Tenstorrent Didn't Sell: An Offer Is Someone Else's Appraisal

On Monday, Tenstorrent, the Toronto-founded AI chip company run by veteran engineer Jim Keller, closed a $1.5 billion funding round that values it at $6.2 billion. That's more than double the $2.6 billion it was worth in late 2024, TechCrunch reported. The number is impressive. It isn't the interesting part. The interesting part is what the company walked away from to get here.

Months ago, Bloomberg reported that Intel and Qualcomm had shown interest in buying the company, and the tech press even ran numbers on those talks. Tenstorrent didn't sell. According to Bloomberg, Keller said the company chose to stay independent and raise more money after its Galaxy servers proved they could run DeepSeek models. In other words, they didn't say no on principle. They said no once they had proof.

An offer is someone else's appraisal

When someone offers to buy you, whether it's your company, your time or your loyalty, they're telling you what you're worth to them, today, based on what they know and the margin they expect to make off you. That number isn't your value. It's their math. And their math always includes a profit that, by definition, won't be yours.

Most people mix the two up. They get an offer and treat it like a verdict: "If that's what they're offering, that's what I'm worth." It sounds humble. It's really just handing off the one question nobody should answer for you: what could you build if you didn't sell? An asset isn't worth the same in every pair of hands; it's worth whatever system it completes, and the buyer always knows which gap you fill better than you do.

Saying no requires a real alternative

Here's the part the independence speeches usually leave out. Turning down an offer isn't an act of courage. It's a calculation, and you can only make it from a specific position. Tenstorrent didn't say no with a pretty plan in a pitch deck. It said no with a product that worked and investors willing to put money behind that evidence.

Without an alternative, "no" is theater. The employee who turns down a promotion that would chain him to a job he hates can only do it if he has another source of income or enough savings to buy time. The freelancer who refuses to be absorbed by her biggest client needs other clients. The partner who won't let the other one "handle everything" needs to know how to do his share. Sovereignty isn't declared. It's backed.

And don't confuse independence with pride. Some people turn down every offer because they can't stand the idea of anyone else profiting from their work, and they burn years and capital for the satisfaction of owing nobody anything. That urge to never depend on anyone only looks like sovereignty from the outside. The question isn't whether to accept outside help or capital. It's how much control you give up when you do.

Independence still has owners

Tenstorrent didn't end up without owners. It ended up with different ones. The round was led by Trace Capital and Innovation X, with Fidelity and T. Rowe Price among the other backers. Every dollar in that round comes with expectations, deadlines and pressure to deliver. Other people's money comes with a calendar, even when it doesn't come with a boss.

The difference between selling and raising money isn't moral. It's structural. When you sell, you hand over the final say. When you raise capital, you hand over a share of the outcome but keep the wheel as long as you deliver. Neither option is clean. What changes is who sets the course when things go wrong. For anyone building something they consider their own, that's usually worth more than the gap between two numbers.

There's also no guarantee it was the right call. In three years Tenstorrent could be worth far more, or it could regret not cashing out when it had the chance. A strategic "no" isn't judged by the outcome, which nobody controls. It's judged by the quality of the reasoning behind it.

The test before you sign

Outside the chip world, the everyday version of this decision shows up more often than you'd think: the company that wants to buy your side project, the employer offering stability if you drop your own thing, the partner or family promising to take care of you if you accept their terms. Before you answer, ask yourself four questions:

• What does the buyer see that I'm missing? If they want you, you solve something for them. Figure out what before you name a price.
• Do I have evidence, or just hope? Saying no with a product that works is strategy. Saying no with a dream is a bet.
• Which decisions do I stop making if I say yes? The real price of a sale isn't what you get. It's what you'll no longer get to decide.
• Can I hold the no? If you'd be asking for the same offer again in six months, you're not turning it down. You're putting it off.

Independence isn't a virtue you have. It's a position you pay for, defend and prove. And it almost always starts with refusing to accept, as fact, the number someone else wrote on you.

The takeaway

The idea: A buyout offer is someone else's appraisal of you, based on what they expect to gain. You can only turn it down if you have an alternative to back it up.
The fact: Tenstorrent raised $1.5 billion at a $6.2 billion valuation after choosing to stay independent, following interest from Intel and Qualcomm in buying it (TechCrunch, Bloomberg).
The PRAXMA mindset: "If you don't know your worth, you take the price you're given. If you do, you need proof to say no."

— PRAXMA, Logic over emotion

If this topic interests you, there's a longer exploration of how to decide clearly when money and time are on the line Spanish only. We also talk about it on the podcast.