Vegetable Oil Is Bad?

Topic for discussing the possible dangers of eating vegetable oil.

My main initial information source is Jeff Nobbs, who wrote a four part series:

  1. What’s Driving Chronic Disease?
  2. Death by Vegetable Oil: What the Studies Say
  3. Why is Vegetable Oil Unhealthy?
  4. The Environmental Impact of Vegetable Oils

I also watched this video: The $100 Billion Dollar Ingredient making your Food Toxic - YouTube

And I watched some “how it’s made” type videos showing how vegetable oils are made (and to compare, I watched similar videos for olive, avocado and coconut oil, which are better). In short, for vegetable oils, they use very industrial processes involving high temperatures, heavy machinery, solvents and bleaching agents. This is necessary to get enough oil yield from small things like seeds or soybeans to keep the products cheap. Olive, avocado and coconut have more oil that’s easier to get out, so they’re typically processed less but are still reasonably economical.

Nobbs explains the main problem with vegetable oil, which he claims is basically uncontroversial. It has a lot of omega 6 fatty acids, which have low stability to heat, light and oxygen. When we eat that, we build our bodies out of more of it, which makes our body more unstable. That results in more free radicals, and our antioxidants being overwhelmed, which results in all kinds of damage leading to more heart attacks, strokes, cancer and other problems. Free radicals are molecules that lost an electron and can take one from somewhere else (which damages the thing they take it from). Antioxidants are molecules that are stable, and they can lose an electron and still be stable, so they can safely give up an electron in order to neutralize a free radical.

The average American now gets 20% of his calories from vegetable oil. There’s a lot of it in processed and restaurant foods. The government started pushing it in the 1960’s and subsidizes producing it. They thought it was healthy, but as with many health claims, they were arrogant and didn’t know what they were talking about. They introduced us to eating transfats which they now recognize as bad, but even without transfats vegetable oils may still be really bad (plus vegetable oils today still have transfats which the government lets them round down to zero on the label).

I have not yet fact checked this stuff or read the studies.

1 Like

multi-post series plausibly arguing that vegetable oil is our biggest health problem from food. this seems potentially important and shocking/disturbing

by convention, “vegetable oil” doesn’t include olive, coconut or avocado oil (which seem to be healthier). it’s mostly soy bean oil and seed oil. but beware of egregious fraud:

A study performed at the University of California, Davis in 2020 determined that a majority of the domestic and imported avocado oil sold in the US is rancid before its expiration date or is adulterated with other oils.[6][7] In some cases, the researchers found that bottles labeled as “pure” or “extra virgin” avocado oil contained nearly 100% soybean oil.[6]

There’s a lot of stuff in the posts. Here’s one quote from part 3:

Researchers studying cancer in mice even suggest that dietary polyunsaturated fats like omega-6 are required in order to induce tumors in mammals. In other words, even when trying to do so, researchers may be unable to intentionally form tumors in lab animals without feeding them fats like omega-6. The same phenomenon is found with liver damage: omega-6 consumption is required for the development of alcohol-induced liver injury, and the severity of the liver damage is correlated with the amount of omega-6 in the diet. Furthermore, replacing omega-6 fats with more traditional fats, like those from coconut, reverses alcohol-induced liver damage, even when maintaining the same level of alcohol consumption [26].

If anyone has a criticism of the idea that vegetable oils are harmful, I would be very interested.

Seems like there’s a lot of olive oil fraud as well. The fraud seems more prevalent with imported brands. I think its better to buy olive oil labeled 100% California.

I already have Kirkland olive oil from Costco (which is from the Mediterranean region, not California), which fortunately appears to be one of the best options.

In Costco Connection, Shauna Lopez, a corporate food buyer, said the company knew of the problems plaguing the olive oil market and had taken steps to ensure that its sourcing had integrity. “We have always been involved in this program by meeting the farmers and touring the mills and the processing plants in order to hold everyone in the olive oil production process accountable for the olives they bring in,” Lopez said. “This ranges from the farmer who is registered with the mill, to the mill that grades and batches the olives as they come in daily.”

