it was almost certainly still alive. (Not this case, but here's an example [2] mentioned in the article.) There would (hopefully) be more regulations on BSL3+ organisms like the viruses being discussed, because anthrax is relatively friendly, but like any other field, all it takes is one incompetent person.
You know how when you watch movies and because you know so much about tech you can tell that most of the stuff done in movies regarding hacking etc is bullshit and not real.
Now I have a lack in knowledge of biology at this level, but can someone maybe explain to me how all those disease mass extinction movies cannot be real? I feel like my lack of knowledge here causes my concerns when reading stuff like this article.
Thing is, some of the stuff I used to laugh at as bullshit can be done now.
Use the enhance feature to magically unblur images to clarity? I've seen ML demos of stuff like that.
Giant realtime database of all phone calls meta data and recordings? Realtime facial/license plate recognition search of millions of people/cars? Not so funny anymore.
Tracking peoples movements in realtime. Sure just cary this phone wherever you go.
Realtime overhead imagery of an entire city? There's a company that flies drones around that lets them rewind the movements of all cars in the city throughout the day...
Magic black box that can decrypt all encryption? Sure, funny only if you overlook underhanded C planted by the NSA.
Hacking into industrial machinery and power systems?
Having millions of computers all under your control to launch attacks?
This was all funny Hollywood BS and conspiracy theories two decades ago. Elaborate animated "hacking" scenes are still funny but what they accomplish is possible.
Yup. Watch_Dogs-style hacking of CCTV cameras? As the other story on HN's frontpage today will tell you (and so will any experience with Shodan), it's totally doable in theory - the only practical reason this doesn't happen is that every camera is broken in a different way, and as a protagonist, to hack them in a field-actionable way, you'd have to have a hundred people in your tech support staff, who manually engage appropriate exploits based on your requests.
I mean that's basically all hacker suites in a nutshell isn't it? A big collection of various tools and exploits for various problems, utilizing the right exploit for the right problem. That's the big thing in Watch_Dogs, you have a pretty big hacker group working together to find exploits and then bundling it together into a suite that the main character can control from his phone. I'm sure with the proper scripting and metadata you can make a DB of CCTV exploits for various popular brands and automatically apply that exploit on the target based off it's brand. It's just a matter of doing all that work.
You'd be surprised by how many of them use the exact same tech, rebranded. For many of them, everything about the server is the same (HTTP headers, directory structure, default logins, etc) but they'll display a different logo and maybe some different CSS.
Maybe the problem is that my sample size isn't large enough, but I've only encountered about a dozen widely used DVR servers that are actually unique tech instead of rebranding.
One aspect such movies almost always get wrong is ease of protection: With only modest countermeasures (face coverings, washing hands, ventilating rooms etc), you can drastically cut down the rate of transmission. A scientist with advanced equipment and knowledge getting infected is extremely unlikely. In the real world it happens, but only to 0.x%, and only to hospital staff spending weeks with thousands of infected people, under often sub-optimal conditions.
It's completely possible to go years without catching a cold if you're willing to devote maybe 10% of your time and a little money to it, and avoid certain situation. We just don't bother because getting a cold isn't (usually) that terrible.
> It's completely possible to go years without catching a cold if you're willing to devote maybe 10% of your time and a little money to it, and avoid certain situation
You picked my curiousity! What do you have in mind?
Ebola is one such virus. In terms of evolution these viruses tend to burn out by killing the host. Evolutionary pressures push viruses towards symbiotic equilibrium. Thus in the worst case scenario there will be survivors who are immune or the virus or the virus will burn out until the only strain left does not harm humans. The later was the case with ebola.
If anything Ebola is too lethal. It tends to burn out locally before it can spread far from its originating point though if someone should immediately travel after becoming infected it could spread to a new location. But in the majority of the cases so far it has burned out before it could spread to a new village, though the largest outbreak to date claimed in excess of 10,000 lives and spread quite far.
All it will take for Ebola to be a truly massive killer is a small mutation that would reduce immediate mortality or that would increase the incubation time. Either one of those would greatly increase the number of people that a single outbreak would affect. I'd rather not have to live through that.
