The Realverse/ Green Future Index 2023 Sees Progress/ The “Manhattan Project” Theory of Generative AI/ America Is Back In The Factory Business
Massimo Portincaso
Founder & CEO at Arsenale Bioyards, Industrial Romantic and Antidisciplinarian Stoic
The Realverse. A while ago I received, as a present, one of those disposable cameras with a black and white film in it. It had been sitting on my desk for a long while, till, last summer, I decided the time had come for me to use it.?
It took us a while to finish the film, as, we wanted to have the right occasion to use it, particularly because it was only 27 pictures we had available, and we wanted to make the most out of them. Eventually, we managed to get the film full, and we had it developed this week. Quite frankly, I didn’t expect the whole experience to lead me to some deep thinking, but it did. And some of the articles below around AI-generated photos and IP made the thinking even sharper...
Of course, it was a lot of fun to explain to my 17-year-old son that this was the way I used to take pictures before mobile phones and digital photography took off. His expression while explaining to him that he had to wind the film after each shot was simply hilarious. Similarly, explaining that you took the pictures, then sent them to develop and then you would receive them back, sometime weeks after the picture had been taken, was an interesting experience, together with him opening the envelope with the printed pictures. But this is not the deep thinking I was referring to.
The moment that got me into thinking was when I looked at the pictures for the first time. It was a very strange feeling. The pictures were on average not so good, I immediately realized the importance of the algorithms that in my phone constantly compensate for the changing environment, the poor light, and challenging conditions, to allow me to take (almost) always good pictures. So, a part of me immediately thought about the impact of technology on our lives, and how it makes things smoother and/or easier. And I found particularly telling the fact that it was playing such an important role by “smoothing” our memories via our pictures. Very often going beyond that, with the use of filters. So, the first reaction was “You know what, by all of this nostalgia for the good old days, technology makes for a better world, I am glad I don’t use films to take pictures anymore…”
But then, another reaction kicked in. Having these imperfect pictures (as the one below) as the official recollection of our experience, made the experience somehow more real, in the sense that the pictures were asking us to complete the recollection with our own feelings and our own memories, and not only this felt good, but also brought a sort of wabi-sabi beauty to it. And what it clearly did, it anchored the pictures in our perception of reality at that point in time.
Now, fast forward from the disposable camera with a black and white film to Midjourney and the “Pope in a puffer coat” (see article below), and you quickly realize that the whole notion of photography and reality gets a very different twist.?
All of a sudden, the perfection of the images creates itself a new reality, that does not exist. And this applies not only to reality but also to our memories. Look for instance at the picture below, you could think that this is from somebody’s family album, while in reality it has been produced using Midjourney. That woman and that situation never existed. Yet, she is real, somehow, in our perception.
The (inadvertently) awarding of the SONY World Photography Award 2023 to a “photo” created with generative AI?is further proof of this complete blurring of boundaries between what is real and what is not real, but still perceived as such.?Luckily, the artist that submitted the picture (Boris Eldagsen) refused the prize with these words:
“Thank you for selecting my image and making this a historic moment, as it is the first AI-generated image to win in a prestigious international PHOTOGRAPHY competition.
How many of you knew or suspected that it was AI generated? Something about this doesn’t feel right, does it?
AI images and photography should not compete with each other in an award like this. They are different entities. AI is not photography. Therefore I will not accept the award.“
He then added:?“We, the photo world, need an open discussion. A discussion about what we want to consider photography and what not.“?Well, I think that not only the photo world needs an open discussion. We all need an open discussion on what is real.
What is happening is quite deep. And while the hype around the metaverse is eventually fading, I think we are ignoring a way more profound development, which is the building of what I would call the “Realverse”, a world that is cognitively real, but that does not exist. Generative AI is only one part of it, as the development in neuroscience will be an even more important component of it.
A lot has been said about the need to explore the ethical dimension of AI and Generative AI. I think it is time to start looking also at the Ontology of Generative AI and Neuroscience, in the “Realverse”, and beyond.?
MIT Tech Review?recently released?its third annual?Green Future Index?and finds “progress… towards a sustainable low-carbon future… wavering as atmospheric carbon levels soar.” As seen in the map above, the Index ranks countries based on “the extent to which [they] are moving toward a green future by reducing carbon emissions, developing clean energy, innovating in green sectors, and preserving the environment, [and] the degree to which governments are implementing effective climate policies.
“Iceland remains top-ranked,” while the US jumped only one spot up to 19th from 20th in 2022. “Emerging economies continue to fare poorly… wealth contributes significantly to a country's ability to define its low-carbon future.” Overall, “the rankings reveal more consistency than progress.” The?IPCC?estimates?“that at current global economic activity rates, it will be less than a decade before the world exceeds the 1.5 degrees Celsius (1.5 °C) global warming threshold set by the?Paris Accord.”
News items:
Deep ocean mining has the potential to deliver an abundant supply of “the metals we need to build batteries” and other crucial tech for tackling climate change. “Potato-sized lumps called?polymetallic nodules… contain manganese, cobalt, copper, and nickel… all used lithium-ion batteries.” Not surprisingly, there are significant concerns about the environmental impact of “deep sea commercial mining.” International waters are governed by the UN’s?International Seabed Authority (ISA), which is now under pressure to come up with rules for deep-sea mining quickly.?
GenAI superfans?and?doomsayers?alike are apt to compare LLMs and the prospects of AGI to the Manhattan Project and the atomic bomb. OpenAi’s Sam Altman agrees that “AI?could?destroy civilization” but claims that?won’t happen?on his watch - and regardless,?AI progress can’t be stopped. The Manhattan Project analogy is inaccurate in many ways, but in “one respect,” GenAi and the atomic bomb are alike. “There is a world before mass access to GenAI and a world after it, and they are not the same.”
领英推荐
News items:
A?new study?indicates that startup acquisitions and hiring of inventors by Big Tech incumbents are “actually hurting” US innovation. An inventor hired by an incumbent earns 12.6% more than at a “young firm,” but their “innovative output declines by 6-11%.” Study co-author?Ufuk Akcigit, Professor at UChicago, says, “The idea that if you allocate more resources to R&D, you’ll grow faster,” he adds, “that’s now broken in the US.”
The business of building factories in the US is booming. Manufacturing-related construction spending reached $108B in 2022.?Government incentives?for building chip fabs and EV battery factories account for a significant chunk of that growth, but even low-cost consumer product businesses in search of faster, more resilient, and flexible supply chains are finding reasons to “come home” after years of manufacturing offshore.
From Paul Revere to Henry Ford, “manufacturing has always been an integral part of American life.” But “construction spending related to manufacturing” hit a 21st century low of?$21.5B in 2003?- two years after?China joined?the?WTO?- a fifth of 2022 spending. This WSJ overview of new factories and the companies building them - both high and low-tech - paints a fascinating picture of how America got “back in the factory business.”
News items:
Apple’s new “interactive goggles,” set to be introduced this June, is being treated with “skepticism [by] current and former employees.” Interviewed anonymously, insiders spoke of “concerns about the device’s $3K price point, doubts about its utility, and worries about its unproven market.”
In late 2022, the US imposed?export restrictions?on selling advanced chips and chip manufacturing equipment to China. In January,?Japan and the Netherlands?also agreed to “bar some shipments of their most high-tech machinery to China.” The?"exact contours of the deal"?remain unknown, but without Dutch and Japanese cooperation, their “assistance could have had China back up and running in as little as a year or two [instead of] decades.” The US, Netherlands, and Japan “collectively control 90%?of the global market for?semiconductor manufacturing equipment, and the restrictions could ensure that China’s domestic chip production falls behind as the global industry advances.”
Or maybe not. According to Bloomberg: “Chinese companies are rising up the list of the world’s biggest spenders on research and development — a sign that perhaps they?won’t need?that Western technology much longer.”?Five years ago, Huawei was the only Chinese company on a global top 25 list of top R&D spenders. Now it’s “joined by TikTok owner ByteDance, Tencent Holdings, and Alibaba.” China’s biggest tech companies - and its government - appear determined not to let export controls slow them down. And they’re willing to pay for it.
News items:
“Pope in a puffer coat” may be one of the most widely seen GenAI-created images yet.?Made using Midjourney?by Pablo Xavier, a 31-year-old Chicago construction worker quickly went viral, and it’s “an omen that marked the accelerating collapse of a clearly distinguishable boundary between imagination and reality.” Is that a good thing? Can GenAI “extend your mind” and help you “see different possible futures?”
As anyone paying attention knows by now, GenAI text-to-image tools like Stable Diffusion, Midjourney, and DALLE-2 were trained on massive datasets of pictures, many of them subject to copyright. LLMs like GPT are similar, except they’re trained on text - much of which also is IP-protected. The ongoing?class action lawsuit?filed by three artists against several text-to-image generators is now one of many, including a?suit?by?Getty?against Stable Diffusion “alleging the improper use of its photos.” The law surrounding GenAI companies’ fair (or unfair) use of copyrighted materials is unlikely to be settled anytime soon. HBR advises developers, creators, and businesses considering or already using GenAI tools- to proceed with caution and “mitigate risk.” Significant hazards exist for all - but there are also “dramatic opportunities” for each.
News items:
“It’s clear the US leads in AI, with advantages in computing hardware and human talent that other countries cannot match.” However, China is quickly?catching up. Government intervention is needed to ensure the US ultimately “wins the AI competition.” Last year’s?ban?on exporting “advanced semiconductor manufacturing equipment and AI chips made with U.S. technology… could prove devastating” to China’s AI ambitions. Meanwhile, immigration policies that cap the hiring of “Chinese [AI] talent” by US companies could inadvertently reverse China’s brain drain and allow it to retain more homegrown AI researchers and innovators.