Hacker News Reader

Comment by gwbas1c | Hacker News Reader

> Ultimately, InventWood is planning to use wood chips to create structural beams of any dimension that won’t need finishing. “Imagine your I-beams look like this,” Lau said, holding up a sample of Superwood. “They’re beautiful, like walnut, ipe. These are the natural colors. We haven’t stained any of this.”

Show the [beep]ing picture!

View on HN

The picture at the top of the article appears representative of the surface appearance: https://www.inventwood.com/superwood-beams

I know it's practically cliche to say so these days, but that picture is giving strong AI vibes. Look at the joints between the big beams. If you needed beams that large to hold up a building, you wouldn't cut them in the middle like that. And you wouldn't have those gaps.

Plus, all the text on the page is future tense, talking about what the super wood beams will be able to do.

So I don't know how much that picture really represents what the wood will look like.

I'm almost positive it's AI generated. Look at the chairs at the kitchen island in the background. No human would draw chairs that look like that.

The picture at https://www.inventwood.com/superwood-facade also looks AI-generated.

The fact that they're using AI images on their landing pages does not give me confidence in the quality of the product.

What about this one? https://www.inventwood.com/superwood-beams

There are lots of posts from them on the internet image search...and it seems they've gone to a number of industry conferences with samples. I agree, that from a PR perspective, their pictures are too polished. They should show a piece in a less polished format.

At this point I'd say I have <5% confidence any image on their site is of an actual sample as opposed to a 3d render, or worse, AI slop.

Edit: Even the photos of wood on a truck is obviously AI garbage when you zoom in: https://images.squarespace-cdn.com/content/v1/67160715a896dc...

wow that image is pretty bad; all sorts of weird inconsistencies in the wood stack that scream AI; I'm ignoring the license gibberish as that could be just removal of identifying info.

What about this: https://www.newhometrendsinstitute.com/consumer-insights/ind...

the image is low resolution, but appears to be from an actual trade show, and an image not from the company.

What is going on with the window reflections (for superwood-facade)? It doesn't look right.

Some of them literally have "visualelectric" in the filename.

(ex: https://images.squarespace-cdn.com/content/v1/67160715a896dc...)

There’s also an image a bit down the page:

https://www.inventwood.com/technology

But I’m not sure if it a real photo or some sort of… analogy? Under “INTRODUCING SUPERWOOD: THE NEXT GENERATION OF BUILDING MATERIALS,” we can see one piece that looks like an actual piece of wood, and then some other ones that looks like an artificial wood-looking finish wrapped around some core which visually looks metallic (but which, I assume, is actually their engineered wood).

I dunno. I guess “being wood” is more about the ecological impact in this case. If it is replacing steel and it looks ugly, well, a lot of steel looks ugly and then gets covered up.

Oh it's 100% AI generated. Look at the back brown leather chair to the right. It's supposed to have a blanket on it, but instead it's a garbled mess

Yeah, inspecting the picture closely there are at least 20 to 30 things that immediately stand out as "AI did this". There are so many things wrong with those chairs on the right. The wood support beams are just not right, nobody would build like this.

That really looks like it's AI generated. Not a good look for a company claiming to make something!

That's an AI generated image

They didn't even bother to AI-upscale it to get rid of the artifacts

This is so spot on. The fact that there isn't a single image showing an ACTUAL comparison, even of small samples, is a massive red flag for a company that is advocating the "aesthetic" properties of it's product.

Then, even worse, they RELY ENTIRELY ON UNLABELED AI GENERATED IMAGES.

I cannot imagine a better way to say "there's a high chance everything we've promised is fake news". Just make sure to squint before looking at our ad copy!

This video is what you're looking for:

https://youtu.be/96Dz-rQtGxI

Here's someone from the InventWood team talking about the product, uses, future, and more at a trade show.

They show a hands-on demo of the wood product and give an animation of the process.

The talk is pretty illuminating. They think this will even be used in the automotive sector in places where carbon fiber is currently utilized.

This is way better than the press coverage and actual website.

Thanks for the link! As pessimistic as I am toward this company it sounds like something that could be quite neat if it actually exists.

However, I lost all hope (of the little I had) after their sales rep said they "densify" their wood. My dude, if you need to spin so hard you pull out a thesaurus for "compress" it's not a great look.

Because they don't just compress their wood! That's the step after putting it in a vacuum and baking it and heating it up so that the lignin depolymerizes. The video does a great job of explaining how this is different from just sticking wood into a press.

> However, I lost all hope (of the little I had) after their sales rep said they "densify" their wood.

You lost hope because they used an industry standard term?

Sure did. You're right it seems to be in use widely (in the wood compression industry), not that it seems to have any distinction from compression... (I say from my uninformed perspective)

E.g. it feels along the lines of "we didn't just heat it, we THERMO-MODIFIED (tm) it." No matter they mean the same thing, doesn't the second one just sound cooler?!? /s

Same ilk as "deconstructed" food, "artisanal" crafts, "synergy", ...

