Artificial Intelligence – Go-getter.

Reply to this post

 

 

 

 

 

A breakthrough that Facebook badly needs.

  • Google DeepMind has reported substantial progress on one of the big three challenges of AI which is exactly what Facebook desperately needs but is unlikely to achieve anytime soon.
  • DeepMind has been able to build a new Go (AlphaGo Zero) algorithm that relies solely on self-play to improve and within 36 hours was able to defeat AlphaGo Lee (the one that beat Lee Sedol) 100 games to 0.
  • RFM has identified three main challenges that need to be overcome for AI to really come of age (see here).
  • These problems are:
    • First: the ability to train AIs using much less data than today,
    • Second: the creation of an AI that can take what it has learned from one task and apply it to another and
    • Third: the creation of AI that can build its own models rather than relying on humans to do it.
  • In my opinion DeepMind’s achievement represents a huge step forward in addressing the first challenge as AlphaGo Zero used no data at all.
  • I do not think that this represents a step forward against the third challenge as the system of board assessment and move prediction (but not the experience) used in AlphaGo Lee was also built into AlphaGo Zero.
  • Hence, I do not think that this system was building its own models but was instead using a framework that had already been developed to play and applying reinforcement learning to improve.
  • What will really have the likes of Elon Musk quaking in their boots is the fact that AlphaGo Zero was able to obtain a level of expertise of Go that has never been achieved by a human mind (see here figure 3).
  • It is almost as if the use human data limited the potential of the machine’s ability to maximise its potential.
  • That being said, it is one thing to become superhuman at Go and quite another to enslave the human race and so I continue to think that dystopia will continue to be thwarted by Dr. Moore (see here).
  • There have been many other attempts to address the data quantity problem but this is the first one that I have seen that has shown real progress.
  • Many of the other digital ecosystems have been trying to use computer generated images to train image and video recognition algorithms but there has been no real success to date.
  • I suspect that taking what DeepMind has achieved and applying it to real world AI problems like image and video recognition will be very difficult.
  • This is because the Go problem is based on highly structured data in a clearly defined environment whereas images, video, text, speech and so on are completely unstructured.
  • Hence, we are not about to see a sudden improvement in Google’s ability to recognise and categorize images and video (which is already world-leading) but the seeds are clearly being sown that will keep Google a long way ahead of everyone else.
  • This exactly the kind of advance that Facebook really needs to make.
  • This is because I have long been of the opinion that while Facebook sits on a massive treasure trove of data, it has very little idea of what any of it is or what it means.
  • This makes it very hard to spot fake news or offensive content which has been the source of many of Facebook’s most recent problems.
  • It also makes it much more difficult to understand what its users do and do not like and therefore much more challenging to tailor its service accordingly.
  • Finally, it will also make it much more difficult for Facebook to keep up with competition in terms of deep and rich services meaning that its users may begin to spend time elsewhere.
  • This is a breakthrough that Facebook badly needs but unfortunately it is Google that owns the IP meaning that it will be Google services that improve.
  • I continue to think that Google comfortably leads the world in AI but recent stock performance and the resulting high valuation keeps me indifferent to the shares.

Microsoft, Huawei & ZTE – Hardware heaven?

Reply to this post

 

 

 

 

 

3 big leaps but with potentially with fatal flaws?

  • Microsoft, Huawei and ZTE have both expanded their hardware ambitions but I question whether enough attention to details has been paid to really make these products really successful.

Microsoft Surface Book 2

  • Microsoft has launched a worthy successor to the Surface Book, substantially upping both the power and the size of the device.
  • Two versions are now available: a 13.5” device and a 15” device and on both, the hinge has been meaningfully reinforced to ensure that the screen does not wobble during typing.
  • Microsoft has included the latest Intel processors as well as graphics from Nvidia to ensure that the performance of these devices is top notch.
  • Both screens detach from the keyboard to become a tablet but it is here where my concerns lie.
  • The single biggest fault of the original Surface Book is the fact that when the screen is detached, the keyboard stops working.
  • In my opinion this removes the best use case for a tablet PC which is to turn it into a portable desktop experience. (see here).
  • This provides both a more productive and a much healthier computing experience.
  • One can attach a separate Bluetooth keyboard to the product, but when the user has already paid up for a great keyboard, this seems to be a slap in the face.
  • It is not clear if this functionality has been enabled on the Surface Book 2 but I think it will make the difference between the perfect product and one that continues to follow the obsolete laptop dogma (see here).