In general, Costco is one of the companies I have a higher level of trust in. (Please correct me if I shouldn’t.)

I read up to the start of the section “How Do Vegetable Oils Affect Energy Expenditure?”. I found multiple small signs of bias. I found writing flaws that suggest he (Nobbs) wouldn’t read the research studies as precisely as I would.

Only two brands [of 22] produced samples that were pure and nonoxidized. Those were Chosen Foods and Marianne’s Avocado Oil, both refined avocado oils made in Mexico. Among the virgin grades, CalPure produced in California was pure and fresher than the other samples in the same grade.

I bought avocado oil at Costco recently and it’s Chosen brand. Costco also sells Marianne’s. That’s a good sign about the judgment of Costco’s buyers.

Here’s one of the defenders of seed oil:

CVD is cardiovascular (heart) disease. he posted four tweets of links to studies. on review of several, they’re irrelevant or dishonest. one of the main issues is they correlate vegetable oils (compared with alternatives like saturated fats) to lower cholesterol measures. they then claim/assume that lower cholesterol means lower heart disease.

the anti vegetable oil side does not dispute that vegetable oils result in lower cholesterol on those measures. that is not the issue. the issue is whether that’s actually good or bad for health. so the studies beg the questions being debated – they assume a conclusion about one of the actual things at issue (the alleged link between cholesterol measures and heart disease).

the biased tweet author has a small patreon and youtube and stuff. he’s an aggressive online vegan:

But again, I would be happy to debate vegan ethics with you. You appear as though you need some education on the subject, and I think getting your ass handed to you in a verbal debate would probably go a long way to ensuring that you post fewer dumb things in the future. :man_shrugging:

he also wrote a long post defending seed oils which i haven’t read any of so far:

notably, vegans can pursue their agenda without defending industrial seed oils. seed oils, in addition to possibly being unhealthy, are also bad for the environment and are also connected with meat production (they squeeze oil out of soy beans and seeds, then use what’s left for animal feed, so if we eat less soy and seed oil then animal feed prices will go up). the factory farm meat industry (including farm-raised fish) and the vegetable oil industry are symbiotic. vegans could agree that seed and soy oils are unhealthy, link them to industrial food processing and factory farms, and then advocate olive, avocado, and coconut oil over meat oils or butter.

Before reading the vegetable oil stuff, I was reading Grain Brain which actually covers vegetable oils (with a negative take) some in a later chapter that i hadn’t gotten to yet (i just did a quick search and saw it). Grain Brain is an anti-gluten/carbs book which puts a major emphasis on defending cholesterol.

A few quotes from Grain Brain to give you some idea of what it claims:

Neurons themselves are unable to generate significant cholesterol; instead, they rely on delivery of cholesterol from the bloodstream via a specific carrier protein. Interestingly, this carrier protein, LDL, has been given the derogatory title of “bad cholesterol.” In reality, LDL is not a cholesterol molecule at all, good or bad. It’s a low-density lipoprotein (hence its acronym), and there is absolutely nothing bad about it. The fundamental role of LDL in the brain, again, is to capture life-giving cholesterol and transport it to the neuron, where it performs critically important functions.

This is why I said “cholesterol measures” instead of “cholesterol” above. LDL is a thing they measure when trying to measure cholesterol (if Grain Brain is right).

If you’ve had your cholesterol levels tested, you’ve probably lumped HDL (high-density lipoprotein) and LDL (low-density lipoprotein) into two different categories—one “good” and one “bad.” I’ve already mentioned these two labels for cholesterol in passing. But contrary to what you might think, they are not two different kinds of cholesterol. HDL and LDL reflect two different containers for cholesterol and fats, each of which serves a different role in the body. Several other lipoproteins also exist, such as VLDL (very low) and IDL (intermediate).