You’re sure eager to judge considering how much knowledge you lack on the subject.

I definitely am negatively biased toward anything I perceive using buzz words or jargon. I doubt I'm the only person with this knee jerk reaction (I say without any statistics or evidence I can link in).

Same with any product that is over or deceptively advertised.

If their stuff was good, provably so, then they'd prove it on their own merits. They wouldn't need to fill a website with 99% AI slop if they had anything else of any substance.

Or, it's definitely possible I'm in the % of people who is overly sensitive to bullshit, and I fall outside the target audience of their AI slop. Similarly to how spam emails include typos to filter out people who are actually paying attention.

This is an extreme overreaction to usage of a regular word that literally describes what they are doing. "Compress" would have been less accurate.

Say what you will about the website but the video is a pretty reasonable sales demo.

> I definitely am negatively biased toward anything I perceive using buzz words or jargon.

Unfortunately every profession with any depth of knowledge will have jargon. Even house staff at a restaurant will talk about 2-tops and 4-tops, etc. It's just jargon, or better yet, convenient abstraction to describe the problem space these people live and breath.

The more layman’s wording you use, the less accurate you will be. The reason for industry specific “jargon” is just that those are words that happen to be the most accurate and descriptive words for whatever materials or processes used.

If you use too much jargon, people outside can’t understand what you are talking about. If you use too little jargon, then whatever you are talking about will be too vague.

cooked and warm are of course, indistinguishable.

in a mechanical sense, a compressed object goes back to normal when the stress is removed, unless you break it. if the work piece did not go back to its original shape, you arent just conpressing anymore. eg. you are likely "cold working" it instead.

if youre applying heat, you might be "forging" or "casting" while applying that compressive force.

theyre all quite distinct

People use more specific terms because they mean something more specific than the generic term. Often within domains of expertise, people know the topic extremely well so they naturally use specific terms, and they also need to speak very specifically.

Densify is a word: https://duckduckgo.com/?q=define%3A+densify&ia=web

The more exciting something would be, the more likely it's BS. It's what survives.

Projects just make it a lot further that way. The consumer wants to believe. The doer has to believe. The invester will take a hit on the cheek for the moon.

Didn't intend to watch all of it but I did! Thanks for the link

Looks awesome, I want it already ;)

This is a relaunch. Here's their launch video from May 2024.[1] The video has some CG of a piece of their wood, plus stock shots of trees, cities, wind turbines, people looking through microscopes, etc.

U. of Maryland announced this specific process back in 2018.[2] They got a paper into Nature.[3]

Densified wood is over a century old. [4] You can buy something called "Lignostone".[5] But that is a material where they took the lignin out and put some plastic in. (Remember "transparent wood" from a few years ago. Same concept, with a transparent plastic.) This new process takes the lignin out and compresses the cell structure.

If you could send in $20 and get a little cube of the stuff to look at and test, this would be more convincing.

[1] https://www.youtube.com/watch?v=JiHABi97rU8

[2] https://www.thechemicalengineer.com/news/new-densified-wood-...

[3] https://www.nature.com/articles/nature25476

[4] https://jwoodscience.springeropen.com/articles/10.1186/s1008...

[5] https://www.roechling.com/us/industrial/lignostone

Where are you seeing AI generated images? https://www.inventwood.com/ seems to have several real pictures of the product.

Yes the final product will preserve the wood grain in some form. The paper has a few images of the final product. They basically boil out all the non-cellulose and compress what remains into a smaller form. I assume the strength is that a super board of the same size would be made of several boards' worth of wood fibers. I haven't read enough yet, but I wonder if this substantially lowers weight / strength. Right now, steel is required for tall buildings because there is an upper limit to how high you can go before wood buckles. There are no 300 m trees after all.

fwiw, I first thought this was going to be some innovation in glue, which allows manufacturers to turn wood chips and sawdust into MDF, OSB, and particle board. These are typically weaker than sawyered beams of the same size because the glue is not as strong as the cellulose fibers that run the length of a beam. (However, they are getting applied more and more in the construction in the US because it is crazy expensive to find a tree that can produce thick enough 40 ft beams, but you can easily get enough sawdust to make an even thicker, precut MDF board.) But I thought no shot, because if you can make glue that is stronger than cellulose, then why bother with the wood?

Someone posted what's probably the study it's based on below, it looks to be Figure 2d and 10e. As suggested, it's just darkened in a nice looking way (without stain) which frankly doesn't seem like it's necessarily a good thing.

https://www.fpl.fs.usda.gov/documnts/pdf2018/fpl_2018_song00...

wasn't this a picture:

https://techcrunch.com/wp-content/uploads/2025/05/SUPERWOOD-...

Normally when you quote someone, and they are making reference to a picture, the picture is logically close close to the words referencing the content in the picture.

I meant wasn't that a non-ai-generated picture of the actual product?

"The company treats it with 'food industry' chemicals"

Reply on news.ycombinator.com