Hacker News Reader

Comment by benzible | Hacker News Reader

See https://thecounter.org/lab-grown-cultivated-meat-cost-at-sca... for a thorough exploration of why lab grown meat will never happen...

> Humbird likened the process of researching the report to encountering an impenetrable “Wall of No”—his term for the barriers in thermodynamics, cell metabolism, bioreactor design, ingredient costs, facility construction, and other factors that will need to be overcome before cultivated protein can be produced cheaply enough to displace traditional meat.

> “And it’s a fractal no,” he told me. “You see the big no, but every big no is made up of a hundred little nos.”

I've posted this comment before, but the linked article is relevant to every story about lab-grown meat as the entire concept is a fantasy, or less charitably, a scam.

View on HN

I can easily believe it's a fractal of hard, but I cannot believe it's a fractal of no, and absolutely not as a question of thermodynamics.

It's clearly possible to make muscle cells (and other components of meat) reproduce rapidly: cattle exist.

It's clearly hard, or we'd have already have all this plus vat-grown organs and (affordable) vat-grown leather.

And some (non-meat) cells are already convenient to grow in vats (you just don't want to get all your protein from nutritional yeast and/or spirulina).

"1000 ways to not make a lightbulb", or whatever the real original Edison quote was — what's the number, 90% of start-ups fail?

The "no" in this context is addressing whether cultured meat will ever be economically viable at scale, not just scientifically possible. The article suggests a compelling argument against this viability, in that the breakthroughs needed for commercially viable cultured meat would also revolutionize the far more lucrative pharmaceutical industry. The fact that these advances haven't been achieved in pharma, despite massive financial incentives and decades of research, is strong evidence of their fundamental difficulty or impracticality.

For instance, even the most efficient cell lines used in pharmaceuticals aren't efficient enough for low-cost production of bulk cell mass. This implies cultured meat companies are attempting to solve problems that have long eluded a much larger, better-funded industry. The absence of these breakthroughs in pharma, where the financial rewards would be enormous, suggests they may be fundamentally unattainable with foreseeable technology.

> unattainable with foreseeable technology

This I agree with; it's just… well, I'm 40 now[0], and a lot of biology that can be done today involved technology that was not foreseeable when I was born.

So, yes, pharmaceuticals getting the hard problems solved first, sure — I'd say that I'd know not to invest in vat-meat startups until pharmaceuticals companies have solved problems I'm oblivious to except for the logical impossibility with such a test.

[0] aaaaaaa

The limits in pharma aren't production, they're figuring out which 10% of drugs will get through all the clinical trials.

While drug discovery may be the most lucrative opportunity overall, production costs remain significant in pharma, especially for biologics and cell therapies, and breakthroughs in these areas are well worth pursuing. CAR-T treatments, which use similar technologies to cultured meat, have manufacturing costs of $95,000-$156,000 per dose [0].

For generics and biosimilars, more efficient production could be a major competitive advantage. The fact that we haven't seen breakthroughs in this area, despite clear financial motivation, suggests these are not easily solvable problems.

[0] https://www.fiercepharma.com/manufacturing/car-t-manufacturi...

CAR-T treatments are expensive because they’re individualized. The process has to be repeated per-individual. That’s completely different from lab-meat.

I was responding to the parent poster's assertion that the only area worth focusing on in pharma is drug discovery, but you're right that CAR-T is very different. Monoclonal antibodies are a more apt analog. Key economic drivers in monoclonal antibody production include fermentation titer and overall yield, as highlighted in the article "Process economics of industrial monoclonal antibody manufacture" from ScienceDirect [0]. These same factors—cell density and yield—are also critical for lab-grown meat production.

The persistence of process-related challenges in both industries, despite significant financial motivation to solve them, underscores their complexity. This relates back to the "fractal no" concept from the original article, where each obstacle in cultured meat production is composed of many smaller obstacles. Importantly, monoclonal antibodies can sell for thousands of dollars per gram, orders of magnitude more than meat, yet even with this immense profit incentive, the pharmaceutical industry hasn't solved these bioprocessing issues. This suggests that cultured meat production, with its much lower profit margins, faces an overwhelming set of challenges,

[0] https://www.sciencedirect.com/science/article/abs/pii/S15700...

> absolutely not as a question of thermodynamics.

it "absolutely" is, anything that exists, is fundamentally a question of thermodynamics, it's unavoidable.