It’s important to understand that when you have a blood cholesterol test, the number that is represented is actually 75 to 80 percent derived from what your body manufactures and not necessarily what you’ve eaten. In fact, foods that are high in cholesterol actually decrease the body’s production of cholesterol. We all make up to 2,000 grams of cholesterol every day because we desperately need it, and this is several times the amount found in our diets. But despite this amazing ability, it’s critical to obtain cholesterol from dietary sources. Our bodies much prefer that we “spoon-feed” our cholesterol from the foods we eat rather than manufacture it internally, which is a complex multistep biological process that taxes the liver. Dietary cholesterol is so important that your body absorbs as much as it can for use.

So what happens if you restrict your cholesterol intake, as so many people do today? The body sends out an alarm that indicates crisis (famine). Your liver senses this signal and begins to produce an enzyme called HMG-CoA reductase, which helps make up for the deficit by using carbohydrates in the diet to produce an excess supply of cholesterol.

so studies say veggie oils are good because they lower cholesterol and therefore improve CVD. but that is not an answer to reasoning like this which agrees that veggie oils lower cholesterol but says that’s actually bad.

This defense of vegetable oils said nothing to challenge my conclusions.

I messaged Jeff Nobbs:

Hi, I’m a philosopher. I write about topics like rationality and epistemology, e.g. Multi-Factor Decision Making Math ( Multi-Factor Decision Making Math ).

I recently started researching food health. I found your articles on vegetable oil. My initial impression is that you’re correct and the information is very important and disturbing.

I wrote lengthy initial thoughts about food health shortly before reading your articles, which I thought would interest you. I started by thinking about broad conceptual explanations. Curiosity – Plants vs. Mammals (Energy, Digestion, Broad Conceptual Explanations, and the Lion Diet)

I deal with topics like how to organize debates effectively, and how public intellectuals can and should be open to error correction, which are relevant to your attempts to debate with and correct people about vegetable oil. Many or your difficulties with persuading the world appear to me to be specific examples of some more generic concepts that I’ve written about. Those include how people approach debate in an irrational, disorganized manner, focus on social status, and are bad at reaching decisive conclusions.

I posted a comment on Jeff Nobbs’ blog:

Death by Vegetable Oil: What the Studies Say | Jeff Nobbs

My initial impression is that the criticisms of vegetable oil are important and it’s dangerous. I want to investigate this issue. So one step I’m doing is writing my own comments on Nico’s arguments in order to think about them.

  1. I agree that harmful effects like these do non-linear harm (I interpret that as the key point here, though neither Nico nor Jeff said “non-linear” or a synonym). I don’t mind the comparison to smoking as a very loose approximation to help give readers some rough sense of scale or importance – it seems to me better than nothing rather than worse than nothing (if Nico or anyone knows a superior way to accomplish the same goal, please share). Even if those passages should be deleted, I don’t see how it would change the overall conclusion against vegetable oil.

  2. My understanding is that the absolute numbers are big on a societal level. People are getting 20%(!) of their calories for vegetable oil, on average (so quite a few people are getting significantly more than that). It’s having large-scale effects like significantly changing the omega-6 levels in human breast milk (from under 5% to 20%+). And chronic disease numbers are up and a lot of people are dying of heart attacks and other things that vegetable oils contribute to. (I haven’t yet fact checked all this or read a sample of the studies myself, but Nico doesn’t seem to be disputing the basic facts.) I think part of Nico’s point is that some studies have small sample sizes and it’d be good to do more research, but that claim is neutral regarding the best current conclusion to draw given the available arguments and research. And I don’t think the risk difference between 1 in 35k and 1 in 23k is actually negligible. It can be negligible when other factors involved in a decision are much more important, but that isn’t always the case (e.g. when choosing which oil to buy off a store shelf, I might be largely indifferent other than health concerns, so there aren’t some other factors dominating the analysis). And a lot of the issues related to vegetable oil seem to be much larger than that.

