Memia 2022.11: #standwithukraine🇺🇦// dual use generative AI☣️😬// the proactionary principle⏭️// beaming down to the rectenna☀️🛰️📡// dandel-IoT🎐// decentralising leadership♾️
A DAO can be formed either online or by mailing the paper form
Kia ora
Another week, another Memia newsletter: scanning across emerging tech and the future as it unfolds. As always, thanks for being here!
iPhone users can take a read of this week’s edition in the new Substack iOS App…let me know what you think. (Hopefully it stretches to more than 102Kb, eh…)
Weekly roundup
The most clicked links in last week’s edition (4% of openers each) were the world sustainable electricity map and Australian military commentator Mick Ryan’s Ukraine Day 19 thread.
🇺🇦 #standwithukraine
There’s nothing extra I can say about the horror of watching a once-thriving city with the same population as Ōtautahi be flattened this week. As discussed at the weekend, there is mounting international pressure for a war crimes tribunal once this invasion is over. (More below).
Once again sharing a few recent links on Ukraine which I’ve found helpful to interpret events:
Firstly: if like me you sometimes can’t tear yourself away from social media right now, this is a welcome suggestion from Rob Holmes:
Wikipedia page on Casualties of the Russo-Ukrainian War: likely more people (military and civilians on all sides) killed in the last 4 weeks than in the entire Russia-Ukraine conflict from 2014-2021 (estimated at 14,000 casualties). Just tragic.
The meaning of “railway strike” has changed for 2022:
Why Can’t the West Admit That Ukraine Is Winning? Hawkish analysis by Eliot A. Cohen (US international relations academic and author of The Big Stick: The Limits of Soft Power and the Necessity of Military Force) in The Atlantic:
“Talk of stalemate obscures the dynamic quality of war. The more you succeed, the more likely you are to succeed; the more you fail, the more likely you are to continue to fail. There is no publicly available evidence of the Russians being able to regroup and resupply on a large scale; there is plenty of evidence to the contrary. If the Ukrainians continue to win, we might see more visible collapses of Russian units and perhaps mass surrenders and desertions. Unfortunately, the Russian military will also frantically double down on the one thing it does well—bombarding towns and killing civilians…
…We should begin making arrangements for war-crimes trials, and begin naming defendants, as we should have done during World War II. Above all, we must announce that there will be a Marshall Plan to rebuild the Ukrainian economy, for nothing will boost their confidence like the knowledge that we believe in their victory and intend to help create a future worth having for a people willing to fight so resolutely for its freedom.”
Elsewhere: Indian imports of Russian oil have quadrupled this month as Europe shuns cargoes: Delhi is maintaining close trading ties with Moscow despite western sanctions, although these imports apparently don’t technically break the sanctions. Whither The Quad…?
China also continues to face both ways. This week I was struck by the over-simplistic, zero-sum outlook of one Chinese commentator, likely not intended for Western consumption:
(translation courtesy of Australia-based “political junkie” Han Yang - click tweet below for full thread):
“Russia’s war takes it to the forefront of competition with the US, temporarily replacing China, giving us breathing space to regroup after Trump’s brutal trade war. Two powers resisting US hegemony is better than one…
…A strategic alliance btw Russia and China is mutually beneficial and reinforcing, providing deterrence against Japan and India…
…Some say why not take the chance to repair relations with US? Its goal for Russia is to disarm, but its goal for China is to completely stop our development and prosperity, make us a client state like Japan and Korea, buy its weapons, and remain only a cheap source of product…
…China will avoid conflict with US not proactively challenge US, but will use our power to force US to accept peaceful coexistence. Russia is the most important partner to achieve this goal. That’s why we can’t abandon Putin during this war. This is in our long term interest”
… This “long term interest” narrative seems to completely discount the interdependent, cross-border, multipolar complexity of the modern global economy — of which China and Russia are still part. Tied together by a mostly uncensored internet, mostly agreed trade rules and mostly shared basic values (eg *don’t* bomb a city full of innocent civilians), the world is not just “US”, “China” and “Russia” any more, if it ever was…the analysis seems way too simplistic and primitive.