Counter-intuitively the speedy kill is actually what makes Ebola comparatively safe, if the victims would live longer they would infect far more others though even in death there is plenty of risk in handling the bodies during burial, especially in remote locations.
I understand that - I'm just pointing out that while Ebola perhaps isn't putting humans at extinction risk, it's still not good by any measure. We (I) don't want anything remotely like it lying around at anyone's disposal.
Glad I'm not the only one who feels this way. I'm also glad that I'm aware to some extent of how what I don't know might obscure my judgement - I think a lot of people don't get exposed to the amount of cross-discipline junctures in knowledge that people in tech do.
Now, I don't want to seem anti-science or anything, but this seems like a massive disaster just waiting to happen, in light of the incidents that led to this ban and also, even more to the point, the 2001 anthrax attacks.[1] Couldn't they at least restrict this research to labs far away from major population centers?
The analogue with software development, I'd say, is when people periodically point out that nuclear power plant or ICBM control systems still run on 8-inch floppy disks!!1! and therefore allegedly need to be completely rewritten. That rationale likewise fails to meet my tempting fate threshold.
But the massiveness of that disaster, or its impendingness, has very little to do with some people in Washington DC mouthing words about how you shouldn't do that.
This is definitely one of those cases where I feel like a lot of people are demonstrating some very magical thinking about the effectiveness of regulations, with all due respect to those involved. Does Kim Jong Un care that some people in Washington DC say that it's illegal to make lethal viruses? Does China? Russia? Iran? Europe? Anyone else in the world, really? Or a small criminal lab located in the United States? Or a criminal with access to a lab in the United States? It's not like there's literally people standing on top of every single gene sequencer in the world, or the US, stopping people from doing things.
The regulations were not all that was standing between us and some deadly virus, because anyone who would make a deadly virus and release it, accidentally or otherwise, still has a very simple path to that goal: "Ignore the regulations the US has." This just gives the good guys a chance to possibly learn some things about that before someone else releases it.
That's not the point of restricting research, though; the point of it is restricting research done by labs which are not bent on using biological weapons on human victims but allow them to fall into the wrong hands, or carelessly cause an incident with improper handling of the materials.
It's possible that it's very difficult to engineer a virus that can do better than anything nature has come up with. Evolution has had billions of years to try all the attack vectors. A worst case 12 Monkeys-style disaster seems unlikely.
But I'm ignorant of this topic. Without handwaving or pointing at the worst case, how much should we worry? Is there any evidence that a genetically enhanced virus can be devastating? https://en.wikipedia.org/wiki/Genetically_modified_virus doesn't have much.
> It's possible that it's very difficult to engineer a virus that can do better than anything nature has come up with. Evolution has had billions of years to try all the attack vectors.
The traits desired in a bioweapon aren't particularly beneficial to the long-term propagation of a viral strain and thus would be selected against, or at least not selected for, by natural processes.
As a general rule, if your argument starts with this premise, you're probably heading down the wrong path. Natural evolution is an extremely limited optimization process with a dramatically different loss function than the ones used by humans.
"It's possible that it's very difficult to engineer corn that can resist pests better than anything nature has come up with. Evolution has had billions of years to try all the attack vectors. Widespread commercial use of genetically modified corn seems unlikely."
> It's possible that it's very difficult to engineer a virus that can do better than anything nature has come up with
A lot of the debate was started because 2 groups did so in 2011, independently of each other finding mutations that allow the deadly H5N1 virus to transmit via air[1][2]. One group also did work to model the chance of these mutations showing up in nature[3]. The techniques by which to do this are fairly well established. Knowledge of how to do it safely is fairly limited, as is access to animals in which to test these modifications.
Humans can suffer very high death rates due to diseases for which we aren't a host. See the Black Death [1]. The host was fleas, riding rodent vectors.
I don't have a source at the moment, but there's evidence that infectious agents become less deadly over time. There seems to be evolutionary pressure to not kill hosts.
