Memia 2021.15: Wanted: frontier productivity📈// MMMMeta manufacturing🦾// Spot the robocop🐕// on yer e-vélo🚴// the autodidactic universe🌌🤯
Skip hip data to get the antimatter / Blue lines are the reason why the temple had to shatter
Welcome to this week’s Memia newsletter, my regular scan across emerging tech and the unfolding edge of the future, viewed from Aotearoa New Zealand.
Aotearoa seen from outside:
World first #1: New Zealand introduces mandatory climate-related impact disclosures for large financial firms:
"We simply cannot get to net-zero carbon emissions by 2050 unless the financial sector knows what impact their investments are having on the climate"
- Climate Change Minister James Shaw
World first #2: Smoke-free nation: New Zealand proposes smoking ban for people born after 2004. (Again, Bravo.👏)
And the opening of the New Zealand-Australia ‘travel bubble’ lays groundwork for airlines’ recovery around the world (🤔…wait and see what happens if/when another community outbreak occurs…)
But, welcome back Aussies!😝
🦠More Covid vaccination momentum:
The UK began to emerge out of lockdown, reaching 50% (1st) vaccination rate and with one prominent scientific model declaring that “herd immunity” reached on 12th April.
Tiny Himalayan country Bhutan vaccinated 93% of its adult population (62% of 800,000 people) in only 16 days. (Clearly guided by maximising Gross National Happiness).
Wanted: frontier productivity📈
Some pretty blunt home truths in this week’s New Zealand Productivity Commission final report from its Frontier Firms Inquiry.
Comparing Aotearoa’s “Frontier Firm” performance with other “Small Advanced Economies” (SAEs) - eg Denmark, Singapore, Sweden…) in 2021 is sobering:
Newly installed Commission chair Ganesh Nana sets a more urgent and action-based tone than his predecessors:
“New Zealand needs to change key aspects of the status quo, to lift the performance of its firms and lift national productivity…Innovation is the key. With it, we have a chance to build a world-class competitive advantage in some markets. Without it, products and production processes become standardised and leaves us trying to compete against lower‑wage economies. Successful SAEs (such as Denmark, Singapore and Sweden) focus their investments on creating world-class innovation ecosystems around their leading firms.”
I haven’t had time to read the whole report but three key headlines stand out for me:
Exporting specialised, distinctive products at scale is the best way to build world-leading firms. Scale, through large “anchor firms” and/or through collaboration among smaller firms, is needed to earn returns on investment in innovation and exporting.
Innovation is key. The Commission recommends learning from international experience and overhauling New Zealand’s innovation ecosystems to enable firms to innovate and export at scale.1
Government investment should be focused on areas of existing or emerging economic strength and competitive advantage.
Hopefully “Nana’s Mana” will ensure this is not just another Productivity Commission inquiry which sits worthily on a shelf while the productive economy melts further into a speculative housing bubble…
On the subject of productivity, I’ve been meaning to write down some ideas about “meta manufacturing” for a while.
I do a fair amount of innovation strategy work with exporting businesses in Aotearoa. I’m often struck - particularly with firms in the primary sector - by the low position in the global value chain, but also the lack of motivation (imagination?) to move upwards. From milk to fruit, seafood to forestry, many established Aotearoa industries seem to be content to continue operating primary-product-centric businesses right at the bottom of the global value pile. I’d observe as a country that nearly *all* commercial sector innovation investment is Horizon 1 - do what we do already, just faster - and any capital-intensive innovation investment competes head-on with imported low-wage labour and short-term cashflow concerns.
IM[H😇]O Aotearoa’s producing firms needs to aggressively leave this conservative short-termist mindset behind - instead, take all of the accumulated expertise and experience which goes into growing milk, meat, kiwifruit, pine trees, whatever…. package it up into high value software / automated manufacturing / services… and sell that instead to other growers worldwide.
I came up with a “MMMMeta” model to try to explain better:
M: Machines are needed to make “stuff” more efficiently
MM: Meta-Machines which make machines (factories to make machines, or software to automate them)
MMM: Meta-Meta-Machines which make machines which make machines (eg 3D printing, generative AI models, completely autonomous “fabs”)
MMMM: Meta-meta-meta-Machines - applied nanotech which leapfrogs traditional manufacturing to just nanoengineer the “stuff” directly.
First cut illustrated below:
Which tier do you want Aotearoa to play in long term?
The usual diverse collection of signals received from the future this week:
Comprehensive thread on current startup trends from upstart VC fund Untapped Capital’s Yohei Najajima (spotting by Saya🙏):
Coinbase became the first cryptocurrency exchange to list on Nasdaq and reached a market cap of nearly US$76Bn - making it one of the 30 largest financial services firms in the world. On the one hand, a tipping point for crypto industry mainstream legitimacy, on the other hand, a “sellout” for the original trustless model behind Bitcoin and other crypto:
“The group’s transmutation into just another middleman operator has been fascinating to watch. It’s unclear if the platform’s 56m users understand or even care that they are not holding coins but Coinbase IOUs, or that most transactions on the platform are not even settled through any public blockchain.“ - Izabella Kaminska, FT
AI-generated pickup lines (spotting by Kay🙏):
Yet more GPT-3: excellently funny AI researcher Janelle Shane tested out a few variants with the prompt phrase:
“These are the top pickup lines of 2021! Amaze your crush and get results! 1.”