ZTE Axon M

  • After being very rapidly commoditised in audio, ZTE is having another go at differentiation with the launch of a dual screen device not very unlike the YotaPhone.
  • The main difference is that ZTE is using two full colour smartphone displays compared to the YotaPhone whose secondary display uses e-ink for an always on display that consumes no power.
  • The aim here is to provide the screen of a tablet in a form factor that can fit in one’s pocket rather than a back-up for when battery is running low.
  • Google Apps can recognise when the second screen is active and run in tablet mode across the two devices but how this works for other apps is unclear.
  • Furthermore, the screen bezels mean that there is a big black line in the middle of the larger display which will be very distracting.
  • I am a big believer in larger screens on pocket sized devices, but until a single screen can unfold or unroll into a large rigid display that is as good as a tablet, this segment is likely to struggle.
  • This has been tried several times in the past and every time the hardware and software compromises being made to get two screens onto a single device have fatally hurt its appeal.
  • I don’t see how the Axon M will be any different and consequently remain cautious on its outlook.

Huawei Mate 10 / 10 Pro.

  • Huawei launched its 2017 flagship products with both devices sporting edge to edge displays pioneered by Samsung and copied by everyone else.
  • The main difference other than slightly different proportions between the devices, is that the Mate 10 is LCD while the 10 Pro uses OLED.
  • However, the main differentiator that Huawei is going for this year is AI where both devices use the Kirin 970 chip, developed in house at HiSilicon which have an onboard neural processing unit (NPU).
  • The idea is that using AI, Huawei claims to be able to prevent the inevitable performance degradation that occurs on all smartphones after months of usage.
  • This aims to compete with Apple’s Bionic A11 chip that also has an NPU but I don’t think NPUs are particularly difficult to produce.
  • AIs work best on processes that are massively parallel which is why GPUs are so good at running AI.
  • This not very difficult to achieve anymore.
  • What is far more difficult, is the creation of the AIs themselves to improve the user experience and here I think Huawei is badly lacking.
  • Huawei has no real AI expertise to speak of and on its own devices it will be competing against the global leader, Google.
  • Consequently, while Huawei may be able to win some short-term differentiation by providing an optimal place to run AIs, this will swiftly be copied leaving Huawei still struggling for differentiation.
  • To really make it, Huawei has to differentiate through the AIs itself and produce algorithms that provide rich and intuitive enhancements to services running on its phones.

Yandex – Homeless.

Reply to this post

 

 

 

 

 

Alice needs a hardware home.

  • Yandex has jumped on the digital assistant bandwagon but with its history, it should be able to produce by far the best product for the Russian speaking market.
  • However, it will be unable to serve the majority of use cases without hardware to carry it into the home or the vehicle.
  • Yandex is the pre-eminent internet company is Russia with 65% market share in search and just seen off a challenge from Uber (see here) to also become the dominant provider of ride hailing.
  • Most importantly of all, Yandex has been crunching data for over 20 years, which according to RFM research (see here), is a major contributor to its RFM rating as No. 3 in AI behind Google and Baidu.
  • Consequently, a digital assistant is an obvious product to launch and is one that has a much better chance of succeeding in Russia than any of the others even if they are taught to speak Russian.
  • The assistant is called Alice and is the result of putting together a series of AI projects that the company has been working on for some time.
  • These include voice search, weather, news, maps and so on.
  • Two of the key features include:
    • First, speech recognition. Yandex claims that the assistant demonstrates near-human levels of accuracy when it comes to understanding speech.
    • This is no great feat in English anymore but in Russian, this is likely to put Yandex meaningfully in front.
    • Second, context. Alice has some short-term memory in that it remembers what the previous question was and is able to answer a follow-on question in the context of the first.
    • This is quite a difficult AI problem to solve and the only other player that I have seen do a decent job of this is Hound from SoundHound.
    • I not seen this ability in Google Assistant, Amazon Echo, Microsoft Cortana or Apple Siri.
  • Alice is available in the Yandex Search app on iOS and Android as well as in beta on Windows PC but this is not where it is most needed.
  • Usage of voice assistants predominantly occurs when the user’s hands are occupied such as in the car or in the kitchen.
  • Consequently, to address this use case Alice needs to be resident in a home speaker of some description and, potentially, in a vehicle infotainment unit.
  • Yandex has stated that there will be further products forthcoming and I am pretty certain that a speaker (probably in conjunction with a known audio brand) will be shortly forthcoming.
  • Given Yandex’s heritage in AI and its dominance in search, it looks unlikely that Amazon or Google will be able put up much of a challenge leaving the Russian speaking markets open for Yandex.
  • It will have more difficulty if it wants to expand overseas but Russian is a big enough market for Yandex to fare pretty well just by staying at home.

