I’m working on growing the US tofu market—wrote a book on the subject (www.brokencuisine.com) and figuring out what’s next.
Besides tofu, I restarted and led the EA Los Angeles group and previously restarted and co-led the EA club at the University of Southern California.
Going to see about connecting 👀
Appreciate the signal boost on point 3!
Thanks Desampo! The tofu production process is pretty specialized, so I don’t think there’s much equipment overlap with other food categories. Could be missing something though!
To flesh this out more, my thinking is something like, WWOTF PR team probably has a theory of impact, which might dwarf the impact of the other ones, and it seems suboptimal if groups are focusing on the wrong one. Maybe their primary theory of impact is shifting the longtermism/AI overton window (attracting as many hits/eyeballs as possible, mass outreach), or maybe it’s attracting allies in positions of power, or maybe it’s inspiring more readers to pursue longtermist careers. Or maybe it’s something else. Getting more of our members to read the book seems aligned with all of those goals, but I wonder if there would be more impactful specific things, tailored towards their theory of impact.
I’m trying to articulate examples of activities that we might pursue, given different end goals. The clearest case for differing approaches would be if WWOTF’s goal is primarily mass outreach—in which case groups might be able to make connections to local media groups that aren’t big enough to warrant WWOTF attention. Or maybe focus more on partnership events that aren’t trying to funnel attendees into our EA group, but rather expose them to some longtermist framework. If the goals are more targeted outreach / career stuff (as I suspect they are, at least with regards to what groups should optimally be helping with), then this dovetails more with our current initiatives/funnel strategy.
Another concern is that this new audience could be categorically different from the standard proto-EA – possibly less altruistic (in the conventional sense of the word), less into impartial welfare, possibly working in different careers, etc. In that sense, there’s the question of how much we should focus on educating them on EA vs referring them to groups/opportunities within longtermism. There’s also the question of anticipating changing group dynamics, if, say, we suddenly start getting like 5x the number of leads/members. These concerns might be mostly unactionable—we’ll see what our new audience is like once they start showing up. It’s likely our main group initiatives won’t change, but I think there’s a chance they could quite substantially. Moving forward, it would be great to see more coordination among groups, sharing if/how things are changing and how we’re responding.
But it’s also possible that creating these isn’t in a student’s comparative advantage, especially if they aren’t already that knowledgeable about the skills they want to teach.
Right, this is what I suspect. It’s naturally more efficient to expand a pre-existing program than create a new one from scratch, especially in highly technical fields.
Do you have more details on what you’re envisioning here? What sorts of workshops do you think would be most useful?
I don’t have a great inside view on this, but the sorts of workshops Sydney has been running seem pretty popular (we had a couple USC fellows attend her “Impact Generator” workshop and they found it both helpful and motivating.) Lightcone in the Bay is doing a ton of that too, and GCP was planning to build out workshops after fine-tuning their Guides program.
sure, I’ll shoot you a PM
So I don’t think people should feel discouraged just from their low offer rate very much.
To clarify, is your main point here that that AI safety orgs could absorb a lot more talent, if folks were more qualified? If so, that’s also my understanding. I doubt, however, that general university groups would be the optimal way to build qualified and motivated applicants, because student clubs inherently spend a lot of time on non-development focused activities. It seems like outsourcing the recruitment pipeline to, say, Cambridge’s AGI Safety Fundamentals program, GCP guides program, plus cross-university skill-building workshops would accomplish the majority of what a successful university AI safety club might accomplish, plus some, with a lot less organizer time.
Given that those programs aren’t fully built out, it might make sense for organizers to spend more of their time helping to build up those programs, rather than devote a ton of time to an AI club at their home university.
What’s the distinction between “community building” and “movement building”?
Good catch—I added “movement building” late last night and it’s way too vague. I meant it to encompass important things that recruiting doesn’t really touch on, like upskilling, but it is way too unspecific. I’ll add a note
Suji is incredible! We’ve been calling it “Shanghai tofu” in our book, since vegetarian chicken can sometimes be misinterpreted as a mock meat.