Reflection:
China and India, two giant nation-state actors taking ambiguous positions on Ukraine “in their own national interests” right now.
From where I sit, most citizens of the world don’t want unconstrained, threatening superpower actors of any hue any more - I hope for a future global, cross-border, “consensus-values”-based, robust governance fabric able to *coax* aggressive nation-state actors back into alignment with a more decentralised (by which I don’t mean US-led) international consensus - and do away with zero-sum rhetoric once and for all.
It would be a long journey to establish a new world order to achieve that… but perhaps recent events are precipitating an acceleration.
(See also below: decentralising leadership).
♨️Hot cold
Not just the Arctic as discussed on Sunday - Antarctica is experiencing a “heatwave” simultaneously as well. What to do…😟
☣️😬⏭️Dual use generative AI / the proactionary principle
In light of the lingering threat of chemical weapons usage in Ukraine, a new paper outlining how it took less than six hours for a drug-developing generative AI algorithm to invent 40,000 potentially lethal molecules is even more worrying than it would be normally.
In one of the most scary dual use technology scenarios to date, computational drug discovery researchers configured a generative AI model normally used to search for helpful drugs into a “bad actor” mode. Some of the molecules developed were modelled to be more potent than the deadly VX nerve agent:
“Our proof of concept …highlights how a nonhuman autonomous creator of a deadly chemical weapon is entirely feasible.
…The thought had never previously struck us. We were vaguely aware of security concerns around work with pathogens or toxic chemicals, but that did not relate to us; we primarily operate in a virtual setting. Our work is rooted in building machine learning models for therapeutic and toxic targets to better assist in the design of new molecules for drug discovery. We have spent decades using computers and AI to improve human health—not to degrade it. We were naïve in thinking about the potential misuse of our trade…Without being overly alarmist, this should serve as a wake-up call for our colleagues in the ‘AI in drug discovery’ community.”
In an interview with The Verge, co-author Fabio Urbina expands:
“I don’t want to sound very sensationalist about this, but it is fairly easy for someone to replicate what we did. If you were to Google generative models, you could find a number of put-together one-liner generative models that people have released for free. And then, if you were to search for toxicity datasets, there’s a large number of open-source tox datasets. So if you just combine those two things, and then you know how to code and build machine learning models — all that requires really is an internet connection and a computer — then, you could easily replicate what we did. And not just for VX, but for pretty much whatever other open-source toxicity datasets exist.”
(They finish the paper with some robust recommendations to reduce potential dual-use risks for AI in drug discovery, including):
“…ethical training of science students and broaden the scope to other disciplines, and particularly to computing students, so that they are aware of the potential for misuse of AI from an early stage of their career”.
Reflections:
Another item to add to the lengthening list of existential risks for humanity.
But to be more sanguine for a moment, fundamentally just the evolutionary algorithm continuing to accelerate: each new set of AI-generated toxins would likely spur development towards a corresponding catalogue of AI-generated antidotes or other defences, possibly within months, weeks or days… (similar to how the Covid vaccines were developed in record time).
Also likely another driver to accelerate the nascent field of Transhumanist nanotechnology: at some point in the not-so distant future, expect to have nanobots floating around in our bloodstreams ready to defend against a whole pharmacy of synthetic toxins.
For a more challenging moral stance, pass this scenario across the radical arguments for the Proactionary Principle, advanced by transhumanist thinkers Max More, Steve Fuller and others. (The Proactionary Principle is the exact opposite of the Precautionary Principle for innovating with new technologies):
“People’s freedom to innovate technologically is highly valuable, even critical, to humanity…The proactionary principle argues that "sufficient study" [of the impact of new technologies] may in some cases be impractical. For instance, in releasing a new life form into the biosphere — whether genetically modified plant, animal, or bacteria — one would have to simulate the biosphere to achieve "acceptable levels of predictability". While the innovator of the new life form might point out that such a simulation would be a heavy burden, the other life forms in the biosphere could suffer irreparable harm in the case of an untested release. More's first principle, freedom to innovate, would place the burden of proof on those who propose a restrictive measure”
(I think this was written before the potential of Generative AI was completely understood!) Personally I fall down on the side of precaution on this one…
[Weak] signals
A roundup of recent tech developments signalling from one or more possible futures…
Three more renewable energy stories to start.