Amazon & Microsoft – One-way street

Reply to this post

 

 

 

 

 

Amazon gets the best of it for now.

  • Amazon looks to be the main beneficiary of the co-operation between Amazon and Microsoft which will see Alexa offer access to Cortana and vice-a-versa.
  • Amazon and Microsoft are working on a co-operation where Windows 10 users will be able to get Cortana to open Alexa and perform its range of functions.
  • Users of Amazon Echo products will also be able to ask Alexa to open Cortana and ask it to perform its various actions.
  • The idea is that users get another easy conduit from which to access Alexa while Cortana is provided with a badly needed escape from the PC where it has been stuck since the collapse of Windows Phone.
  • Cortana was originally designed to operate on a mobile device and consequently was taught how to work in a range of domains that are used on mobile.
  • The problem is that most of these domains are irrelevant on a PC and as a result, Cortana is fairly useless where it is predominantly present today.
  • This is exacerbated by the fact that Cortana has not really been taught how to work with the Office applications making the user experience for its main use case on a PC pretty poor.
  • For example, asking Cortana to read my email results in a Bing search for “read my email” and it is quicker and easier to open documents in Office with a mouse than to ask Cortana to do it.
  • I think that Microsoft’s artificial intelligence is actually better than Amazon’s as a result of the data it has been crunching via Bing but very little of this has found its way into Cortana.
  • Consequently, Amazon has come up with a better product that is far more useful in the environment where it is present (speakers in kitchens and living rooms).
  • Hence, I don’t see much of a use case for Alexa users to begin asking Cortana to do things but having access to Alexa via a PC could prove to be quite useful.
  • This is particularly the case as Alexa is very good at shopping and controlling the smart home potentially making device control remotely from the office much easier.
  • As a result, I think that Amazon is the main beneficiary of this collaboration in the first instance.
  • However, if Microsoft’s AI continues to be better than Alexa’s then there is scope for a much deeper collaboration where Microsoft’s AI could be used to power some of Amazon’s services.
  • The only problem here is that this could result in cross over between Microsoft and Amazon Web Services who are fierce competitors in the cloud.
  • Hence, a deepening of this collaboration looks unlikely at the moment but may become a reality if Amazon’s AI continues to languish.
  • Although Amazon appears to have gotten the better of this deal, I still cannot stomach the valuation leaving me with a strong preference for Microsoft’s shares.

Google – Brain boxes

Reply to this post

 

 

 

 

 

Clever devices are useless without volume.

  • Behind the carefully orchestrated event was a series of strategies aimed at driving penetration of devices which to date have been very disappointing.
  • Google made up for slightly below par hardware by maximising its leadership in AI to provide best in class functionality as well as some unique features that no one else is likely to be able to copy for some considerable time.
  • However, the key to badly needed volumes will be execution as Google completely bungled the open goal left by Samsung after its Note 7 disaster.

Pixel 2 / 2 XL

  • Google has updated the Pixel phones and has moved to OLED displays.
  • In contrast to iPhone X, Google has opted to make use of the always on display feature that allows key information to be displayed when the screen is off with almost no impact on battery life.
  • Why Apple declined to make use of this excellent feature on the iPhone X is a complete mystery.
  • What really sets the Pixel 2 apart are the new features such as Google Lens which offers the best image recognition and the fact that Google uses AI to do with one camera while everyone else needs 2.
  • However, Google openly admitted that volumes of Pixel have been disappointing and its offer of a free Google Home Mini is clearly aimed at driving badly needed volumes of this device.
  • Pricing remains punchy at $649 for the Pixel and $849 for the XL making the comparison to better looking Samsung s8 and iPhone 8 inevitable.
  • I suspect price is going to be an issue for users considering this device.

