AI is no match for the quirks of human intelligence ⊗ The future of AI is a conversation with a computer ⊗ The industrial re-revolution

This week →{.caps} AI is no match for the quirks of human intelligence ⊗ The future of AI is a conversation with a computer ⊗ The industrial re-revolution ⊗ Hey, Facebook, I made a Metaverse 27 years ago ⊗ Network intersubjectives

A year ago →{.caps} The most popular article in issue No.149 was How Discord (somewhat accidentally) invented the future of the internet by David Pierce.

◼{.acenter}

Something to note if / when you think “oh no, AI again” or “oh no, not the **verse again!” A lot of the articles I include are there for their central topic, yes, but also often for a slightly different perspective or because they are poking at a hard problem in different ways, surfacing intriguing twists in perspective or knotty challenges and potential solutions. The top four today do some variation of that, for example the first piece below starts from AI but explains some quirks of our own intelligence, which are interesting in their own right, but also because of how they constitute challenges for an hypothetical Artificial General Intelligence. Nothing super new on the AI side, but an excellent read for what it assembles.

AI is no match for the quirks of human intelligence

As mentioned above, Herbert L. Roitblat reviews some of our brains’ quirks and how AI does not address them, perhaps never will completely. It’s a much more fleshed our and relevant twist on the ‘computer can’t be creative’ argument we often see, plus it can even be read just out of curiosity about the brain. To circle the article back to creativity, the author’s description of insights and of Kahneman’s “slow, deliberate, and more accurate system” lines up very well with the design squiggle, variations of which happen in writing and various creative endeavours.

[I]nsight problems are characterized by a kind of restructuring of the solver’s approach to the problem. In path problems, the solver is given a representation, which includes a starting state, a goal state, and a set of tools or operators that can be applied to move through the representation. In insight problems, the solver is given none of these.{.highlight} […]

With insight problems, it is often difficult to determine whether any progress at all has been made until the problem is essentially solved. They’re often associated with the “Eureka effect,” or “Aha! moment,” a sudden realization of a previously incomprehensible solution. […]

The main barrier to solving problems like this is to abandon the default representation and adopt a more productive one. Once the alternative representation is identified, the rest of the problem-solving process may be very rapid.{.highlight} […]

Analytic problem solving is directly applicable to systems that gain their capabilities through optimization of a set of parameters. On the other hand, if the problem requires divergent thinking, commonsense knowledge, or creativity, then computers will continue to lag behind humans for some time.

The future of AI is a conversation with a computer

Again, much of the AI talk in here we’ve already seen, but worth a read for the useful parallel with geofoam and for a few other details. First, it’s another example that AIs just make shit up, it doesn’t differentiate between inventing a turn of phrase and inventing facts, it can spew out text that looks like a good approximation of an article, but the content might be false. Kind of an important detail.

Second, much of the improvements in recent years are due to the scaling of the data that goes into them, what happens when ‘everything’ as been used for a model and it’s still not good enough? Google has a 1.6-trillion-parameter model, at some point you have to think it’s getting redundant.

Third, even when people write critically about AIs, they still use words and phrasings like “thinking,” “didn’t understand,” “dialog.” Hard to debunk the premise of AIs having actual intelligence when you assign them human traits. Also, noting “low attention text” for later.

A reliance on scale, though, is inextricably linked to the statistical approach that creates uncertainty in these models’ output. These systems have no centralized store of accepted “truths”; no embodied understanding of “what the world is like for humans” and, hence, no way to distinguish fact from fiction or to exercise common sense.{.highlight} […]

This, I think, is why AI writing is so much more exciting than many other applications of artificial intelligence: because it offers the chance for communication and collaboration.

The industrial re-revolution

According to Ed Conway, economists are obsessed with the industrial revolution and usually focus on their own pet event to explain how it unfolded. Conway shows that of course what followed was not made possible by one thing, it was a long sequence of inventions building one on top of the other; the Leblanc process, Portland cement, Aluminium, nitrogen fixation, etc. Why is that important? Because carbon dioxide is created in pretty much all of these processes, so when we talk about decarbonating the global economy, it’s not just cars and power stations, it’s re-inventing all of these technologies. It’s recreating the industrial revolution.

There is carbon dioxide created in the manufacture of Portland cement, of steel, of glass, in the making of most chemicals, in the production of nitrogen-based fertilisers, in the electrolytic reaction at the heart of the Hall–Héroult process that creates aluminium, in the refining of most metals, in the minting of silicon chips and solar panels and the making of lithium ion batteries. […]

…literally some of the most incredible technological leaps the human race has ever achieved. And we literally need to do it all over again. Not just that: unlike in the Industrial Revolution which was really spread out over centuries, we are aiming to do this in a matter of decades.

