UPSIDE Foods has placed a $50 million stalking-horse bid for the US assets of Believer Meats, a cultivated meat company that raised nearly $400 million, built a large-scale facility in North Carolina, and obtained FDA and USDA clearance before running out of cash in late 2025.
Thank you for reading this post, don't forget to subscribe!The transaction illustrates a broader pattern of asset recycling in the cultivated meat sector, where infrastructure, cell lines, and adjacent facilities have each changed hands differently. Critically, Believer’s IP (including cell lines, media formulations, and process control software) is excluded from the sale, leaving any buyer with the physical plant but not the biological and process knowledge needed to operate it effectively.
UPSIDE Foods has bid $50 million for Believer Meats’ North Carolina plant. The interesting question isn’t the price, it’s which parts of a dead cultivated-meat company are actually worth buying.
There’s a particular kind of news that tells you an industry has changed phase. Not the funding announcements, not the “world’s first” press releases. Those are the vapour pressure stuff of a hot market. The signal that matters is when companies start buying each other’s carcasses. And that’s exactly what happened earlier this month.
UPSIDE Foods has submitted a $50 million stalking horse bid for the US assets of Believer Meats, the company that until recently held the title of operator of the only commercial-scale cultivated meat facility in the world. For anyone who hasn’t had to sit through a restructuring before: a stalking horse bid is just the opening, court blessed offer in a supervised sale. It sets a floor. Other parties have until 20 July to come in higher (and any qualifying bid has to clear roughly $52.25 million once you add the breakup fee) with an auction pencilled in for 28 July if rivals show up. So this isn’t a done deal. It’s a starting gun.
But the mechanics of the auction are the boring part. What I find genuinely instructive (as someone who spends his days thinking about what’s actually inside a bioreactor suite) is what this transaction reveals about where the value in a cultivated-meat company lives, and where it doesn’t.
A short, expensive obituary
Believer Meats was Future Meat Technologies before a 2022 rebrand, and it was one of the best capitalised names in the entire sector: Close to $400 million raised, with ADM and Tyson among the backers. It poured a reported $150 million plus into a 200,000 square foot plant in Wilson County, North Carolina, with a nameplate capacity of up to 12,000 tonnes of cultivated chicken a year. It cleared the FDA and USDA in 2025. By the standards anyone was using three years ago, that is a company that had won.
It ran out of cash in December 2025, ceased operations, and was in receivership by February. The construction firm that built the place is in court claiming it’s owed $36.4 million. A bank is sitting on a $25 million secured term loan. The lights are off in the largest regulatory-cleared cultivated meat facility in the United States.
I want to dwell on that sequence for a moment, because it’s the whole story: nearly $400 million, a finished plant, full regulatory clearance and it still wasn’t enough. Hold that thought.
Three ways to recycle a dead startup
What’s quietly remarkable is that the Believer bid isn’t an isolated event. It’s the third distinct flavour of asset recycling the sector has produced in about eighteen months, and the three of them map almost perfectly onto the three things a cultivated-meat company is actually made of.
Buy the steel. That’s UPSIDE. The company has raised over $600 million, but in 2024 it quietly shelved its own ground-up “Rubicon” megaplant near Chicago and chose instead to wring more out of its existing EPIC site in California. Acquiring a finished, inspected, 20,000 litre scale facility is the capital efficient version of building one. You skip the construction risk, the commissioning headaches, and the regulatory walk through, and you inherit someone else’s stainless steel at cents on the dollar. Smart, if your bottleneck was ever really the building.
Buy the biology and give it away. That’s the Good Food Institute. When SCiFi Foods folded, GFI won an auction for eight bovine cell lines and two serum free media formulations, then handed them to Tufts’ Center for Cellular Agriculture to validate and open source. These weren’t just any cells, they were CRISPR immortalised and suspension adapted, which is to say someone had already done the hard, multi year part of turning a primary tissue sample into something that will actually grow in a stirred tank. GFI essentially composted one company’s IP into a public good. No bioreactors changed hands. Just the genome and the recipe.
