Futuristische ontwikkelingen

Deze afdeling is voor algemene topics die niet passen in wat reeds voorzien is. Ze moeten wel aansluiten bij ons thema.
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
29/03/2022 | Lydia Heida

Dit systeem combineert een fotonische processor met edge computing en een netwerk van nanomaterialen dat met complexe bewerkingen om kan gaan, voor ultrasnelle dataverwerking.

Steeds meer toepassingen vragen erom rekenkracht weer terug te halen dichtbij waar het gebeurt. Dit heet 'edge computing'. Zo voorkom je vertraging, zoals bij cloud computing, die te groot kan worden. Zelfs bij nieuwe communicatiestandaarden zoals 5G en 6G.

Edge computing introduceert wel een dilemma: zware computerkracht naar de lokale toepassing verplaatsen is niet wenselijk. De klassieke benadering van computers gaat nu eenmaal gepaard met veel dataverkeer tussen processor en geheugen.

In HYBRAIN (Hybrid electronic-photonic architectures for brain-inspired computing) willen onderzoekers dan ook een aantal innovatieve oplossingen bijeenbrengen.

De ingang van het systeem bestaat uit een fotonische processor die werkt met licht, omdat geïntegreerde fotonica erg goed is in het verwerken van veel data.

De data wordt vervolgens verdeeld over twee lerende netwerken: het één is gebaseerd op in memory computing, en is opgebouwd uit zogenaamde 'memristors': elektrische weerstanden die in staat zijn om hun instellingen te onthouden, zelfs als ze worden uitgeschakeld. Een netwerk van memristors kan lineaire operaties uitvoeren zoals vermenigvuldiging en optelling van grote datastromen.

Daarnaast krijgt het systeem een netwerk van nanomaterialen dat geen eigen ordening heeft, en met complexe, niet lineaire bewerkingen kan omgaan.

Omdat het totale systeem uit drie delen bestaat, is communicatie nodig tussen de onderdelen, maar niet zo'n intensief transport tussen geheugen en processor als in een conventioneel systeem.

Doel is om het het energieverbruik van HYBRAIN in de buurt te laten komen van het menselijk brein. Dan is een belangrijke stap gezet. Dit is bruikbaar in uiteenlopende technologieën: van autonoom rijdende auto’s, smartphones tot de deeltjesversneller van CERN in Genève, met zijn massieve hoeveelheden data.

Het HYBRAIN project wordt geleid door UTwente. Het consortium bestaat verder uit de Universiteit van Münster, University of Oxford, het Italiaanse bedrijf Trust-IT en IBM Research in Zürich. Het project loopt vier jaar en heeft een budget van drie miljoen euro vanuit Horizon Europe.
 
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Giulia Carbonaro • Updated: 25/05/2022 - 15:06

You might know Dyson for its sci-fi looking vacuum cleaners, futuristic-looking fans and its state-of-the-art hairdryers. But Dyson has been working on a secret project you probably do not know about and could realise a dream that humanity has been nurturing for decades now: robots.

And not just any robot (you may already consider some of their vacuum cleaners as robots, from a technical perspective). Dyson has revealed that it is betting on the construction of robots that will be able to do household chores. The company expects the new technology to be available by 2030.

Imagine mighty robotic hands gently picking up the dishes from the rack and setting up the table for lunch. A video shared by Dyson accompanying the announcement of their new business idea on Wednesday shows that this fantasy that looks straight out of a 1950s sci-fi film is now (almost) reality, as Dyson's prototypes are already able to pick up teddy bears from the floor and find crisps fallen into the cracks of the couch.

"There's a big future in robotics. Saving people time, performing chores for people and improving people's daily lives," said Jake Dyson, son of the billionaire founder of the company Sir James Dyson, in a promotional video.

According to Dyson, the new robots will be able to handle delicate tasks and deal with fragile objects, something that robotics has until now struggled to achieve to a satisfactory level.

Jake Dyson said the company has been doing research on robotics "that nobody is aware of" for the past 20 years.

For the past ten years, according to Dyson, the company has been sponsoring research at London's Imperial College. They're now setting up three more offices, two in the United Kingdom and one in Singapore.

Dyson said that the research is "very top secret" and the company is only now revealing snippets of information to attract the best engineers around the world. Dyson is looking to hire 700 robotics engineers to work on the "brain" of these household chores robots.

If successful, the move will be a revolution in robotics, finally bringing this type of robot, which is already used in factory settings, into people's homes.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
31-05-2022 10:12, door Cédric Van Loon

De Amerikaanse Frontier-installatie is de eerste exascale supercomputer die de TOP 500-lijst haalt. AMD is hofleverancier van de installatie met CPU's en GPU's.

In de nieuwe TOP 500-lijst van supercomputers wordt voor het eerst de exascale-barrière doorbroken met de Amerikaanse Frontier supercomputer. Die werd eerder dit jaar in productie genomen nadat de Amerikaanse overheid in 2019 budget had vrijgemaakt voor het exascaleproject. Het doel was 1,5 exaflops, vandaag haalt de machine 1,1 exaflops. De totale kost van het systeem wordt op 500 miljoen dollar geschat.

Frontier steekt de Japanse Fugaku supercomputer voorbij in de lijst. Die laatste haalt 442 petaflops in de HPL-benchmark. Volgens Fujitsu, de fabrikant van de Fugaku-installatie, heeft het systeem wel een theoretische topsnelheid van meer dan één exaflop.