☀️⚡🔋27 times as much electricity and make it renewable
Provocative article by Australian academics Paul Burke, Emma Aisbett and Ken Baldwin in The Conversation: to really address climate change, Australia could make 27 times as much electricity and make it renewable.
Their newly-published study (carried out as part of a multi-disciplinary team under the Zero-Carbon Energy for the Asia-Pacific project) analyses the potential for Australia to produce and export not only clean energy but also “green” value-added commodities, eliminating emissions that would have taken place elsewhere.
They calculate that about 2% of Australia’s land mass (168,000km2) would be needed for solar and wind farms - about 7,000 terawatt-hours of solar and wind generation per annum – 27 times Australia’s current electricity output and use, which surplus can then be exported either as electricty, or embodied in “green” hydrogen, steel and aluminium.
☀️🛰️📡Beaming down to the rectenna
The UK government is reportedly considering a £16 billion proposal to build a massive solar station in space — an big satellite equipped with solar panels which would beam energy back down to Earth, helping the UK achieve net-zero by 2050.
The satellite would be 1.7km in diameter, weighing around 2,000 tonnes and operate in geosynchronous orbit. The terrestrial “rectenna” would be roughly 6.7km by 13km, so likely to be placed offshore. The satellite would deliver 2GW of power to the UK.
The diagram below shows the proposed CASSIOPeiA design concept by UK inventor Ian Cash.
🧮Ledger of Power
Crypto’s massive carbon footprint dominates any discussion about blockchain’s energy efficiency - but there are several real-world blockchain projects underway which aim to increase the efficiency of renewables, mainly by transparently managing supply and demand. CoinTelegraph magazine profiles the blockchain projects making renewable energy a reality:
“It‘s a whole new era where you can take the sun‘s rays and turn it into a currency.”
🤔Hmmm…
🧾Tax those DAOs
Across the ditch, Australian Senator Andrew Bragg gave a speech to the Australia Blockchain Week conference on Monday, stating that decentralised autonomous organizations (DAOs) pose an “existential threat to the tax base”. He announced new crypto legislation called the Digital Services Act (DSA): saying that DAOs “must be recognized and regulated as a matter of urgency”:
“Recognising the fact that DAOs are self-regulating and transparent, with an in-built system for governance. Nonetheless, the legal recognition of DAOs means we should legislate minimum standards:
• A responsible person within the jurisdiction.
• A consumer protection framework distinguishing between wholesale and retail DAOs.
• Standards for audit, assurance, and product disclosure as appropriate.
• Limited liability
• Replaceable rules to standardise DAO governance protocols.
The Treasury will need to address these issues, leaving the field open for DAOs to continue to live up to their name. Any attempt to prescribe a code be[sic] self-defeating.”
Well, indeed.
Several US states including Wyoming have advanced DAO legislation already. (From Wyoming’s DAO FAQs (and I had to check this wasn’t a spoof…)
“A DAO can be formed either online or by mailing the paper form.“
…Government in the 21st century, eh.
🎐Dandel-IoT
Many plants - for example the Dandelion - have evolved mechanisms to disperse their seeds using the wind.
A paper in Nature this week extends this concept to IoT - Wind dispersal of battery-free wireless devices demonstrates millimetre-scale devices weighing only 30 milligrams using off-the-shelf parts and powered by lightweight solar cells for various sensing and computing applications. (h/t spotting Dad!)
🥽Augment the garage
Veteran AR / VR strategist Robert Scoble has some sensible strategy advice for budding AR startups:
“Entrepreneurs: Avoid the elephants in augmented reality….