  3. If what Nico says is correct (I haven’t read the study yet, but Jeff seems to acknowledge most of it), then I am concerned that the Rose Corn Oil Trial was inadequately disclaimed in the blog post, and I’m not sure that it should have been included at all. Jeff wrote, “The two-year trial had only 54 participants, so while the results are dramatic, the data from this trial should be taken with a grain of salt.” That disclaimer doesn’t mention serious problems like not controlling how much fat each group gets, not meeting statistical significance tests, and unclear “remaining event free” math. So after reading Jeff’s reply, I still tentatively agree with Nico on this one. However, even if we ignore this trial, it wouldn’t change the overall conclusion about vegetable oil. One way I might change my overall conclusion is if a substantial portion of Jeff’s research was shoddy, and this point could potentially, along with other arguments, contribute towards determining that.

  4. Jeff gave a good answer here. He highlighted the unexpected part of the results. So that wasn’t bad cherry-picking. Without having read the study, I have nothing to add currently.

  5. Again without reading the study, just based on what Jeff and Nico said, Jeff’s response seems reasonable for why he left out the fourth group. So far what I’m seeing is that in three examples chosen by a critic, Jeff’s research appears to hold up fine in two out of three cases. That seems adequate, not like the kind of systemic research quality issue that could change the overall conclusion about vegetable oil. Nico’s other point here is that he thinks the study conditions don’t map to realistic human eating. Jeff responds that unfortunately restaurants reuse oil in deep fryers for days and for many batches of food. That seems believable to me and Nico gave no argument disputing it. Nico further says to get an equivalent amount of vegetable oil from McDonald’s fries you might have to eat 21oz which is a lot of fries. Yeah but people eat many different things with vegetable oil and it adds up. A large amount of both processed and restaurant foods contain vegetable oil, so many people get vegetable oils from multiple foods on their plate (and drinks like non-dairy milk) in most of their meals. Nico’s undetailed skepticism about how much vegetable oil people eat leaves me wondering if he read through all of Jeff’s first 2 posts in this series and was trying to be fair/objective. If he thinks the 20% of our calories from vegetable oil statistic is wrong, Nico should check the source on it and directly challenge it instead of commenting about fast food like this. In Nico’s defense, my perspective is biased by having also read Jeff’s parts 3 and 4, so I got more information and detail than Nico would have gotten from only parts 1 and 2, and I don’t clearly remember which information I got from part 3 that Nico wouldn’t have seen yet (I do remember that part 4 focused on environmental issues so it had less relevant info).

I commented:

What's Driving Chronic Disease? | Jeff Nobbs

Suppose you eat a regular American diet with vegetable oil, and then you switch to a diet very low in vegetable oil and with a decent amount of omega-3s (e.g. eating fish twice a week). Roughly how long does it take to heal the composition of your body? E.g. how long before the fatty acids in your muscles drop from ~25% omega-6 to under 5%?

One of the main arguments for vegetable oil is that it lowers cholesterol and therefore lowers heart attacks. A lot of the studies check that it lowers cholesterol and then conclude that it lowers heart attacks based on a pre-existing assumption that high cholesterol is a main cause of heart attacks.

It seems to be uncontroversial that vegetable oil does lower cholesterol according to a particular way of measuring cholesterol. (But Grain Brain says they’re really measuring a thing that helps carry/transport cholesterol, not cholesterol. That book also says that eating more cholesterol is good for you.)

Some studies looked at actual health outcomes, like death, and found that some people eating more vegetable oil had lower cholesterol but worse health outcomes. Studying a proxy like cholesterol levels is much easier – you can just control people’s diet for a few weeks then do a measurement. Looking at actual health outcomes requires multi-year studies.