Buy something adjacent and cheap, then retrofit. That’s Aleph Farms, which a couple of years back picked up a manufacturing facility in Modi’in from a cancer-therapeutics company, VBL Therapeutics, for a reported sum in the single digit millions. The logic there is that a biopharma cleanroom built for something else is a lot cheaper than a purpose built meat plant, and the gap can be closed with a retrofit and a technology transfer from your pilot line.
Steel, biology, real estate. Three companies, three different answers to the question which part is worth saving?
The part the buyer doesn’t get
Here’s where it gets interesting for the engineers in the room, and it’s buried in the court filings rather than the press release.
The Believer asset package includes exactly what you’d expect from a walk-through of the plant: bioreactors, media and process tanks, centrifuges, freezing systems, wastewater treatment, and (crucially) the automation and production control systems. The physical capital. The things bolted to the floor.
But the receiver has drawn a hard line around the Israeli parent entity’s intellectual property, and the carve out is telling. Explicitly excluded from the sale are the frozen cell lines, the media formulation recipes, and the process diagrams and control software. In other words: a buyer can acquire the entire body of this facility and still not own its brain.
Think about what that actually means at the level of a process. You can have the 20,000-litre vessels, the SCADA layer, the chilled-water skids and the centrifuge train and none of it knows what to do. The cell line is the engine. The media formulation is the fuel chemistry. The control software and process diagrams are the gearbox and the firmware. The encoded knowledge of how this specific organism behaves in this specific equipment, the feed schedules, the setpoint trajectories, the hard-won “we run it like this or it crashes at day six” tribal wisdom that never makes it into a P&ID.
That’s the asset that doesn’t depreciate and doesn’t transfer in a fire sale. Stainless steel is fungible; you can buy a 20,000 litre vessel from a catalogue. A validated, suspension-adapted cell line paired with a chemically defined, cost viable medium and a control strategy that holds them together at scale. That is the entire game, and it’s exactly the part tangled up in litigation between a US receiver and an Israeli trustee.
The building was never the bottleneck
Which brings me back to the thought I asked you to hold.
Believer had the plant. It had the regulators’ blessing. It had something close to half a billion dollars. What it apparently did not have was a process whose unit economics close. A media cost low enough, a cell density high enough, a volumetric productivity good enough that the number on the spreadsheet for a kilogram of chicken stops being a punchline. There is a yawning gulf between “we’ve done dozens of runs at 2,000 litres” and “we make money at 20,000 litres,” and that gulf is not made of concrete or stainless steel. It’s made of titre, media cost-of-goods, and how gracefully your control strategy handles a ten fold jump in vessel volume where your oxygen transfer, mixing time and shear environment have all quietly changed underneath you.
So when UPSIDE inherits Believer’s hardware, it inherits a magnificent, FDA cleared shell. The expensive problem (making the biology pay inside that shell) comes with neither the building nor, thanks to the IP carve out, the previous tenant’s notebook. UPSIDE will be bringing its own cells, its own media and its own control philosophy into a plant tuned for someone else’s process. That’s a non trivial commissioning exercise dressed up as a bargain.
What I think it actually means
I don’t read any of this as the doom narrative the headlines reach for. Infrastructure changing hands instead of being demolished is what a maturing field looks like, not a dying one. The first wave of cultivated-meat companies effectively ran a very expensive, distributed R&D programme for the whole sector, and the survivors and second movers now get to buy the de risked outputs (sometimes the steel, sometimes the cells, occasionally both) without repeating the spend.
The lesson I’m taking, from the upstream seat, is one I’d have bet on anyway: the durable value in this industry was never going to be the capex. It was always going to be the process. The cell line, the medium, and the control intelligence that wraps around them. The things that turn a tank of warm broth into a predictable, observable, controllable manufacturing operation. Companies will keep coming and going. The stainless steel will get passed around like a hand-me-down. But the brains are what everyone’s actually fighting over in court, and that tells you precisely where to point your engineering.
Buy the building if it’s cheap. Just don’t mistake it for the company.