> It's possible that it's very difficult to engineer a virus that can do better than anything nature has come up with
As someone finishing a PhD in computer-aided drug discovery, this seems to be the situation in a lot of cases. Nature often finds it trivial to get around drugs and back to the viral behavior it wants. For example, a lab might make an allosteric drug that reduces the activity of a viral enzyme. In response to treatment with the drug, a few copies of the virus emerge with point mutation that makes them more active. Now a patient has two populations of viruses, and the disease no longer cares about the treatment. The surprising thing isn't just that this happens, but that it happens so rapidly, often requiring just a few days or a single mutation.
A quick example of the above is the I585T substitution in [1]
One big-picture concept is that genes and proteins are more than their immediate sequences/transcripts. On top of their current identity, they all exist on an mutational "landscape". Remember that some codons are redundant, while some amino acid changes would require two or three base mutations. Therefore, the sequence of a virus encodes both its current identity, as well as nearby evolutionary "alternatives". When a recurring threat comes by, the virus might have encoded in its genes a common defense, maybe through a particularly mutagen-exposed gene location or something like that. In the above case, the virus may have had millenia of exposure to weird dietary natural products and whatnot, and encoded easy rate up/down mutations for its essential enzymes, or orthosteric disruptor mutations for its promiscuous binding sites.
This sort of fuckery will have to become a consideration in designing the antivirals of the future.
And, to toot my own horn: In an evolutionary sense, our fight against viruses is particularly interesting. Now that computers are helping us accelerate drug development, our brains and computers could be considered as part of our immune systems. In the same way that metal weapons made us safe against macroscopic threats (lions etc), computers and public health infrastructure will become our arsenal against microscopic threats. If computer-aided drug design can lead to faster approval of cheap drugs, we can design hundreds of antiviral molecules and chase a virus all over its evolutionary landscape, until it has nowhere left to run.
For out-of-field readers -- An anecdote of cancer cells (not viruses) becoming reliant on a drug in [2]
I believe there is reason for concern w.r.t. modifications of flu viruses, for one thing. That is, they know what makes flu easily transmissible among humans, and they know a bit about how to alter nontransmissible viruses to be more so. Given the belief that things like the 1918 flu are believed to be viruses that made this jump on their own, it's quite reasonable to worry that we could do it in a lab.
Your first sentence is true, but ignores the fact that scientists aren't creating viruses from scratch. They are using what nature made and modifying it. Many treatments rely on a viruses ability penetrate cells. Why recreate what already works so well?
If I recall correctly, there is one gene from a particularly lethal virus can be transferred to the flu to make an incredibly deadly nightmare. I don't know why anyone would do that, but it's possible. IIRC this has already been done in a lab. Trying to do the same but targeted seems like something certain types of people would try to do and could easily go awry.
I don't trust people's due dilligence enough to be okay with this.
Edit: remember when Ebola was never supposed to leave the CDC fascility and a nurse was exposed to it and left the hospital? That's what happens when there's the highest level of scrutinity and security.
They have a strong desire to match american machismo. They probably lack the funds to build novel biological weapons their own. They are extremely good at break into computer systems. Companies excel at ignoring their vulnerabilities until it is too late.
So to bring a technology angle to this, there isn't anything remotely equivalent around this for Cybersecurity where it has some parallels. Everyone understands how bad a pandemic is - the risk of lives is obviously more important than data - however with the reliance of technology lives can be indirectly affected, such as the case of WannaCry shutting down hospitals and associated healthcare technology.
Vulnerabilities researched and weaponized by the government, ended up in the public, and caused global damage. There is no oversight into how these are produced or their risks.
Great ! Now this is going to open new rooms for more mistakes and may lead to more deaths which can be blamed to some scientific failure.
We humans collectively as one entity clearly have no discipline.
The current set of laws are themselves in first place not providing any peace in the lives of most people. And we have no time to fix that issue and our impulsive anxiety issue is making some weak willed men to open up new roads to destruction. It may yield some benefit but it is not essential.
I have 0 confidence in today's governments that they are capable of containing any damage caused by this law.
If entire of humanity is going to become extinct then we can take a high risk effort. All the day every time we need not take life threatening risks.
Opening up of this law means any minute now in best case scenario there is a 1% chance that there may be a deadly virus in the air u just breathed in. And based on history we humans are capable of leading ourselves into worst case scenarios and that puts the percentage of risk anywhere between 1% to 99%.
Now even if something worst case happens i am sure the human species would not go extinct as the billionaires and few millionaires would have means to survive any death prone event only with nobody left to buy their products anymore.
I'm not entirely clear on the scenarios in which this type of research would be beneficial (I'm a physician):
- AFAIK research on vaccines does not require altering an organism to make it more lethal. You would want to cover as broad a spectrum of serotypes/serovars of the existing organism as possible. I can see how someone might argue that you could pre-empt a more malignant variant but surely this is outweighed by the likelihood of creating systemic risk.
- What exactly would we learn form creating new, highly virulent organisms that can't be learned using the same technology on non-virulent organisms?
- It bothers me that this is a push to create more biological weapons from the military complex (especially terrifying considering the infantile intellects we currently have in congress).
Trial and error isn't great for this kind of stuff but it works great for AI research. Let's create software that can do genetic engineering of viruses right the first time.
As much as I dislike the subject, providing a monitorable outlet is better than a full prohibition. Prohibition denies important signals to things that will occur anyway.
As long as we progress, it will always be like that.
Technology is, in a fundamental way, a force multiplicator. It gives you power to do more things that you could do with only your biological body and mind. That power can be used for destructive means too, so it follows that as we get more power, we open new, exciting ways to destroy ourselves.
The downvotes seem weird to me. Maybe it felt off topic?
Dr. Francis S. Collins is in several of the videos, and the third video is all about this particular controversy of dangerous experiments at the dawn of genetic engineering(for example one guy wanted to splice a cancer causing gene in e. coli bacterium).
It seemed relevant to me. It is what the article is talking about afterall.
Yes because black tar heroin produced in a basement comes with the following disclaimer: ".... and the work must produce knowledge — such as a vaccine — that would benefit humans".
Stop making everything political, if this was done under a different administration you would have applauded it because of it being a scientific effort. Chances are this was decided by a panel and not the person you despise.
Being a-political myself, it always amazes me how for BOTH parties, the political issues aren't even about the issue at hand anymore. It's about proving the other side wrong.
This country has a sordid history when it comes to using/testing bio weaponry against those it designates as "non-people." Beginning with blankets coated in syphilis weaponized against the native population to experiments on black airmen at Tuskegee.
Trump referred to "very great people on both sides" of the Nazi rally in Charlottesville that led to the brutal murder of anti-hate protester Heather Heyer. He hesitated to denounce the endorsement of David Duke, a former Grand Wizard of the KKK. He appointed Jefferson Sessions, rejected from a federal judgeship for his racism, to the top law enforcement position in the nation. Under his watch they dismantled anti-violence police oversight programs and narrowed anti-terrorism efforts to draw attention away from domestic white-supremacist groups.
Stephen Bannon, brought into the White House as a senior advisor, helped coalesce support for Trump on the web, forming groups on forums like 4chan and Breitbart rife with eugenicist-apologia and genocide normalization strategies.
This is all well documented, here's a great read on it:
I'm anti-Trump as can be, but I'd suspect this moratorium was always intended to be temporary while the protocols to protect us from accidents were checked and revised if necessary. There's genuine merit to understanding what makes a virus vary in lethality and how to combat that.
There's plenty of things to bash this administration on. I don't think this is one of them.
"Federal officials on Tuesday ended a moratorium imposed three years ago on funding research that alters germs to make them more lethal."
or a little further down:
"In October 2014, all federal funding was halted on efforts to make three viruses more dangerous: the flu virus, and those causing Middle East respiratory syndrome (MERS) and severe acute respiratory syndrome (SARS)."
The pathogen to be modified must pose a serious health threat, and the work must produce knowledge — such as a vaccine — that would benefit humans. Finally, there must be no safer way to do the research.