I don't think they were denying thermodynamics. Merely denying that it trumps economics.

Economics itself is a product of thermodynamics.

Which is both true and also missing the point.

The phrase "not a question of $x" does not mean $x can't be asked at all, but that the question is unimportant.

In this case, cows exist, therefore thermodynamics is resolved to "yes you can make a cow for at most the cost of a cow, in the general case", and I don't need to show detailed working to know that. Existence proofs make it obviously "not a question of" any law of nature, including thermodynamics, that we can't reproduce that existence proof synthetically.

This doesn't tell me how to do it, how to make it cheap enough for anyone to care, what the actual ultimate limits are, etc.

As someone who has zero stakes in this (well beyond the planet) and limited knowledge in the biological sides of this.

This feels incredibly dismissive? Sure I will concede that most likely the 9 year timeline mentioned in that article is incredibly optimistic and very likely will not come to fruition.

But I think it is also naive to think, regardless of how much you know on a particular subject, that there are not solutions to problems that have not been developed yet or that there are already examples to those problems that are not public knowledge (it even mentions that the paper was written with NDA knowledge).

Then it is even more naive to try to paint a "wall of no" seemingly in a way to shut down research on this subject.

Just look at the silicone chips allowing us to have what once would have been considered a super computer in our pockets compared to when computers were tubes. Of course at that time anyone suggesting that a personal computing device would even be possible would likely be laughed out the room, the core technology didn't even exist to think that maybe some day it would be possible.

Are these companies and that paper making lofty promises that may be borderline lying to investors, maybe. But I don't think we should be saying it is impossible at this point in time.

Fusion might be good comparison here. Yes we can make it happen, yes we have multiple solutions. But still even with decent investment we are not near economical energy production. And it is not like we have not tried reasonably.

Somethings are just rather complicated at scale and both cultured meat and fusion belong to this category.

That I agree with, which is why I mentioned that the 9 year timeline is likely too ambitious.

But there is a difference between it being very complicated and likely relying on technology (or a fundamental understanding of something we don't know about) we don't have yet, and a "big no" and "never". Which is what this paper seems to be saying (reinforced by the comment I was replying to also saying "never").

That is the part of this that bothers me. Yeah we should admit that it is going to be very hard (if not impossible) and we should examine the reality of the timeline claims. But we should not be discouraging the research on both of these topics at this time by making claims that it will never happen.

I wouldn't bet against it ever happening, mainly because there is already a pretty advanced highly scalable meat manufacturing biotechnology process called, you know, the actual animal. Wouldn't recommend betting against a thing that already exists, it's not like fusion where we don't have small viable energy producing reactors yet (the sun doesn't count). Timeline though? Who the hell knows. Progress in biology has been unbelievably fast over the past few decades in a really unpredictable way (sometimes rapid improvements in existing technologies, sometimes unexpected breakthroughs that open up new horizons). Maybe the core challenges to replicating what animals do will take another 30 years to figure out, maybe a breakthrough will happen in the next two months, who knows. But a bunch of random chemical reactions kicking around for 3 billion years figured out how to make meat so don't think it's unreasonable to think we'll be able to figure it out too.

Right, and I think thats why the article bothered me and I made this comment.

We can criticize the timeline and the promises made to investors, that is incredibly valid.

But "never" is naive.

The only time I am really comfortable with saying "never" is something like, we will never travel faster than the speed of light. That seems to be fairly well proven at this point in time and if we ever fight a loophole around it, likely the core law will remain true.

Yeah, and it's especially naive when in addition to animals existing, the tech involved happens to be an area of enormous societal importance for health reasons (plenty of investment), the problem happens to be somewhat existential (climate change is getting tons of funding right now), and the science of biology is advancing at a frankly extraordinary rate.

Even GFI, the tech hype bros in that article that's being responded to, is showing a pretty big lack of vision. "This gradual progression is necessary; you can’t just throw a small amount of cells into a large bioreactor and hope they’ll start dividing. Cells are “fastidious,” Hughes told me, and have strict metabolic requirements for growth, including oxygen tension." <- pretty comfortable saying that under those constraings, 100% agree on the fractal of nos. But like, yeast and other things we grow in bioreactors aren't that picky. The hype bros want to try and convince people it's happening now so they're pitching some nonsense vision of mass manufacturing based on tech that obviously can't scale, and taking their vision of how things are going to work as ground truth is kinda silly. Almost by tautology, a successful bioreactor based approach is going to be one that uses bioengineering to create cells that don't need that sort of pickiness. It doesn't exist yet, sucks for the hype bros, but how sure are you that nobody will ever figure out how to engineer yeast to work as meat so you can use bioreactors from defunct breweries or something?