Google Home.

  • Two new products were introduced:
    • First: Google Home Mini ($49) which takes direct aim at the best-selling Amazon device (Echo Dot) in another clear attempt to drive badly needed volume.
    • Second: Google Home Max ($399) which goes up against Sonos and Apple HomePod.
  • The broadening of the portfolio should help Google increase its penetration of the home but the smart home piece is still badly lacking.
  • Google claims that 1,000 devices from 100 manufacturers now work with Google Home but it failed to demonstrate any and instead concentrated on products from Nest.
  • Google also launched routines which is exactly the same as the Amazon Echo function of the same name and something that all smart home controllers need in my opinion.
  • The integration of Google Home with other Google devices and the functionality being added is far ahead of anything else available but the smart home bugbear continues to rankle.
  • This means that anyone serious about smart home is likely to choose Amazon simply because they know that anything made for the smart home will work while the same cannot be said for Google.
  • This needs to be fixed and will remain the reason for Google’s potential defeat at the hands of Amazon because elsewhere it is by far the best product available.

Google Accessories.

  • Two companion products were launched which deepen the cross-device functionality as well as highlight Google’s core AI strengths.
    • First: Pixel Buds ($159). These take aim at Apple’s popular AirPods (also $159) and while the design looks inferior, the functionality is excellent.
    • This includes an exciting implementation of Google Translate that works with the Pixel phone to enable usable voice translation in 40 languages.
    • It also allows easy access to the best in class Google Assistant in a similar way to AirPods.
    • The difference here being that Google Assistant is a substantially better service than Siri.
    • Second: Google Clips ($249). This looks like a regular GoPro or Yi camera but the differentiator lies in its functionality.
    • The idea with clips is to position the device during an event or gathering and leave it to gather the best photos and video clips.
    • Again, this is Google using its leadership in AI to differentiate and if this feature works well, I suspect that it will be a very good reason for users to buy this product.
    • The number one use case for GoPro and Yi cameras is family despite their sporting image and it is this use case that Google is taking aim at.
    • If it works well and gains traction, this spells more trouble for GoPro which has struggled with software and ecosystem from Day 1.

Take Home Message.

  • Google has substantially deepened its cross-device capability with the new launches as these devices should all work extremely well together.
  • I think that Google comfortably leads the industry in this capacity.
  • Furthermore, much of the functionality that Google has demonstrated should make its way onto the Android devices from other manufacturers driving which should really help penetration.
  • How well they work on the hardware of others is a concern as manufacturers tinker with Android that always seems to result in inconsistent and subpar performance of apps and services.
  • Consequently, in terms of driving deeper and richer services for its ecosystem users, this was a successful event but the real question remains what volumes will Google’s own hardware achieve?
  • These services will obviously work better on Google controlled hardware where the endemic fragmentation and lack of software updates are not an issue.
  • Execution and marketing are the two things I am looking for from Google as to date, these have been woefully lacking.

Amazon vs. Everyone – Battle for the home pt. V

Reply to this post

 

 

 

 

 

Amazon goes better, smaller cheaper.

