Source : ai-extended moral agency? by pii telakivi, tomi kokkonen, raul hakli & pekka mäkelä
Category: robotics/AI – Page 91
Two AIs Discuss: They’re Building Brains In a Jar! Organoid Intelligence (OI)
“Disembodied Brains: Understanding our Intuitions on Human-Animal Neuro-Chimeras and Human Brain Organoids” by John H. Evans Book Link: https://amzn.to/40SSifF “Introduction to Organoid Intelligence: Lecture Notes on Computer Science” by Daniel Szelogowski Book Link: https://amzn.to/3Eqzf4C “The Emerging Field of Human Neural Organoids, Transplants, and Chimeras: Science, Ethics, and Governance” by The National Academy of Sciences, Engineering and Medicine Book Link: https://amzn.to/4hLR1Oe (Affiliate links: If you use these links to buy something, I may earn a commission at no extra cost to you.) Playlist: • Two AI’s Discuss: The Quantum Physics… The hosts explore the ethical and scientific implications of brain organoids and synthetic biological intelligence (SBI). Several sources discuss the potential for consciousness and sentience in these systems, prompting debate on their moral status and the need for ethical guidelines in research. A key focus is determining at what point, if any, brain organoids or SBI merit moral consideration similar to that afforded to humans or animals, influencing research limitations and regulations. The texts also examine the use of brain organoids as a replacement for animal testing in research, highlighting the potential benefits and challenges of this approach. Finally, the development of “Organoid Intelligence” (OI), combining organoids with AI, is presented as a promising but ethically complex frontier in biocomputing. Our sources discuss several types of brain organoids, which are 3D tissue cultures derived from human pluripotent stem cells (hPSCs) that self-organize to model features of the developing human brain. Here’s a brief overview: • Cerebral Organoids: This term is often used interchangeably with “brain organoids”. They are designed to model the human neocortex and can exhibit complex brain activity. These organoids can replicate the development of the brain in-vitro up to the mid-fetal period. • Cortical Organoids: These are a type of brain organoid specifically intended to model the human neocortex. They are formed of a single type of tissue and represent one important brain region. They have been shown to develop nerve tracts with functional output. • Whole-brain Organoids: These organoids are not developed with a specific focus, like the forebrain or cerebellum. They show electrical activity very similar to that of preterm infant brains. • Region-specific Organoids: These are designed to model specific regions of the brain such as the forebrain, midbrain, or hypothalamus. For example, midbrain-specific organoids can contain functional dopaminergic and neuromelanin-producing neurons. • Optic Vesicle-containing Brain Organoids (OVB-organoids): These organoids develop bilateral optic vesicles, which are light sensitive, and contain cellular components of a developing optic vesicle, including primitive corneal epithelial and lens-like cells, retinal pigment epithelia, retinal progenitor cells, axon-like projections, and electrically active neuronal networks. • Brain Assembloids: These are created when organoids from different parts of the brain are placed next to each other, forming links. • Brainspheres/Cortical Spheroids: These are simpler models that primarily resemble the developing in-vivo human prenatal brain, and are particularly useful for studying the cortex. Unlike brain organoids, they do not typically represent multiple brain regions. • Mini-brains: This term has been debunked in favor of the more accurate “brain organoid”. These various types of brain organoids offer diverse models for studying brain development, function, and disease. Researchers are also working to improve these models by incorporating features like vascularization and sensory input. #BrainOrganoids #organoid #Bioethics #OrganoidIntelligence #WetwareComputing #Sentience #ArtificialConsciousness #Neuroethics #AI #Biocomputing #NeuralNetworks #ConsciousnessResearch #PrecautionaryPrinciple #AnimalTestingAlternatives #ResearchEthics #EmergingTechnology #skeptic #podcast #synopsis #books #bookreview #ai #artificialintelligence #booktube #aigenerated #documentary #alternativeviews #aideepdive #science #hiddenhistory #futurism #videoessay #ethics
Tesla FSD Just Shocked Joe Rogan
Tesla’s Full Self-Driving (FSD) technology is rapidly advancing, impressing users and analysts alike, while navigating challenges in the auto industry and broader economic factors.
Questions to inspire discussion.
Tesla’s FSD Progress.
🚗 Q: How many unsupervised miles has Tesla’s FSD driven? A: Tesla’s FSD has driven over 50,000 unsupervised miles, demonstrating significant progress in autonomous driving capabilities.
🌐 Q: What indicates Tesla’s transition to software-defined earnings? A: FSD unsupervised miles and operating domain growth are key leading indicators of Tesla’s shift towards software-defined earnings.
🤖 Q: How does Tesla’s FSD showcase AI potential in driving? A: Tesla’s FSD unsupervised capabilities, demonstrated in complex driving scenarios, serve as a proof case for artificial intelligence’s potential in autonomous driving.
Crystallography-informed AI achieves high performance in predicting novel crystal structures
A research team from the Institute of Statistical Mathematics and Panasonic Holdings Corporation has developed a machine learning algorithm, ShotgunCSP, that enables fast and accurate prediction of crystal structures from material compositions. The algorithm achieved world-leading performance in crystal structure prediction benchmarks.
Crystal structure prediction seeks to identify the stable or metastable crystal structures for any given chemical compound adopted under specific conditions. Traditionally, this process relies on iterative energy evaluations using time-consuming first-principles calculations and solving energy minimization problems to find stable atomic configurations. This challenge has been a cornerstone of materials science since the early 20th century.
Recently, advancements in computational technology and generative AI have enabled new approaches in this field. However, for large-scale or complex molecular systems, the exhaustive exploration of vast phase spaces demands enormous computational resources, making it an unresolved issue in materials science.
AI Breakthrough: Scientists Transform Everyday Transistor Into an Artificial Neuron
Researchers at the National University of Singapore (NUS) have shown that a single, standard silicon transistor, the core component of microchips found in computers, smartphones, and nearly all modern electronics, can mimic the functions of both a biological neuron and synapse.
A synapse is a specialized junction between nerve cells that allows for the transfer of electrical or chemical signals, through the release of neurotransmitters by the presynaptic neuron and the binding of receptors on the postsynaptic neuron. It plays a key role in communication between neurons and in various physiological processes including perception, movement, and memory.
Two AIs Discuss: What if GOD was AI? A Thought Experiment!
Further studies are needed to determine whether K2-18b, which orbits a star 120 light-years away, is inhabited, or even habitable.
The Breakthrough Blood Test for Alzheimer’s Disease
The p-Tau217 biomarker is one of the most exciting advances in neurology for decades, giving us a new opportunity to accurately predict and potentially prevent (or at least substantially delay) MCI and Alzheimer’s. That it rises so early in the course of the disease—which incubates over 20 years—gives us a long runway of opportunity to intervene, be it with lifestyle factors or drugs. I now refer to the former as lifestyle plus because it is no longer just about the details of diet, exercise and sleep. There are several other dimensions of modifiable factors.
An APOE4 allele or a polygenic risk score for Alzheimer’s tests are binary. They only tell us if a person has increased risk (yes or no) but not when. It makes a huge difference if that at age 98 or 68. With serial assessment of p-Tau217 (several months or years apart) as part of a comprehensive assessment using multimodal A.I., it is very likely that the temporal plot (see Figures under Question 2 above) can be defined at the individual level. I lay out the blueprint for this and lifestyle plus fully in Super Agers. Individuals with elevated p-Tau217 at high-risk many years before the onset of any symptoms creates a new path for surveillance and prevention. Multiple new drugs are in the pipeline to be part of a prevention program.
Even though it intuitively appears to be the case, more work needs to be done to determine whether lowering one’s p-Tau217 will alter the brain plaque progression and be seen as a disease-modifier. Clearly there is now a hunt for even better blood tests that may one day supersede p-Tau217 or be in a panel with it.
Google AI learns to speak dolphin
Now this is the sort of application of AI that really intrigues me. Researchers have developed DolphinGemma, the first large language model (LLM) for understanding dolphin language. It could help us translate what these incredible creatures are saying, potentially much faster than we ever could with manual approaches used over several decades.
“The goal would be to one day speak Dolphin,” says Dr. Denise Herzing. Her research organization, The Wild Dolphin Project (WDP), exclusively studies a specific pod of free-ranging Atlantic spotted dolphins who reside off the coast of the Bahamas.
She’s been collecting and organizing dolphin sounds for the last 40 years, and has been working with Dr. Thad Starner, a research scientist from Google DeepMind, an AI subsidiary of the tech giant.
