“I don’t love the idea of being [an artificial intelligence] house cat, but what’s the solution? I think one of the solutions that seems maybe the best is to add an AI layer.” – Elon Musk, Code Conference 2016

An AI layer to your brain, he means. Not happy with simply improving technology that’s only been around for the last century, innovative entrepreneur Elon Musk has his sights set on the way humans communicate, something that hasn’t been vastly improved on in over 50,000 years of evolution.

Musk has referred to something called “neural lace” several times recently, most notably at Vox Media’s Recode Code Conference in June 2016; however, not many details were known about how Musk envisioned this technology being implemented. You know, the Musk way of doing it. He suggested at the conference that he might be willing to tackle the challenge himself, and a few months later, teased a few times that he was in fact working on the idea.

Making progress. Maybe something to announce in a few months. Have played all prior Deus Ex. Not this one yet.

— Elon Musk (@elonmusk) August 28, 2016

Maybe next month

— Elon Musk (@elonmusk) January 25, 2017

The announcement came in the form of a startup called Neuralink Corp, the initial details for which were originally reported by the Wall Street Journal. He seemed to have collected some impressive scientific minds and combined them with personal funding to initiate the company’s work. Other than speculation about what types of products could be created by the company and Musk’s initial idea of a direct interface with the brain’s cortex, not much information was available. More recently, a few more pixels were filled in on Musk’s vision.

Tim Urban of WaitButWhy.com, one of Musk’s preferred correspondent contacts (i.e., Internet writers), has been selected again as the person given the most in-depth information and access to Musk in order to publish a very detailed piece on what Musk has been up to.

Urban previously discussed and published details about Musk’s work on SpaceX’s vision for Mars and Tesla (with lots of direct access to Musk himself), and now has published a very long, yet very informative, piece on Musk’s NeuraLink company. He calls the company’s overall goal a “Wizard Hat”, and after seeing how much access Urban had to Musk and his new Neuralink team to gather information, that label is probably pretty accurate.

In Urban’s piece, he focuses on understanding what the business side of Neuralink will involve, as it’s the business models of Tesla and SpaceX which enable and drive their innovations. “We are aiming to bring something to market that helps with certain severe brain injuries (stroke, cancer lesion, congenital) in about four years,” Musk is quoted as saying.

The incredibly complicated nature of the human brain, a multi-million year biology project in the making, unsurprisingly presents numerous challenges for scientists wanting to direct the flow of information into and out of it. Understanding all of the details of “how” the brain functions isn’t the challenge, though. It seems to come back down to engineering. As summarized by Urban, after some 1,000 interviews with multi-disciplinary (and amazing) science people, Musk put his team together and Neuralink Corp. was born to start working on it.

The first major challenge described in Urban’s piece is the concept of bandwidth, or rather, how many brain neurons (cells in the brain which essentially provide the 1’s and 0’s of brain signals) can be read by electrodes at a time. He quotes the Neuralink people as needing around a million neurons to be read in order to really achieve something revolutionary.

If you’re familiar with computer chips at all, the comparison to Moore’s Law is a decent metaphor here. According to this law, the number of transistors on a computer chip doubles every 18 months, and this has led to computers becoming both smaller and faster. If you liken “transistors” to “electrodes”, you can see the engineering challenge for neural lace companies.

Then there’s also the question of whether people are going to be willing to let their brains be voluntarily experimented with. Musk’s cult following might give him a trust advantage for seeking out willing participants, but skull surgery may turn out to be too much even for them. According to Tim Urban, the Neuralink team is acutely aware of this concern, and has thus made “non-invasive” implantation a huge focus for brain-interface technology to really take off. Also of issue is accessibility to the technology to make the implantation possible. In Urban’s discussions with him, Elon Musk likened the technology needed to what Lasik surgery machines do.

In summary, in order for Neuralink Corp. to achieve the innovative leap which will change the world forever with direct brain-interface technology (the “Wizard Hat”), they’ve got to make electrode manufacturing about as advanced as computer chip manufacturing, and they’ve got to be able to install whatever electrode device is developed into brains in a very non-invasive, automated way. Also, they will need to figure out brain-friendly WiFi, some serious miniaturization solutions, and develop a “neuron signal” to “human language” dictionary.

Easy peasy, lemon squeazy. [Yeah, that was sarcasm.]

Quite honestly, it’s not the medical procedure that concerns me, but rather the potential of not being able to block spam that has direct access to my brain. Between Minority Report and my daily email battles, yikes! Sure, there are already brain-implanted devices that solve problems; however, I think there’s a difference between correcting functions the brain is supposed to have and giving something unfiltered access to adding something that wasn’t already there. I can put my phone away if I don’t want to deal with a Twitter freak out deluge. I can’t exactly do that with my brain. You know, just saying. The WaitButWhy piece gave me even more reasons to worry, so I feel justified.

Tim Urban’s piece also detailed some pretty amazing things that could come out of the neural lace field that sound like science-based versions of telepathy and magic. The ultimate goal, though, was to enable human brains to be as functional as artificial intelligence in order to avoid all the pitfalls of superintelligent AI. Actually, to be a bit more clear on Elon Musk’s vision for all this brain-interface technology, he wants the interface to connect to a super-human-collective AI cloud which feels just as much a part of you as any other part of your brain does.

For instance, when you have a thought, you don’t consider which part of your brain’s anatomy created it. It just happens and you consider it a part of your being. Imagine a super computer as part of that “you” system, and congrats! You’re [kind of] getting where Musk is headed with Neuralink. Or at least that’s the long term goal of what he’s starting with the company. You know, kind of like the moving the baton forward thing he aimed for with SpaceX and getting to Mars.

I’m going to quote Urban on this, actually:

He started Neuralink to accelerate our pace into the Wizard Era—into a world where he says that “everyone who wants to have this AI extension of themselves could have one, so there would be billions of individual human-AI symbiotes who, collectively, make decisions about the future.” A world where AI really could be of the people, by the people, for the people.

Where Neuralink will come down amongst current competitors already in the field (Facebook, Braintree, etc.) is obviously yet to be seen, but it’s yet another reminder that when Elon Musk says there’s a challenge needing to be solved, there’s a good chance he’s not going to wait for someone else to do it.