The results are… hilariously bad. So many to choose from, my personal top pick:😅
“I will briefly summarize the plot of Back to the Future II for you.”
How to scan your QR code in Shanghai (spotting by Stu🙏):
Also via Alvin Foo, the incredible video below shows the new Mollii Suit from Swedish startup ExoNeural Network, designed for treating spasticity (muscle stiffness) and other motor disabilities. The suit has 58 electrodes which can be individually programmed for each user, using low level electric current to produce basic tension in the muscle, stimulating the antagonist to the spastic (stiff) muscle. (For example, if the biceps is spastic, the tricep is stimulated which in turn makes the biceps relax). More in this interview with founder Fredrik Lundqvist. Amazing life-changing innovation.
🐕Spot the Robocop:
NYPD are using a specialised version (it looks like) of Boston Dynamics’ Spot robot on Police call outs:
I was pretty impressed by seeing the Wisk Cora in real life at the weekend (see below)… and then I saw this concept video from eVTOL-car startup ASKA… next level again:
🚴On yer e-Vélo:
In an amendment to a draft climate bill passing through parliament, France will offer owners of old cars the opportunity to scrap their vehicles in return for a 2,500 Euro grant to buy an electric bike.
Big Tech’s guide to talking about AI Ethics:
MIT Technology Review’s Karen Hao lists 50-ish words you can use to show that you care without incriminating yourself. Savage and very funny.
(Also, Tired: Ethics of AI // Wired: Ethics of VC).
To which I say:
"Your current version of Ethics is now out of support. Please upgrade to the latest version of Ethics."
🌌The Autodidactic Universe🤯
A team of researchers working alongside Microsoft have published a pre-print research paper which combines theoretical physics, computer science and philosophy of science to describe “an approach to cosmology in which the Universe learns its own physical laws”. Basically the laws of physics, such as we have observed, might be continuously evolving through neural-network-type unsupervised “self-learning”.
“This paper is one of a growing number that attack the question of why these laws [of physics]? …Why the vast hierarchies of scale and why the particular ratios of parameters of the standard model, setting the values of the masses and mixing angles?
It is sobering to contemplate that not one problem of this type has ever been solved, going all the way back to the measurements of the electron’s mass and charge. Roughly speaking, we are faced with a single stark choice:
Either: There are no rational reasons for any of these choices. The universe might have been very different, but there will never be a reason why it took the path we observe it on.
Or: There is at least one rational explanation - in which case we are obligated to find it.”
Get your neurons around that over your Wednesday morning espresso.
Around the traps this week:
🛩️Back down here on Earth - Wisk’s latest Cora eVTOL prototype was on display at Tūranga in Ōtautahi last weekend, a thing of beauty to see up close. Speaking with the Wisk staff, this is aimed at autonomous urban air mobility, room for 2 people, weighs 1600kg without passengers, so far 23 minutes of flying time (no passengers) from its test base near Tekapo.
Wisk has serious backing from both Boeing and Google co-founder Larry Page’s Kitty Hawk and can afford to take its time - the technology has a way to go, not to mention the “Vertiport” infrastructure and air traffic control upgrades needed to support a mass rollout. But in 25 years time… would anyone still own a car?
The government’s Digital Council advisory group published their findings on Trustworthy and trusted automated decision-making (ADM) in Aotearoa.
So cool to see this new initiative from University of Waikato’s new AI Institute: MBIE-funded, NZ$13 million platform TAIAO (“Time-Evolving Data Science / Artificial Intelligence for Advanced Open Environmental Science”) to advance cutting edge environmental data science in Aotearoa:
“It will build a new open source framework to implement machine learning on time series data, provide an open available repository with datasets to improve reproducibility in environmental data science, and build capability in fundamental and applied data science, accessible to all New Zealanders.”
More in this article by Matthew Scott in Newsroom: High-tech tool to help scientists see the wood for the trees.
Eclectica from around the internet this week:
🎧Probably my favourite album of all time, Massive Attack’s Blue Lines was released 30.Years.Ago. Other than making me feel very old indeed👴, I enjoyed RNZ Music101 programme’s highly listenable, memory-jogging 30th anniversary deep dive, which reminded me of just how magnificent and timeless that creative work is.
Interesting thread on Twitter: China’s European ghost towns: the tourism trend that went wrong.
🌎When two worlds collide🌏: this simulation is hypnotic to watch (spotting by Alice🙏)
😶And finally, spotted on TradeMe this week. I dare ANYONE to top this:
As always thanks🙏🙏🙏 for reading and for excellent link spotting by readers this week…more again next time!
Ngā mihi / Cheers
My italics. But, *obviously*.