  • Amazon has raised the bar in the digital assistants battle right before Google’s hardware event on 4th October with an update to the Amazon Echo home speaker and a big upgrade to the user experience.
  • Amazon has launched two new speakers to replace the Amazon Echo, one of which is has extra hardware to optimise the smart home experience.
  • On top of this Amazon has created an over the top experience that allows the user to tie together a range of Alexa skills that it hopes will make Alexa much more intuitive to use.
  • Two new Echos have been released to replace the original.
    • First: The new echo is smaller, has better sound quality,comes in 6 colours and will cost $99 which is half of what the original did at launch.
    • It has a new microphone array that should improve audio performance in terms of noise cancellation and wake word making the overall experience less error prone.
    • Importantly, this is now cheaper than Google Home and looks set to continue what has become a race to the bottom in smart speakers.
    • Second: Amazon Echo Plus has all of the above but also includes a built in home-hub that enables the automatic discovery and set-up of devices that support Zigbee.
    • The device costs $150 but also includes a Phillips Hue smart light bulb to get the user started.
    • This will enhance the smart home functionality but it in no way covers all available devices.
    • Third: Amazon has launched Alexa Routines that allows the user to tie together a series of actions into one command.
    • The user will now be able to say “Alexa, I am going to bed” and the lights will be turned off, doors locked, TV turned off and so on all in one go.
    • It will also be possible to schedule these sorts of actions.
    • This will not work with all of Alexa-enabled devices and skills but I think it represents a further step forward.
  • With this update, I think that Amazon has achieved two goals:
    • First: It has put itself ahead of Google in the hardware race with an improved device that is now meaningfully cheaper than Google’s offering.
    • Whatever pricing Google was considering for Google Home 2 may now be quickly re-thought.
    • Second: The horrible user experience using Alexa’s skills may now take a big step forward.
    • A lot depends on how good this experience is and how well it works but if it is good, it will bring Alexa into line with what I consider to be smart home best practice (see here).
  • Amazon has had by far the most aggressive roll-out of hardware that supports a digital assistant of any of the major ecosystems.
  • There are now a total 8 different types of home device that all carry the Alexa digital assistant with a large number of third party devices in the works.
  • This is critical because a large majority of the usage of digital assistants occurs when user’s hands are busy meaning that the smartphone is almost always useless in most use cases.
  • This gives and advantage to those that provide a physical device present in the home that use an audio wake word.
  • Google has this with the home but there is only one device whereas Amazon now has 8 all which are much better at controlling the smart home than Google is.
  • This puts Google on the back foot right ahead of its launch despite the fact that it has a much better product when considering the performance of the assistant and its ability to correctly respond to enquiries.
  • Google’s response on 4th October will be key to its outlook in the smart home and based on its performance to date, I am not optimistic.
  • I continue to think that Google is at risk of suffering a VHS vs. Betamax-like defeat in the smart home.
  • I don’t like either Amazon or Google on valuation grounds preferring instead Tencent, Baidu and Microsoft.

Artificial Intelligence – Dystopia

Reply to this post

 

 

 

 

 

Humanity saved by Gordon Moore.

  • There is considerable disagreement over the dangers presented to the human race by AI, but I think that it will be the laws of physics that prevent dystopian predictions from coming true.
  • At the Tech Crunch conference, Google’s head of AI was quick to dismiss Elon Musk’s concerns that AI could present an existential threat to humans or cause a third world war.
  • Artificial super-intelligence is when machines become more intelligent than humans.
  • Top achieve this, computers need to continue evolving at an exponential rate for the next 23 years and three huge AI problems need to be solved.
  • RFM has identified (see here) these problems as:
    • First: the ability to train AIs using much less data than today,
    • Second: the creation of an AI that can take what it has learned from one task and apply it to another and
    • Third: the creation of AI that can build its own models rather than relying on humans to do it.
  • Progress against these three goals is incredibly slow and only the very best companies are making any real progress at all.
  • Everyone else claims to be working on AI but in reality are using advanced statistics to make predictions that have an improved probability of being correct.
  • Even with the best minds working on these, I think it will be decades before these problems are even close to being solved.
  • However, the real reason why I think AI will not overtake the human race comes down to Moore’s Law.
  • If one extrapolates the exponential pace of computer capability over the last 40 years, one can predict that computer intelligence will overtake that of humans by 2040.
  • This is what most of the predictions of artificial super-intelligence are based on and where much of the fear comes from.
  • However, I do not think that the current breakneck pace of Moore’s Law can continue.
  • 10nm is currently the cutting-edge geometry for semiconductors and beyond around 5nm the laws of physics start to misbehave.
  • This means that doubling the number of transistors in the same area of silicon every 18 months will no longer be possible using the transistors we know.
  • It is this doubling that has underpinned the exponential improvement in computer capability over the last 40 years and without it, I think this improvement will slow to a crawl.
  • In order to continue beyond this point a new form of transistor is required which could prove as fundamental a change as the shift from triode vacuum tubes to silicon transistors.
  • Alternatives to silicon transistors are at such an early stage of development that it seems inevitable that Moore’s Law will grind to a halt long before a viable alternative is found.
  • I suspect that this will mean that the pace of improvement of computer capability will also slow down to the point where artificial super-intelligence drops way below the visible horizon.
  • Hence while I think that Elon Musk is right to think that humans are in trouble if machines ever become more intelligent than man, it is so far away in time that Google is also right not to be worried about it.
  • Dr. Moore can be content that he has added saving the human race to his list of accolades.

