Q&A: What does the Facebook acquisition of CTRL-Labs mean for Neurotechnology and Augmented & Virtual Reality?

__________

While sit­ting at din­ner with an unsus­pect­ing friend, my phone blew up with odd­ly-late mes­sages from fel­low neu­rotech­nol­o­gy com­rades. CTRL-Labs announced their impend­ing acqui­si­tion by Face­book and inte­gra­tion into Face­book Real­i­ty Labs (respon­si­ble for Ocu­lus et al.), for enough mon­ey to buy my guilt-induc­ing­ly-expen­sive New York cof­fee for a mea­ger 550,000 years. Woah—so very many ques­tions. I fig­ured I’d share (and con­sid­er this an open invi­ta­tion to reach out and throw in your 100 bil­lion cents-worth).

As a quick note before pro­ceed­ing, aside from its prag­mat­ic com­mu­ni­ca­tion impli­ca­tions, I don’t much mind whether or not CTRL-Labs and oth­er EMG tech­nolo­gies hold philo­soph­i­cal cre­dence as true neu­rotech­nolo­gies; what mat­ters is that they’re brand­ed as such. It’s a debate that’s been waged else­where, so I’ll leave it be!

Q: What does this mean for the neurotechnology field?

Consumer neurotech

• Con­sumer neu­rotech­nol­o­gy has been prop­er­ly legit­imized: there will now be an entry in the tech­nol­o­gy ver­nac­u­lar for “that neu­ro com­pa­ny FB bought.” Neu­rotech will be a true cat­e­go­ry, and will prob­a­bly begin to appear on the web­sites of the more dar­ing ven­ture cap­i­tal firms. This is non-triv­ial: when Face­book pur­chased Ocu­lus in March of 2014, what fol­lowed was a huge influx of invest­ments into AR/VR (the suc­cess­es of which have admit­ted­ly been ques­tion­able at best). As many VCs are quick to note, VC is a social dance, and bil­lion-dol­lar sur­prise exits to a FAANG com­pa­ny is a pow­er­ful way to get that game started.
• I’ve spo­ken to sev­er­al entre­pre­neurs in con­sumer neu­rotech­nol­o­gy today and, as expect­ed, all of them are excit­ed. This is a win they can point to; a vision they can show might lead to a con­crete out­come. It is and will remain moti­vat­ing for entre­pre­neurs, their employ­ees, and their present and future investors.
• Even though I don’t per­son­al­ly mind it, this acqui­si­tion is going to cause a has­sle in nomen­cla­ture. To actu­al­ly under­stand how CTRL-Labs dif­fers from a cen­tral ner­vous sys­tem-BCI requires a bit of basic anatomy/physiology knowl­edge. For peo­ple not in the know (i.e., almost every­body, includ­ing most VCs and engi­neers), this adds an addi­tion­al lay­er of complexity/misdirection to how peo­ple will think about neu­ro. Absent detailed infor­ma­tion, humans employ under­spec­i­fied mod­els of seman­tic domains; that’s what will prob­a­bly hap­pen in con­sumer neu­rotech­nol­o­gy. In the same way, due to my igno­rance of quan­tum com­put­ing, any piece of infor­ma­tion I learn about quan­tum com­put­ing will apply to the entire field of quan­tum com­put­ing, where­as in real­i­ty, it’s just anoth­er piece in a com­plex puz­zle. I sus­pect the same will hap­pen in neuro.
• Con­trol (as opposed to affec­tive com­put­ing and the like) is going to be the flag­ship of neu­rotech­nol­o­gy, because that’s what CTRL-Labs does. It’s pos­si­ble that investors, jour­nal­ists, and tech­nol­o­gists will, on aver­age, fall prey to the same under­spec­i­fied mod­el trap and col­lapse all “use-cas­es” into con­trol. (If you want to get a sense for how I cre­ate a tax­on­o­my for con­sumer neu­rotech, you can skim this).

Medical neurotech

I’m will­ing to wager that more of the med­ical neu­rotech­nol­o­gy com­pa­nies I know and will get to know over the com­ing months are going to empha­size the con­sumer-crossover angle. “We’re a med­ical device com­pa­ny now, but we’ll have a patient-fac­ing app which would fit in nice­ly with the health data focus at [insert your com­pa­ny of choice].” It’s a rea­son­able notion, and it’s very rea­son­able for entre­pre­neurs and investors to be enticed by the idea; who doesn’t like the sound of a $1B pre-rev­enue acquisition?

Venture capital

• Well, for one, there’s been a large neu­rotech­nol­o­gy exit now. That’s enticing.
• This is almost cer­tain­ly going to put neu­ro onto the radars of gen­er­al­ist VCs and/or spe­cial­ist VCs who just haven’t spe­cial­ized in neu­rotech yet but have plen­ty of adja­cent knowl­edge in con­sumer elec­tron­ics, tech­ni­cal­ly inno­v­a­tive hard­ware, med­ical devices, AI of var­i­ous sorts, etc. This is a good thing, but neu­rotech founders should think care­ful­ly about how to approach the edu­ca­tion of non-neu­ro (read: most) investors. Have primer mate­ri­als pre­pared, have cur­rent investors on hand will­ing to help in the edu­ca­tion­al process, and prac­tice explain­ing neu­rotech­nol­o­gy to peo­ple in a way that’s impar­tial and only as infor­ma­tive as your audi­ence wants.

Q: What does this acquisition get Facebook?

• A bunch of tal­ent. A bunch.
• I don’t think this acqui­si­tion was about neu­ro; this acqui­si­tion was about win­ning AR/VR (note: Aug­ment­ed & Vir­tu­al Real­i­ty). As some­one who pre­vi­ous­ly worked on a fair­ly intri­cate VR inter­face, I’d argue the mod­ern miss­ing piece in AR/VR is the abil­i­ty to have degrees of free­dom of con­trol com­men­su­rate to what we have over our laptops—i.e., the same degrees of free­dom of our hands. That’s what Face­book got with this acqui­si­tion: the erad­i­ca­tion of the bar­ri­er between VR and full-depth, intri­cate appli­ca­tions. To quote a close friend who works in AR/VR, in ref­er­ence to this acqui­si­tion, “It’s an arms race out here” and Face­book just found its new arms in the arms (get it? EMG?).

Q: What are the characteristics of EMG use-cases? Pros/cons? Does it obviate the need for head-worn BCIs?

• In prin­ci­ple, EMG will even­tu­al­ly give your com­put­er access to any­thing it wants to know about your hands. If we’re con­strain­ing our­selves to the phys­i­cal con­trol mech­a­nisms we’ve evolved to use, then hand kine­mat­ics recon­struc­tion is as good as it gets.
• EEG (or oth­er non­in­va­sive modal­i­ties), on the oth­er hand, enable inter­ac­tions we haven’t evolved; where­as as hominids we’ve always used our hands to manip­u­late things, we have not lit­er­al­ly used our brains. The ques­tion is whether non­in­va­sive sens­ing of con­trol sig­nals that don’t cooc­cur with actu­al or intend­ed phys­i­cal motion ulti­mate­ly leads to as many degrees of free­dom as hand manip­u­la­tion, and what sorts of manip­u­la­tions are bet­ter-suit­ed to abstract con­trol (brain) vs. evolved con­trol (hands/arms).
• EEG is bet­ter for motor impair­ment use-cas­es because it doesn’t rely on neur­al sig­nals prop­a­gat­ing down the spinal cord.
• We still need cen­tral ner­vous sys­tem neu­rotech­nolo­gies to han­dle cog­ni­tion and emo­tion. To appre­ci­ate the sig­nif­i­cance of this, con­sid­er the trend (with all its com­plex pos­i­tives and neg­a­tives) toward under­stand­ing the men­tal states of users in fin­er-grained detail so that we can adapt tech­nol­o­gy, and even phys­i­cal envi­ron­ments, to these states. There are a big class of prob­lems that impli­cate cog­ni­tion and emo­tion, and EMG won’t help us there.
• So, no, CTRL-Labs does not obvi­ate head-worn BCIs!

These are my ini­tial thoughts. They’ll change and become more detailed over time. Mean­while, please reach out and share your lat­est and great perspectives.

– Avery Bedows is the Neu­rotech­nol­o­gy Spe­cial­ist at Loup Ven­tures, a ven­ture fund invest­ing in fron­tier tech com­pa­nies automat­ing the world and build­ing new ways to expe­ri­ence it. Dis­claimer: I active­ly write about the themes in which we invest or may invest. My opin­ions here are not intend­ed for use in mak­ing any invest­ment deci­sions; pro­vid­ed sole­ly for infor­ma­tion­al purposes.

 

News in Context: