Apple (AAPL) announced its long anticipated Vision Pro mixed reality headset this week, with the base model starting at US$3,499. It will be available early next year and will be 7x the base price of Meta’s (META) Quest 3 headset (announced by Meta late last week). The difference in price is correlated to what’s on board - the chipset, resolution (~4K/eye vs 2064x2208 for Quest, sensor stack, “AR” capability ... .and the other ~5,000 patents linked to the device). Apple’s headset is powered by Apple designed chips, the M2 System-on-a-Chip (SOC) which powers Apple’s latest suite of products and devices. However, working alongside the M2 is Apple’s newest chip, the R1, which is tasked with processing all of the information from the sensors (12 cameras, 5 sensors - including LiDAR, 6 microphones) in 12 milliseconds, the speed at which Apple claim is the threshold for your brain to notice any delay. By comparison, Quest (using a single Qualcomm Snapdragon chip) does have major issues with motion sickness due to the delay.
There are a couple of other points worth expanding on:
This is VR dressed up as AR. Just like getting a new iPhone, when you register the device, you have to scan your face from all directions. This is saved as your avatar (or as Apple call it - a “persona”) so that when you’re chatting on FaceTime, Apple uses this persona (along with your real live expressions) to present a live render of your face. The same is true for your surroundings. Unlike Magic Leap which augments digital content into a real world environment, the Vision Pro’s cameras (LiDAR etc) capture and render your living room and present this to you as an AR experience. When someone looks at you, you will see them (via the outward facing cameras) standing in your view (like an AR experience) and they will see a render of your eyes (capturing live expression) on the outside of the visor. This is called “EyeSight” (see image below left next to the Pixar image).
For sports fans (and fans of live music and events), there is a lot to get excited about. Disney CEO Bob Iger presented his vision for Disney with Vision Pro. Of course this included immersive Star Wars and Marvel experiences, but also sports. As Disney owns ESPN there is a real intent for them to show immersive sports experiences via the Vision Pro. For example, buying courtside seats for the NBA playoffs (and watching replays augmented onto your coffee table). Perhaps you’d like to sit in the Red Bull F1 car as it speeds around the Las Vegas strip…..or be right up the front for Glastonbury and Coachella. It’s suggested by a number of sources that Apple have created their own cameras and platform called “Apple Immersive Video Format” to enable this. A partnership with Disney to deliver this is a compelling value proposition!
However, in saying all of this, the downslide is that we’re facing a future where we’ll have a computer strapped to our face (or embedded in our head), all of which will lead to a continued rise in online bullying, anxiety, depression and disengagement from meaningful connection. Apologies for the cynicism, but this does remind me a little too much of the movie WALL-E, where the future of humanity is overweight, sedentary and transfixed on embedded AR! Of course WALL-E was produced by Pixar Animation Studios which was founded by Steve Jobs and now owned by Disney.
Of course, if we think Mixed Reality is scary, you then have Neuralink (private) who have just filed a patent for its craniotomy machine (below) which will be used for the safe and accurate craniotomy prior to the ‘installation’ of the neuralink brain implant via the surgical robot. They highlight that a manual craniotomy (where a piece of the skull is removed to expose the brain) is a “time consuming and high-risk procedure” and that there is “little tolerance for human-caused inaccuracies in the craniotomy procedure”. This is in addition to a device implantation robot for which there is a separate patent.
Now onto other news!
The prospect of beaming solar energy directly from space is a little closer to reality, with a team from Caltech launching the Space Solar Power Demonstrator - an orbiting experiment to test space-based solar power. According to Caltech, the experiment has started and they are already seeing encouraging results. Meanwhile, The European Space Agency is drawing up their own blueprint for a possible European space solar network, whilst China have plans for a prototype by 2028. Of course there are a number of challenges relating to the efficiency of this technology, and no doubt the foil hat brigade will be having a meltdown, but it will be fascinating to see how this (and other solar technologies) will advance in the coming years.
Beijing is advocating directly to Arm (private) to work more closely with Chinese companies. Softbank-owned Arm is (alongside Taiwan Semi and ASML) one of the most important chip companies in the world as it is the Arm designs which are utilised for the most advanced semiconductors used by Apple (M2), Nvidia, Qualcomm (Snapdragon SoC) and all leading edge devices. China’s deputy minister for science and technology has advocated directly to Arm to work with Chinese universities and institutions. At present, the US, Netherlands and other countries (i.e. Japan) are cracking down on China’s access to all advanced semi technology. Such restrictions have prevented Arm from being able to share their most advanced designs with China. Meanwhile, Softbank is supposedly awaiting final approvals from China for the transfer of Arm’s China unit to Softbank, a precursor to Softbank’s much anticipated (and needed) spin-off of the company.
Softbank (9984.JP) isn’t the only investor caught in the crosshairs of strained US/China relations. This week we saw VC giant Sequoia Capital (private) splitting its China business (which was an early investor in Alibaba, Meituan, PDD and ByteDance) into a separate entity. The company will also split its Indian and South East Asian business into a third entity. Such a move is catalysed by US moves to restrict investment into China, in particular investment into deep tech areas such as AI and semiconductors.
A former ByteDance (private) executive alleges that the company passed on TikTok information about Hong Kong users to the Chinese government back in 2018, when citizens were actively protesting China’s “reunification” with the island nation. The report, published by the WSJ, suggests TikTok data including network information, SIM card IDs and IP addresses were used to identify and locate users. Such allegations are quite damning and play into the hands of policymakers in the US, UK, EU, Australia and elsewhere who are investigating risks the app poses to users. ByteDance continue to deny that any information about users is ever passed onto the Chinese Communist Party.
GitLab (GTLB) surged 30% after announcing (much) better than expected earnings and announcing a new AI product within its existing DevSecOps platform (which is the practice of integrating security testing at every stage of the software development process). However, using AI in this way is more about GitLab getting to product parity with the likes of AWS CodeWhisperer and Microsoft’s GitHub copilot (below).
According to The Information, Microsoft (MSFT) are charging users of their AI features a fee, with a reported 100 customers already paying $100k for up to 1,000 users per annum. At present there are an estimated 600 users of Microsoft’s AI features, including BoA, Walmart, Ford and Accenture. Microsoft is supposedly assessing two options to price AI in their future product. One would be to charge for a premium tier to the basic Office subscription while the other would add the AI feature to Office for all corporate users, who would then see a premium baked into all future subscriptions. The reality is that AI comes at a considerable cost to those offering it (i.e. Microsoft, Google) and ultimately we are likely to see a rise in software prices across the board as those costs (data centre, R&D) are passed on to the end consumer. This is probably fine for Microsoft and Google because they will see significant synergy from the vertical integration of data, cloud, servers and infrastructure (i.e. chipsets). However, it is likely to be a major issue for those in the middle who will rely on APIs from Microsoft and Google, but do not have the synergy (therefore their ability to pass on the cost without denting free cash flow will become worrisome). Expect to see AI accelerate the gap between big tech and the rest!
Microsoft is also looking outside of their four walls to power this increased demand for AI, looking to spend potentially billions of dollars over multiple years with Nvidia-backed CoreWeave (private), a startup which provides clients with access to over 45,000 Nvidia GPUs for AI applications. Having started life as an ETH mining venture, the advent of AI has been a godsend for CoreWeave, being able to quickly pivot demand for those GPUs from crypto to AI almost overnight.
The same can’t be said for other crypto companies….which brings me to another point (a bit of an “op-ed”!) - the blockchain industry’s flight of capital and talent. What we are seeing, and will continue to see, is the once popular blockchain sector struggling to pull in capital and retain talent (both of which will be shifting focus to AI). On the talent front, it’ll be very difficult (as a whole) for blockchain startups to retain top talent (i.e. engineers) who will now be in huge demand from AI startups (with deeper pockets and more attractive ESOPs). A quick LinkedIn search indicates that there are 17k jobs for AI-related engineers in the US vs <2k for blockchain-related engineering jobs! On that front, we are going to see VC firms getting inundated with pitch decks from AI startups (just as they were inundated with pitches from blockchain startups 12-18 months ago). The struggle for many of those startups will be showing a level of differentiation or having unique IP which isn’t dependent on a Microsoft/Google/OpenAI API/plugin. Later this month I will share quite an exhaustive ecosystem map on the ‘foundational layer’ of AI which is currently tightly squeezed on an A3. In the meantime I’ll share a preview representing ~25% of a summarised ecosystem! (nb. the dotted lines represent investment, not consolidated ownership).
Of course, along with talent and capital flight, there is also the latest regulatory crackdown, with the SEC taking action against Binance (private) and Coinbase (COIN), who collectively account for half of all global trading in digital assets. On Monday the SEC filed a complaint against Binance alleging an array of 13 civil charges like improperly mixing customer funds with those of a firm owned by the CEO. This followed action on Tuesday, with the SEC announcing that they would sue Coinbase alleging it violated US securities law by failing to register as a broker, national securities exchange or clearing agency. Coinbase have articulated their frustration, indicating that they want to see solutions rather than enforcement. The GC of Coinbase said “the solution is legislation that allows fair rules for the road to be developed transparently and applied equally, not litigation”. What we’re likely to see here is a very dragged out legal battle between Coinbase and the SEC, which will further buoy state regulators to rein in certain parts of Coinbase (and others) business.
So there you have it, a three pronged attack on crypto and blockchain - capital, talent and regulation - making it increasingly difficult for many in the industry to survive.
Thank you and have a great week and weekend ahead!
Charlie Nave
LinkedIn or E-Mail (cnave@granitebaycap.com)
Associate Professor (Practice) Monash Business School and Monash Centre for Financial Studies (MCFS)
Granite Bay Capital is an innovation focussed consultancy with a deep focus on the companies at the leading edge of innovation across major themes such as AI, ubiquitous computing, sustainability, automation and longevity. Any views expressed in this article are those of the author and do not constitute financial advice.