The first cow genome was only published 15 years ago - at least wait for biology and medicine to plateau and for the incentive of fixing climate change to go away before convincing yourself it's never gonna happen!

> and other factors that will need to be overcome before cultivated protein can be produced cheaply enough to displace traditional meat.

This elides over the cost differential and where that "cheaply enough" wall is. Animal grown meat is currently rising in costs and will continue to rise in costs as both demand is rising and the habitat for those animals is constricting due to climate change. How much cheaper does lab grown meat and how much more more expensive does animal grown meat need to be before it becomes viable. From what I've read they are closer than this argument assumes.

another classic case of differential accounting - if the climate adaptation cost attached to farming was charged in the price of meat, maybe the no would be on the other foot.

It seems like all the arguments there are predicated on the "meat" being actual muscle cells in a vat, though. Wouldn't it be just as likely that the successfully engineered food ends up being just, like, kelp with GM'd amino acid byproducts or something? Or chicken-of-the-woods with more random proteins thrown in? (I mean as long as they don't kill the culture, it doesn't matter if it benefits the host organism.)

I'm sure that's all harder than it sounds. But I doubt it's the same "wall of no" as muscles in a vat.

I wish I could get back the five minutes it took to read that stupid article. The argument for the second half is “his report seemed really technical to us and nobody has ever cited it in the four years since it was published so it’s a massive conspiracy by every scientist ever because everything he said was so right.”

Either every single company is wholesale lying about key parts of their processes and outputs…or this is just another hitpiece by Big Ag with exaggerated claims.

> See https://thecounter.org/lab-grown-cultivated-meat-cost-at-sca... for a thorough exploration of why lab grown meat will never happen...

The NYT carried another opinion piece by the same author this year[1] along with a number of reader rebuttals[2], including this one from Mindy Kursban:

> It is surprising that in his obituary for cultivated meat, Joe Fassler spends so few words on the viability of the science and cites more chief executives than scientists.

> Just recently, the chair of Tufts University’s biomedical engineering department, David Kaplan, who has published more than 1,000 peer-reviewed papers, told[3] a room full of cultivated meat scientists that progress over the past five years gives him “complete confidence that we’re going to get to where we need to go.”

> There are many eminent scientists all over the world who share Dr. Kaplan’s view, but readers would not know that from reading Mr. Fassler’s essay, which argues that cultivated meat is a project of Silicon Valley “dreamers with a fancy prototype, a pitch deck and a good amount of natural charm.”

> This groundbreaking science may be taking longer than initially thought, but it’s making strides, so it’s a shame that Mr. Fassler spends so little time exploring it.

[1] https://www.nytimes.com/2024/02/09/opinion/eat-just-upside-f...

[2] https://www.nytimes.com/2024/03/02/opinion/lab-grown-meat.ht...

[3] https://www.youtube.com/watch?v=iGzXAuiYKGU

all we need to do is break the second law of thermodynamics, what's the big deal?

Last I checked the cow doesn't have a problem producing steak and thermodynamics applies to that too so not sure what the argument is here

Ruminants are actually very efficient protein generating machines.

They (like us) have been evolving under thermodynamic pressures for millions of years, I would like us to be able to not have to produce them to have high quality protein/fat/etc, but I haven't seen anything to say it's remotely possible.

We only sequenced the first cow genome 15 years ago and plenty of plants and fungi produce high quality protein/fat/etc - it's not like we're starting from scratch the whole point of bioengineering is that we can cheat using stuff nature figured out for us. It's certainly not scientifically impossible. In the zero interest rate hype cycle over the past couple years plenty of startups convinced themselves that they could totally make it work with cell culture and stupid expensive bioreactors and media and stuff since that was the only tech that was 'ready' even though at best all you can get is an overpriced lean meat only replacement for ground beef, and this article seems to take for given that this approach === meatless meat. I think it's useful to keep the questions of 'are these tech bros going to figure out how to make producing meat in class 8 cleanrooms viable' and 'are bioengineered animal-free meat replacements going to happen in the next few decades' seperate

Reply on news.ycombinator.com