Hey, Facebook, I made a Metaverse 27 years ago

Not exactly, but Ethan Zuckerman did create a 3D MOO back when hand-coded HTML pages were still quite new. He shows that people have been metaversing for a while now, much has been widely forgotten, and Zuck’s version is not only boring, it’s not that good, and will of course end up as a walled garden. Many have said the same already, yet a fun read for the quick history lesson, and the ±three decades of experience behind the argument.

Stephenson, of course, wasn’t being entirely original either. His vision of the metaverse owed a debt to Vernor Vinge’s 1981 True Names and to a series of William Gibson novels from the ’80s. Both of those authors owed a debt to Morton Heilig’s 1962 Sensorama machine, and on and on we go, back in time to Plato’s shadows on a cave wall. […]

Facebook can claim originality in at least one thing. Its combination of scale and irresponsibility has unleashed a set of diverse and fascinating sociopolitical challenges that it will take lawmakers, scholars, and activists at least a generation to fix. […]

The metaverse isn’t about building perfect virtual escape hatches—it’s about holding a mirror to our own broken, shared world. Facebook’s promised metaverse is about distracting us from the world it’s helped break.{.highlight}

Network intersubjectives

Gordon Brander comparing the up and downsides of three network topologies (centralized, distributed, and decentralized), and the spectrum of variations between them. Also shared for the “Ostrom commons,” and for the piece in the context of Brander’s research for Subconscious, a tool for imagination.

Federations are pluralistic by nature. Different groups can have different governance models, tailored to the goals of the community. This grants federation more variety than centralized one-size-fits-all networks. It also grants federation more variety than p2p networks, since a protocol can never have requisite variety to encompass the complexity of human nature. […]

[I]t may be more helpful to view network topologies as a design space, where different models, and mixed models can be designed to produce particular outcomes.

Asides

• 😨 🤬 🎥 Scientists get it, lots of citizens get it, The Economist (!) gets it. ‘Leaders?’ Doesn’t really look like it. See what three degrees of global warming looks like. “If global temperatures rise three degrees Celsius above pre-industrial levels, the results would be catastrophic. It’s an entirely plausible scenario, and this film shows you what it would look like.”
• 🇧🇹 🇸🇷 🇵🇦 🌳 👏🏼 Forget net-zero: meet the small-nation, carbon-negative club. “A tiny but growing club of “carbon negative” forest countries is emerging, with Suriname - a small rainforest country north of Brazil - already a member and Panama expected to be certified later this year. What they have in common is strong protection of their carbon-absorbing forests alongside increasingly tough measures to hold down climate-changing emissions, including efforts to adopt renewable energy, electrify transport and cut waste.”
• 🐋 💩 🌳 Whale-based geoengineering! I’m in! Poo from the world’s largest animals have a stunning affect on ocean ecosystems—and even carbon capture. “A million additional whales defecating close to the surface would be like having massive ocean fertilizer machines—absorbing as much carbon as forests covering a continent”
• 🦤 🤯 Endangered birds experience 'virgin birth,' a first for the species. “Though rare in vertebrates, parthenogenesis occurs in sharks, rays, and lizards. Scientists have also recorded self-fertilization in some captive bird species, such as turkeys, chickens, and Chinese painted quail, usually only when females are housed without access to a male. But this is the first time it’s been recorded in California condors.”
• 🪵 🏠 🇳🇱 Twenty per cent of new homes in Amsterdam to be constructed from timber. “The city of Amsterdam has mandated that 20 per cent of all new housing projects in the Dutch capital must be constructed with wood or other biobased materials from 2025.”
• 🇫🇮 🇪🇪 🤔 🤔 “We’re building tunnels not walls” says Angry Birds co-founder behind plan to link Helsinki and Tallinn via undersea tunnel. “All the station areas are being designed to accommodate 50,000 new people moving to the area so we’re adding 200,000 people in the Helsinki/Tallinn metropolitan area, which currently has a population of two million people.”
• 👾 🤑 🇫🇷 Ubisoft is Making 'Play to Earn' Blockchain Games. “That both Ubisoft’s CEO and CFO both made clear the company’s future would involve the blockchain is certain to be controversial. NFTs, cryptocurrency, and the blockchain are widely associated with scams, environmental damage, and manipulative marketing. While Ubisoft said it was still learning, its heavy investment in Animoca shows the company is well past the point of deciding whether or not to support blockchain endeavors.”

{.miscellany}