Quantum Bayesian Networks

May 28, 2019

Xanadu.ai, a quantum computing Theranos, will never make a profit

Filed under: Uncategorized — rrtucci @ 8:44 pm

Last night, Xanadu.ai, a quantum computing company, published the following paper in arXiv

“Quantum-inspired algorithms in practice”, by Juan Miguel Arrazola, Alain Delgado, Bhaskar Roy Bardhan, Seth Lloyd,

All 4 authors give Xanadu as their affiliation. The last one, Seth Lloyd, also gives MIT as an affiliation.

The main author, Arrazola, has also published a popular description of the work in his Medium blog:


This Theranos-syndrome-showing company claims that its main goal is to build hardware, a CV (Continuous Values or Variables, squeezed light) quantum computer. I am told that they just closed their second round of funding, and yet…they have never presented any hardware that can calculate anything more efficiently than a cell phone does. Even if they had, it is doubtful that such hardware could be error corrected in a fault tolerant way.

And today we learn that they spent so much time and manpower on a very abstract subject that has almost zero overlap with their CV hardware. Is this company ever going to make a profit? I am extremely doubtful that it will.

In my opinion, last night’s paper is a vanity paper for Seth Lloyd, whose obvious narcissism has been hurt by Ewing Tang’s brilliant work, which I described in a previous blog post:


Actually, the paper backfired to some extent, because Tang’s algorithm seems to work better than her complexity theory upper bounds predicted. I bet that Tang, or others, will improve her algorithm so that it works better for high rank matrices. Xanadu bozos, I hate to tell you this, but mathematical algorithms are a moving target, they almost always can be improved, especially algorithms that are so new and novel.

Despite this silly paper, the fact remains that Seth Lloyd’s Machine Learning algorithm is as dead as a door nail. It always has been, even before Ewin Tang’s brilliant work. I say this because Lloyd’s ML algorithm uses something called QRAM. QRAM is a very distant pipe dream, even more distant than far easier Fault Tolerant quantum computing, which is itself a distant dream.

To summarize, for Xanadu to be spending so much time and manpower studying a distant pipe-dream algorithm that is irrelevant to their business, instead of working on their under-performing or non-existent hardware, does not bode well for the company. I am extremely doubtful that Xanadu will ever make a profit, or make a device that calculates anything better than a cell phone does. Xanadu investors: you are being fleeced by your VCs and by Très Void.

Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: