Dark Oxygen/ Sparse Convolutional Neural Network/ Blue Origin’s Engine Explosion/ A Comparative Perspective on AI Regulation
Massimo Portincaso
Founder & CEO at Arsenale Bioyards, Industrial Romantic and Antidisciplinarian Stoic
It has been a super-hot and dense week. Hence only a very quick thought about the more and more recurring heat waves, as the week didn’t leave time for much more…
I had already started to think about it last year, with the previous heat waves, but is now more evident than ever. I cannot stop thinking that it is going to get worst and worst, faster and faster, as the reinforcing loops starts kicking in. And then the question arises: is it already too late? Have we already “fucked” the planet (pardon my language)? I am not a scientist and there are people way more qualified than me to answer that question.
But the one thing I know, is that the moment we reach that break-away point, being it a complex adaptive system with strong reinforcing loops, then it is going to be fast, very fast and we will not be able to intervene with the instruments that we have been using till that point in time. ?
Hopefully we manage to get things back on track, but every day that passes I am less confident we will be able to revert things and stop climate change… If till a few years ago I was convinced that we can, and that we will reverse, or at least stop climate change, I see now our chances to succeed as 50% at max, that is my sad assessment of the reality. And I also started thinking about scenarios where climate change becomes non reversible… I am not a doomer, quite the opposite, but I also like at reality, and this is what reality is telling us…
?
New research shows that there’s “a vast biosphere with a global volume nearly twice that of all the world’s oceans” right beneath our feet, where “abundant microbes produce unexpectedly large amounts of ‘dark oxygen’ even in the absence of light.”
The study upends much of science’s previous assumptions of “subterranean realms [as] oxygen-deficient dead zones. Karen Lloyd, Subsurface Microbiologist at U.T., compared the discovery of dark oxygen to “the scale of oxygen coming from the photosynthesis in the Amazon rainforest.”
News items:
Baldness is typically considered “a cosmetic concern, not a medical one,” however, hair loss “takes a huge toll on mental health,” and studies show that cancer patients even “consider refusing chemotherapy” rather than lose their hair. A new treatment, using a “protein called osteopontin,” discovered by studying “hairy moles,” may offer new hope to the hairless and “raises some intriguing questions about aging cells.”
For experimental physicists, looking for “useful information” in “particle experiments” is like looking for a needle in a haystack. “Detectors capture and analyze vast amounts of data… [but] often only a small portion of it actually matters. A new ML tool called a “sparse convolutional neural network (SCNN)” could filter out the garbage data and “vastly accelerate [researchers] ability to do real-time data analysis.”
“It’s increasingly likely that SCNNs — an idea originally conceived in the computer science world —will soon play a role in the biggest experiments ever conducted in neutrino physics (DUNE), neutrino astronomy (IceCube), and high-energy physics (the LHC).”
News items:
Robo-trading is “playing a growing role” in European energy markets, particularly as “the continent ramps up its reliance on intermittent renewable energy.” Events like “clouds settling over solar farms” can lead to wild price swings. “That type of volatility and the massive amount of data needed to track supply and demand opens up lucrative opportunities for traders savvy enough to harness satellite images, weather patterns, and even social media posts to get ahead of price changes.”
In June, one of Blue Origin’s BE-4 rocket engines exploded “about 10 seconds into the test.” BE-4 is already “years behind schedule,” and this setback could jeopardize its attempts to have its Vulcan rocket approved by the US Space Force to fly valuable national security missions.”
According to “industry expert” Ben Brockert, there are three essential stages to rocket engine development: Development, qualification, and acceptance. BE-4 is supposed to be in stage three by now, but now its immediate future is in doubt. Blue Origin’s numerous setbacks have “many in the industry, both competitors and customers, fear[ing] a SpaceX launch market monopoly.”
领英推荐
News items:
NASA’s Perseverance rover has found “diverse types of organic molecules” on Mars. The “Martian organic matter” doesn’t prove there’s life on Mars, but “it could give us important clues about whether Mars was able to host alien life.”
The results are in! A new study shows that GenAI tools like ChatGPT can help level the playing field between good writers and bad writers in the workplace. “453 marketers, data analysts, and college-educated professionals” were recruited to write “press releases, short reports, or analysis plans.” Half of the participants were given the opportunity to use ChatGPT as a writing assistant.
Those that did “took 40% less time to complete their tasks, and produced work that the assessors scored 18% higher in quality than… participants who didn’t use it.” Proficient writers took less time to complete their assignments, and “those who were assessed as being weaker writers produced higher-quality work once they gained access to the chatbot.”
News items:
Amidst all the excitement about GenAI image platforms like Midjourney and Stable Diffusion, a burgeoning cottage industry emerged amongst “creators” generating trend-driven packages of clip art to sell on Etsy. “Thought leaders” like Patryk Marketer claimed to have a “surefire system to generate a passive income stream from AI-generated art.” A few months in, the picture Marketer paints is not so rosy. He says now, “Once you get into it, you find out that it is not as easy as people say. It’s challenging to make enough money to sustain your lifestyle. The thing I have learned—it was a hard lesson—[is that] trends on Etsy change so fast, super-fast.”
?
An overview of the different approaches being taken by governments in the EU, UK, and US to regulate AI. The EU aims to be “a world leader” in regulating AI, and its “sweeping” AI Act proposal could be passed by the end of 2023 but is “unlikely” to go into effect until mid-2025. The EU’s leadership on “personal data protection” with GDPR gives them recent experience with regulating tech.
UK PM Rishi Sunak and his Conservative party have adopted a pro-innovation approach to AI regulation and pledged to “avoid heavy-handed legislation.” Sunak reportedly reached out to President Biden to propose a “global AI watchdog authority” inspired by the IAEA.
The US federal government’s approach to regulating AI has thus far been “piecemeal.” October 2022 saw the release of the Blueprint for an AI Bill of Rights lays out “five principles… intended to guide the responsible use of AI systems.” In Congress, “there is an active appetite… to oversee and potentially regulate AI,” but thus far, that’s mostly led to more hearings than concrete action, despite the recent announcement of the SAFE Innovation Framework. State-level regulation is patchwork at best, with California’s Assembly Bill 311 the only “comprehensive AI framework” currently being considered.
There is still much work to be done with AI regulation. “Whether there will eventually be a true global standard for AI regulation remains to be seen.”
News items:
British tabloid The Daily Mail is suing Google for using “hundreds of thousands of online news stories” to train Bard — its ChatGPT competitor. Google allegedly “harvested” around 1M articles from the Daily Mail and CNN “without either copyright holder’s knowledge or permission.”
Owen Meredith, chief executive of industry group The News Media Association, said, “If allowed to continue unchecked, [GenAI] has the potential to become a serious threat to publishers’ business models, which would undermine the future sustainability of journalism, inhibiting public access to trusted information. If quality journalism cannot be funded, we all lose out.”
Researcher | Consulting Advisor | Keynote | Chief Innovation / Learning Officer. AI to Transform People's Work and Products/Services through Skills, Knowledge, Collaboration Systems. AI Augmented Collective Intelligence.
1 年the AI writer link points at the wrong article (though it was interesting too!)