[There are] a ton of ideas but many are obviously going to be built by Apple, Nvidia, Unity, Amazon, Meta, Snap, etc. The elephants.
Things like concerts in your family room.
What about augmenting your garage? I haven't heard ANYONE talk about that yet. At big companies or small. So, someone could become "the AR garage" company/entrepreneur.
The real lesson here is don't start a company that will compete head on with Apple or Unity. Build companies that will take over some other part of the home than the family room, or home office, where tens of billions are being invested.
Could you build a company that augments the garage?…Another idea? Augment the microwave oven. Become the company that does that and adds a lot of fun and utility onto just the microwave that many people have in their homes.”
Blue Ocean Strategy, 2022 edition.
🍗Birdbot
Turns out ostriches run more efficiently than people and this robot proves it:
📋Replicating RNA
Finally, in a new insight Into the possible origins of life, for the first time researchers have created an RNA molecule that replicates.
Mind expanding
Some longer links to stimulate those neurons this week!
♾️Decentralising leadership
I featured Time Magazine’s insightful interview with Ethereum co-founder Vitalik Buterin on Sunday - if you haven’t read it, recommended, he is one of the world’s most highly tuned minds.1 A short extract:
…Buterin worries about the dangers to overeager investors, the soaring transaction fees, and the shameless displays of wealth that have come to dominate public perception of crypto. “The peril is you have these US$3 million [Bored Ape] monkeys and it becomes a different kind of gambling”…
…Buterin hopes Ethereum will become the launchpad for all sorts of sociopolitical experimentation: fairer voting systems, urban planning, universal basic income, public-works projects. Above all, he wants the platform to be a counterweight to authoritarian governments and to upend Silicon Valley’s stranglehold over our digital lives.“
Australian futurist Ross Dawson observes how the very concept of “leadership” is shifting away from an implicit social hierarchy/centralised command and control model towards a more decentralised, “soft power” interpretation, epitomised by Vitalik and his “leadership” (actually I hate the word) of the Ethereum Foundation.
(But hold that thought when thinking about how the world immunises against aggressive superpower nation states in future…)
🌐21st century battlefield
The first great information war: Ukraine & civic resistance on the 21st century battlefield by Jennifer Cassidy
📊🧍Databodies
Bodies of data or “databodies”? Julia Keseru continues her series exploring the connection between bodily integrity and our online reality:
“As of next summer, tax filers in the United States will only be able to access certain online services if they go through an identity verification process that uses facial recognition software…
…predatory technologies became our digital default because of the naive assumption that the Internet would keep our bodies invisible. As a result we now have a largely unregulated technology industry, with data protection regimes that cannot guarantee the integrity of our physical bodies.”
Thought provoking.
Gems
A great catch of memes in the net this week:
🏓Arduino ping pong
This Arduino Arduino hobby project is a PID-controlled “ball on a plate” machine - with the sole goal of keeping a ping pong ball bouncing for as long as possible. The location of the ball is calculated from the time difference when the ball’s noise is hitting the 4 microphones. (Spotted via the eclectic @massimo.)
🧑💻🧑💻🧑💻🧑💻🧑💻🧑💻🧑💻🧑💻🧑💻How many software engineers does it take to…
Remember Windows Vista?
From 2006: former Microsoft developer Moishe Lettvin recounts how a team of at least 24 were responsible for input into the design of *that* bemusingly complex shutdown menu.
(Spotted via Dan Hon’s daily blog Things That Caught My Attention - in which he carries out a retrospective over 15 years later.)
👷Construction automation
All bricks should be laid like this from now on.
💯Meme appeal
Exquisite.
Thanks as always for letting me into your inbox every week, and especially everyone who reaches out with feedback and links - appreciated!🙏🙏🙏
Catch you again on Sunday.
#standwithukraine🇺🇦
Ben
Dipping into Vitalik’s blog is a regular intellectual treat.
Create your profile
Only paid subscribers can comment on this post
Check your email
For your security, we need to re-authenticate you.
Click the link we sent to , or click here to sign in.