Does anyone know of something convincing that cholesterol (or its carrier molecule) is actually bad for you or causes heart attacks? I have actually seen some mainstream/pro-vegetable-oil type people admitting or agreeing that cholesterol isn’t the right issue, for example:

He begins by attacking the lipid hypothesis: the idea that fat in the diet causes elevated blood cholesterol which causes cardiovascular disease. This no longer needs attacking: mainstream recommendations have followed the evidence and have evolved from a low cholesterol diet to a low fat diet to a low saturated fat diet to avoidance of trans-fats and of excess calories from fat.

He seems to be denying/conceding the main claimed advantage of vegetable oils while not offering any new advantage nor denying their main disadvantages (less stable to light, heat and oxygen. plus correlated with weight gain on equal calorie diets in studies for reasons that I think are unknown).

But his overall conclusion nevertheless seems to be to defend the status quo and vegetable oils. Here’s one more quote:

It is probable that the dangers of saturated fat have been exaggerated, but no reputable cardiologist would claim that saturated fat is good for you or that fat should make up 70% of the diet. Cardiologists have become more liberal about allowing some cholesterol and fat in the diet, but they still consider fat intake as part of the complex equation of risk factors. Fat contains 9 calories per gram compared to 4 calories per gram for protein and carbohydrates, and the calories can add up fast to contribute to weight gain.

(ugh/lol at the idea that since fat has more calories per gram you’ll get fat if you eat it)

Nobbs is CEO and (co?)founder of

They launched their first product today, called Cultured Oil. (Previously I had basically no idea what sort of thing were working on – just something anti vegetable oil.)

They don’t say what it’s made out of, and also refuse to admit that they’re hiding the ingredients and the manufacturing process.

They talk about vegetable oils having a bad manufacturing process, but then they hide theirs and just ask you to trust them with no transparency!? Except they don’t even openly ask for your trust or admit to any lack of transparency. They don’t e.g. claim that their process would be easy to copy by big food companies if it wasn’t secret, so they have to hide a lot of information until they’re a big company.

Zero Acre FAQ:

What is fermented to make Cultured Oil?

Those plant sugars can come from a variety of sources, including sugar beet, sugarcane, cassava, and other plant sources.

That’s the closest you get to an answer, which doesn’t answer. They also don’t even try to give some health information, which a lot of anti-vegetable-oil people might care about, such as whether it contains any seeds and what sort of lectins it has. Instead they give biased information claiming e.g. that it’s sorta, kinda paleo in spirit. They also don’t say how much omega-3 it has, what the omega 3 to 6 ratio is, or how it’s different/better than the high-omega-9 vegetable oils that are already being manufactured.

They try to say it’s healthy because it contains the same fats as people ate in the past, but you could say that about vegetable oil. That’s a biased argument.

They have $40,000,000+ funding:

Zero Acre Farms puts microbes (and $37M) to work on a better alternative to vegetable oil – TechCrunch

It appears to be funded primarily by environmentalists not health people.

Crunchbase says it was founded Jan 1, 2020, meaning Nobbs started the company before using his blog for marketing with an undisclosed conflict of interest. And he already knew he had a founder/CEO job to keep him busy when he promised a part 5 “coming soon” in his vegetable oil series, and then never delivered. Nobb’s first blog post was Jan 16, 2020, and he appears to have had multiple posts pre-written in order to get a lot of posts out quickly.

I’m uncertain if that date is accurate though – it seems like a somewhat generic date because the exact start date is hidden. The seed funding round of 3.6 million dollars was announced Nov 4, 2020 which, due to the size, probably means the company had been doing stuff for a while before that. Nobbs, Zero Acre, etc., are vague (seemingly on purpose) about stuff like who the executive team of the company is, when it was founded, how much money they have, what kind of pitch got them that money, etc. Anyway my best current guess is that Nobbs was fully biased by a predetermined conclusion he was betting his career on while writing every one of those blog posts.