Google vs. Amazon – Battle for the home pt. IV.

Reply to this post

 

 

 

 

 

Amazon increases its aggressive land grab. 

  • Not content to sit on 70% market share, Amazon is aggressively compensating for the lack of Alexa on smartphones by effectively giving the devices away and pushing e-commerce as hard as it can.
  • A land grab strategy makes complete sense because the more Amazon can drive Alexa usage, the more data it will generate and the better it can become.
  • Usage is the key to making all digital assistants better and this is the one area where Amazon has huge ground to make up compared to Google.
  • Amazon has launched yet another Alexa device which costs $20 but this is immediately credited back to the user when it is registered with an Amazon account making it effectively free.
  • The latest addition to the family is called the Amazon Dash Wand which can be used to scan bar codes or Alexa to order products from Amazon.
  • Alexa is present on the device and while this is clearly aimed at driving e-commerce, there is no reason why it can’t be used to answer inquiries or control the smart home.
  • The one thing it won’t do is play music or radio but when the whole device costs $20, it is obvious that the audio experience would not be worth the effort.
  • At the same time, Amazon is also offering $50 off the Amazon Tap reducing the price of the portable speaker to $79.99.
  • The two weaknesses of Amazon in the digital assistant space are that it is inferior to Google and that Google Assistant is present by default on every Android smartphone that ships.
  • This means that if Google can convince users to use their smartphones to access the digital assistant, then Amazon will be at a big disadvantage.
  • However, at the moment over 60% of all digital assistant usage occurs when the user’s hands are busy with another task which obviates smartphone usage as the device almost always has to be removed from a pocket to be activated.
  • This, combined with the fact that Google is still really struggling in the smart home (see here), is why Amazon still has the upper hand which it is showing no sign of losing.
  • This move is clearly aimed at seeding as much of the market as possible before Google can get its act together.
  • If a large number of households have Alexa which is working nicely with the other smart devices they have at home, it will be increasingly difficult for Google to win them back even with a superior product.
  • This is particularly relevant given that the market is still lowly penetrated in USA and is almost non-existent overseas.
  • Given Google’s very slow progress, I am increasingly of the opinion that we are witnessing a repeat of the VHS vs. Betamax battle.
  • I continue not to like either Alphabet or Amazon (even if it wins the smart home) on valuation grounds, preferring instead Tencent, Baidu and Microsoft.

Huawei – The AI of others

Reply to this post

 

 

 

 

 

Huawei needs its own algorithms to succeed in AI.