Funny enough, if you freeze and thaw it a couple times, the crumb becomes very bread-like. Stew it in a creamy, sweet sauce, refrigerate it overnight, and it will have an almost “tres leches cake” consistency. Such a cool ingredient!
Thanks Joel. And absolutely, I’d love to chat and hear your China story!
Thanks for sharing your concerns here and taking time to write this all out. I agree that framing these ingredients in a sensitive way is really important.
I would second having Chinese people on your staff, as well as looking into ways your project can benefit the rural villages whose people developed the tofu.
I’m definitely looking for other allies and partners!! If you know anyone, I’d love to connect :)
I’ve heard a general rule to avoid implicitly playing into these stereotypes is to ask yourself whether the ethnic descriptor is necessary to get the point across. In this case, I don’t think it is, because tofu is already Chinese.
I’d actually disagree pretty strongly on this point—I think “Chinese” is a pretty important descriptor.
For folks in China, calling something “Chinese” is generally a symbol of pride, the same way American products might brag that they’re “made in America.” There are definitely cases where this isn’t true (i.e. Trump’s “China virus”), but bad examples don’t seem like a reason to not use the label in positive ways.
Besides pride and respect, there seems to be important information value. In non-Asian American communities, tofu is treated as this homogenous pan-Asian thing. The reason I think this is problematic is a) it’s culturally insensitive, b) it’s factually inaccurate, and c) it compresses Asian food into a monotony that’s harder for folks outside the communities to understand and appreciate.
Tofu is not just a Chinese thing, but the varieties we’re using are mostly Chinese. In this sense, labelling them as Chinese helps fight the cultural flattening. It also allows folks to dig deeper into how these ingredients are different. (There are some “pan-Asian” tofus, like firm/soft/silken, but our team isn’t focused on those.)
The alternative, not calling them Chinese, seems worse to me. Almost like saying, don’t call Italian olive oil “Italian.” It would probably be seen as disrespectful to producers in China, like not giving them credit, and it would restrict consumer understanding.
I’m less certain about the “rare” label. It seems that the word can have both positive and negative connotations, and sometimes provide important information value.
The main reason I call these tofus rare is to distinguish them from ordinary tofus. Differentiating them seems important because a) most US non-Asian Americans don’t love ordinary tofu; b) this dislike stems from inherent characteristics of ordinary tofu, rather than branding, meaning that negative perceptions will likely continue into the near-term future (happy to share more—my team’s done a bit of research into this question); c) if “rare” tofus can’t distinguish themselves from ordinary tofus, they will run into the same consumer opposition; d) “rare” tofus are fundamentally different from ordinary ones and could have wide appeal amongst anti-tofu foodies.
I think it’s also accurate to call them “rare”, as they aren’t common in most of the world. Many of these tofus aren’t even common within China.
Consumers find rare/scarce things more exciting and appealing. This is true for everything from rare collectibles, exclusive events, limited edition specials, etc. I don’t think the connotations are inherently positive or negative—sometimes they can be othering, other times they can increase interest, but it really depends on context.
Business often exploit, commoditize, and market “exotic” or “foreign” culture. I think this is pretty sad for many reasons, including that it marginalizes folks in our communities who identify with that culture. I think marketing rare tofus to western cooks, in the way we’re hoping to do, however, has almost the opposite effect—it engages folks with ingredients they otherwise would not know about, creating bonds between the U.S. and China at a time of historical tension, over common concerns—fighting climate change and enjoying delicious food. I’m also not trying to “own” any cultural aspect—the more folks working on this project, the happier I will be.
It’s possible there’s a better synonym—curious if anyone has thoughts.
Have you considered that by increasing the variety, availability and tastiness of tofus, you could be extending the amount of time a current vegetarian/vegan remains committed? I think that as a movement we tend to focus on new converts, and neglect maintaining those who currently avoid animal products, especially in light of evidence that many vegetarian/vegans eventually revert to eating meat.
This is a great point, Bryan! I wonder how a “support vegetarians” angle would compare to a “help meat eaters reduce” branding. More broadly, it seems like it might hinge on how helpful it is (to the long or short-term success of the movement) to grow the # of vegetarians vs flexitarian allies. I’d expect those goals to have more overlap than difference, but it seems worthwhile to at least think about.