Nobbs seems to aspire to run a mega food corp doing some industrial process on a huge scale with no transparency, not to e.g. run some small, open company. He’s trying to be like the people he attacked. He attacked them because he wants to take their jobs and marketshare, not to rid the world of stuff like them. His attacks gave many readers the other impression though (wanting to rid the world of that stuff). He wants his ideas about health – which have no Paths Forward, no openness to debate, no better error correction – to determine people’s diets. He wants to be the new ruler instead of setting up a better system. He doesn’t understand issues like “Who should rule?”, error correction, or how there’s systemic corruption in the industry affecting all large players and he’s, if successful, just going to end up corrupt (if he wasn’t already). There are mechanisms (like interaction with government and many others) which cause this corruption, which he has written nothing about (he has no plan to defend himself or avoid corruption) and he isn’t open to discussion about this.

Another thing about Nobbs, which I’ve seen with a lot of people, is he gives the impression he’s more of a maverick or original thinker than he is. He doesn’t cite sources enough and wants you to think that he figured this stuff out himself. (He cites academic papers and facts/stats but not other thinkers and writers who he got explanations and arguments from.) But TechCrunch says another recent but older startup was already trying to make oil by fermentation. And paleo diet people, among others (like the book Grain Brain), were already saying vegetable oil is bad before Nobbs.

His seed funding comes from a VC using “Outliers welcome.” as a slogan. I think they’re just liars. I see nothing on their website to persuade me the slogan is true, and no acknowledgment that it’s a claim that reasonable people would meet with heavy skepticism. And the website has indicators that they’re social-climbing conformists who like trendy iconoclast-style marketing.

How do you welcome outliers? A reasonable way to try is to become a serious intellectual who can write and has a lot of philosophical knowledge including about topics like fallibilism. If that was your method, you’d write some stuff and explain it. Is there any other method that would work well? Nothing immediately comes to mind and they don’t suggest anything or acknowledge that welcoming outliers is actually a hard problem (while fooling yourself is easy). I don’t think the tolerance tradition is good enough for this because tolerating something is like putting up with it, not minding it or being willing to leave it alone, rather than welcoming it, liking it and actually funding it.

Everyone sucks. They’re all a bunch of biased tribalists.

PS: It’s still my best current guess that vegetable oil is bad. The above is just about side issues.

If you want to avoid eating vegetable oil, you have to largely avoid processed food and restaurants, not just switch which bottles of oil you have in your kitchen.

If you want to make only easy changes, then stop buying bottles of vegetable oil since they’re easy to replace (with olive, coconut, or avocado oil, and/or butter, tallow, or lard). And the worst offender is deep frying, which you could avoid reasonably easily.

Deep frying involves high heat and oil reuse (so more light and oxygen exposure over time, and also it’s heated repeatedly). For example, apparently McDonalds changes their fryer oil every 1-2 weeks (they filter it more like daily to get out bits of food). Sautéing in single-use vegetable oil at lower heat is not as bad as deep fryers, though it still has issues (omega 6 fats, solvents, bleaches, etc.).

McDs uses highly refined oil with the taste removed and then assumes that taste and color stability means the oil is still fine for cooking. There’s something really wrong with using common sense metrics like taste, smell, texture and color to judge oil that is super unnatural. Those metrics work pretty well on natural foods, though sometimes fail. They are much less reliable when modern science and industry tweaks stuff. An obvious example would be adding dyes to a food and then thinking that the good-looking color means it’s probably safe/healthy. A little more subtle is adding gums/emulsifiers/etc to stuff so the texture seems non-rotten, but which could hide rottenness. I think it’s basically the same with the oil: they made it unnatural enough that the commonsense ways to check if it’s still good food are a lot less effective. With the oil they actually make it taste bad during processing and then remove the taste… So the common sense metric actually said it was no longer good to eat earlier on and they hid the problem and then now they are saying things kinda like “it tastes fine so it’s fine”. But when something tastes bad, that is an indicator of a possible health problem in the food, and fixing the taste doesn’t mean fixing the problem – the problem could easily involve molecules other than the ones we taste.

1 Like