  • Huawei abandoned its habit of launching a new phone at IFA 2017 and instead focused on a new chipset called the Kirin 970 that promises all usual the bells and whistles as well as artificial intelligence.
  • Huawei made some bold claims regarding its hardware performance as well as power efficiency thanks to its 10nm geometry but I get the impression that it intends to drive differentiation through its embedded neural processing unit (NPU).
  • This is a part of the chipset that has been specifically designed to run AI algorithms more quickly and more efficiently than running them on the CPU or in the cloud.
  • The result should be faster processing of AI tasks resulting in better services that drain the battery less.
  • This is all well and good but what really matters is what users of Huawei devices will notice, to whom they will attribute the value created and for what they will pay.
  • The Kirin 970 NPU supports Huawei’s own APIs as well as Google’s TensorFlow and Facebook’s Caffe 2 meaning that AI created by these two ecosystems will also run optimally on the NPU.
  • The idea is that the algorithms are created in the cloud, downloaded to the device where they run locally improving both speed as well as privacy as the data will not leave the device.
  • I have long believed that this type of AI will be limited to functions where the algorithms are very well established.
  • In the early days this is likely to be image processing such as facial recognition or computer vision.
  • This is where I think Huawei will begin to struggle as I believe that it has very little AI of its own meaning that the Kirin 970 will spend almost all of its time processing the AI of others.
  • The AI of others will be running on the devices of all of Huawei’s competitors meaning that Huawei will be competing purely in hardware performance.
  • When other chipmakers come to market with their own NPUs, it will then be a straight fight based on hardware performance.
  • When it comes to AI, users are going to place value of the depth, richness and intuitiveness of the services themselves meaning that to improve its differentiation, this is where Huawei needs to compete.
  • Of this there is no sign meaning that while the Kirin 970 may help Huawei increase market share, it will do nothing to enable it to increase the prices of its phones.
  • The net result is that until Huawei can outsell Samsung by a factor of 2 to 1 in terms of volume, it will really struggle to increase its margins beyond the 2-4% that everyone else (except Samsung) is stuck with.
  • Google is the only company that really makes money from Android but I continue to be cautious as its valuation is already pretty full.
  • Tencent, Baidu and Microsoft remain my top choices.

 

Apple – SiriKit?

Reply to this post

 

 

 

 

 

There are very good reasons to launch SiriKit.

  • Apple has made some changes to executive responsibility for Siri that I see as a sign that things are not going very well and that changes are required.
  • Given Siri’s weakness, I don’t see much downside for Apple in allowing third parties to implement the digital assistant in their hardware presumably using an SDK called SiriKit.
  • Responsibility for Siri has moved from services (Eddy Cue) to software (Craig Federighi) which I think is pointing to much deeper integration of Siri into the Apple ecosystem.
  • The way this kind of development works is that the services are developed on top of the finished product of the software department.
  • With Siri as part of the software department it can be much more deeply integrated as the software is created and refined which should allow its functionality to be meaningfully enhanced.
  • However, what is unlikely to change is that fact that Siri is just not that smart and is easily outperformed by Google Assistant and even Amazon Alexa on occasion.
  • This is due to the fact that Siri has not been in existence for very long and that its global learning capability is hobbled by Apple’s implementation of differential privacy (see here).
  • The net result is that I think Siri is falling behind in the AI race and moving Siri to software will not really solve the problem.
  • To really improve, Siri needs to be used and this is where I think the problems really begin.
  • Usage of Digital Assistants primarily occurs when users’ hands are busy which currently means in the car and in the kitchen.
  • Apple’s position in both of these areas is quite weak and a $500 Home Pod that is nearly 4x more expensive than Google Home and 10x more expensive than the cheapest Amazon Alexa device is unlikely to help penetration.
  • Apple’s strategy to date has been to drive differentiation and desire through software that can then be monetised by selling hardware at premium prices.
  • This is why it keeps all of its software to itself but I think Siri can be an exception:
    • First: I do not think that Siri is differentiating for Apple because it is a substandard service.
    • Consequently, if it was removed from Apple products or allowed to appear on the products of third parties, I don’t think it would affect Apple’s ability to price its hardware at a premium.
    • Second: Siri is driven by AI and the AI community is far more open and collaborative than Apple has been historically.
    • For example, DeepMind published its method for creating AlphaGo which in my opinion was then immediately copied by Tencent to create its own AI Go player.
    • Apple has opened up a little bit and has begun sharing and publishing some of its methodologies for Siri (see here) which I suspect will increase over time.
  • As a result, I see only upside for Apple in making Siri available for third parties to put on their devices.
  • I think that Siri on third party hardware is very unlikely to damage Apple’s hardware business and at the same time could result in many more devices in the places where digital assistants are most used.
  • The net result would also be more data collection and learning that would help make Siri better.
  • This would represent a big departure from the way Apple has been doing businesses and there is a possibility that Apple has become too big and too set in its ways (like Nokia was) to make a departure of this magnitude.
  • Consequently, I think the probability of Apple launching SiriKit is pretty remote which is will to allow Google and Amazon to continue dividing the market between them.
  • My top picks remain Tencent, Microsoft and Baidu.