My hunch would be to focus first on high-end flexitarian chefs, to establish that the ingredients aren’t “just for vegans,” then try to expand within the veg community. But I could be wrong.
I suspect that people who already eat tofu would be the initial base for the new-to-the-West tofus.
I’m actually not so sure about this. From a flavor perspective, rare tofus taste pretty different from one another and have different use cases, so liking one type wouldn’t necessarily be strongly correlated with liking another. That said, there might be greater willingness to try?
I’d be interested to hear more about the tofus you have in mind.
There are quite a few “great fit” varieties. The ones that seem easiest to import and have the clearest uses cases are probably:
Shanghai tofu (上海素鸡 shanghaisuji) - ultra high in protein, rich/eggy/custardy flavor, somewhat bready structure when frozen, very distinct strengths. Left whole, we’ve turned it into unique sorts of cakes, dumplings, “crostinis”, protein crumbles, etc.
Spongy tofu (千页豆腐 qianyedoufu) - a fishcake-like tofu, pre-seasoned, delicious blended or left whole as a main, can be dense and chewy or light and silky.
Fermented tofu (腐乳 furu) - a seasoning that could be as diverse in flavor as aged cheeses, but that hasn’t yet been systematized.
Yuba (油豆皮 youdoupi) - the film that forms atop heated soymilk, rich flavored, high in protein, really interesting uses as a pastry shell, probably the easiest tofu to season.
I also love tofu gan (pressed tofu 豆腐干)! We’ve found it tastes surprisingly great thin sliced and added to grilled cheese, or also cooked with apple/sugar/cinnamon spices.
There are a number of rarer tofus that I hope can one day be commercialized in the west, but that are already hard to find in China:
Juicy tofu (包浆豆腐baojiangdoufu) - a meltable tofu popular in Yunnan province.
Charcoal ash tofu (荞灰豆腐 qiaohuidoufu) - a tofu from Guizhou province that’s coated in ash and slow cooked for a few days. It’s like tofu gan, but with a very tender texture and nice smoky flavor.
Bubbly tofu (泡豆腐 paodoufu) - a thin tofu sheet from Guizhou province that’s dried then cooked in hot rocks to puff up. It needs to be reconstituted (soaked) in liquid before using and has a wonderful texture.
PS I guess you might have come across this, but the world’s top restaurant has been popularising tempeh.
This is too cool! I might have to add a rest stop post EAG…
Thanks for your pointers, Nate! I’m also curious—what sorts of tofus have you tried?
I’m with you on the challenge of building a new market from the ground up. That said, on the supply side, I’d expect importing palettes of rare tofus would be a lot easier than, say, building a production site or finding suitable co-packers for a new plant-based meat company. Especially since a lot of these tofus are already imported, just on a small scale, but from producers that have a lot of capacity. (It also helps that they can all be frozen, unlike simple tofus.)
That’s all assuming there’s demand—which like you said is tough without existing use cases. My current take, and I could be wrong, is that simple tofus just don’t have great product-market fit within western cuisines, and that this is unlikely to change in the short term. This is because of taste, textural, and cooking requirement issues that seem fundamental to firm/soft/silken tofu. (Silken tofu may be a slight exception—it seems to do well blended into baked goods and sauces, but I don’t think this will substitute for much animal product consumption.)
Rare tofus, while currently unused in western cooking, have a lot of seemingly better-fit use cases, which we’ve been looking into for our book. (From crazy yuba/tofu skin pastries, to delicious stand-alone protein mains, to natural dumplings, … ) My bullishness is more of a long-term bet, that rare tofus will eventually fare better than ordinary ones, since the culinary signs seem to point that they could. If these novel culinary uses could get chefs excited, which is something other alt proteins have had trouble with, I think that would have big trickle down effects on demand.
Interesting idea! Funny enough, black bean tofu is actually made with black soybeans, not black beans. The standard beans don’t curdle the same way—too much starch content—so they can only form starch not protein gels. Would be cool to see more of them, though, and be able to better personalize towards different restaurants