De Frontier supercomputer werd samengesteld door Cray met AMD als hofleverancier. Iets na de aankondiging kocht HPE Cray voor 1,3 miljard dollar. Een jaar later werd de hele line-up geïntegreerd in het aanbod van HPE.

AMD en Cray bouwden Frontier

Frontier bestaat uit 74 rackkasten die elk 128 blades bevatten. Elke blade heeft een derde generatie AMD Epyc 64C 2 GHz CPU aan boord en vier AMD Instinct MI250x GPU's. In totaal bevat Frontier 9.408 CPU's en 37.632 GPU's die allemaal heel wat warmte opwekken. De koelinstallatie verbruikt elke minuut 22.000 liter water, luchtkoeling is niet langer haalbaar bij dit type installatie.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
Mark Klaver 19 juni 2022 17:03

Europa's eerste exascale supercomputer JUPITER komt volgend jaar in het Duitse onderzoekscentrum Forschungszentrum Jülich te staan. De computer gaat zo'n 500 miljoen euro kosten, waarvan de helft door het EuroHPC JU, een samenwerkingsverband van Europese landen voor de ontwikkeling van supercomputers, betaald wordt. De rest wordt in Duitsland zelf gefinancierd.

De JUPITER gaat vooral gebruikt worden voor onderzoek naar klimaatverandering, het beheersen van pandemieën, duurzame energieproductie en kunstmatige intelligentie.

Met exascale wordt een snelheid van minstens 1 exaflops bedoeld: 1 triljoen (de Nederlandse triljoen, een 1 met 18 nullen) 'floating point operations' per seconde. De computer zou daarmee op de eerste of tweede plek in de TOP500-lijst van snelste computers ter wereld komen te staan, nek aan nek met de Amerikaanse Frontier, die een peak-snelheid van 1,68 exaflops haalt.

Om die snelheid te behalen is flink wat stroom nodig, tot 15 megawatt (15.000.000 watt). Dat is heel wat(t) natuurlijk, maar een stuk minder dan de Frontier, die 21 megawatt verbruikt. Het is de bedoeling dat de JUPITER geheel op groene stroom gaat draaien, en zelfs de restwarmte van het waterkoelingssysteem hergebruikt wordt, bijvoorbeeld voor de verwarming van het onderzoekscentrum.

Het is nog niet bekend wie de processors gaat leveren voor de JUPITER. EuroHPC's huidige topmodel, de LUMI (die in Finland staat), is opgebouwd uit 75.264 AMD Epyc-cpu's, met 4.704 AMD Instinct MI250X-accelerators. LUMI's peak-snelheid ligt rond de 550 petaflops (0,55 exaflops) trouwens, dus de JUPITER zal een flinke upgrade zijn.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Daniela La Marca, April 2022

AI research is progressing at a rapid pace and now Google has unveiled its Pathways Language Model (PaLM), the largest language model presented so far.

In a blog post, the Google research team explains that its Pathways Language Model (PaLM) has created an AI that with more than 540 billion parameters is larger than the Megatron Turing-Natural Language Generation Model (MT-NLG) of Microsoft and Nvidia. But most importantly, in various tests, the model was so convincing that it could keep up with human performance.

During the development of the Gopher language model, the researchers at the Alphabet company, Deepmind, realized that simply scaling the model does not necessarily lead to better results. That's why they have meanwhile developed a much smaller model called Retro, which due to a peculiarity can still keep up with AI that contains up to 25 times the parameters. In fact, Retro can access a database of two trillion pieces of text to look for passages of similar language that could improve its predictions.

The fact is that Google no longer relies on sheer size when developing PaLM, although it is of course huge. The new PaLM system combines the power of AI with a form of multitasking to increase performance. This multitasking ability, which Google calls Pathways, has now been used for the first time to support a language model. The researchers have thus succeeded in creating a model whose considerable quantitative values are not only used for qualitative statement. The PaLM model also performed excellently in the usual tests for assessing the performance of a language model.

In almost all tests, PaLM is said to have left the competition far behind. These were mostly monolingual tests such as simple question-and-answer tests, fill-in-the-blanks, sentence completions, reading comprehension and logical reasoning tests, and tests in which it is important to draw the right conclusions from statements made in natural language. According to Google, PaLM was able to demonstrate abilities in such tests that are at the level of language comprehension for 9- to 12-year-olds.

The system also performed "strongly" when it came to translation tasks. The same applies to demanding learning with comparatively little information, the so-called "few-shot" learning. Again, PaLM could have achieved average human scores on these tests.

Despite pathways and retro, the research team believes that an increase in performance is possible, which can result from pure scaling of the models. This suggests that the PaLM model's 540 billion parameters may soon be exceeded. But of course, this sheer size of AI models can also lead to increasing opacity, making biases in current models increasingly difficult to detect. And this is a serious ethical problem that currently has no solution.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
Published 16 June 2022

In a paper published in Science Advances, researchers at the University of Oxford have developed a method using the polarisation of light to maximise information storage density and computing performance using nanowires.

Light has an exploitable property – different wavelengths of light do not interact with each other – a characteristic used by fibreoptics to carry parallel streams of data. Similarly, different polarisations of light do not interact with each other either. Each polarisation can be used as an independent information channel, enabling more information to be stored in multiple channels, hugely enhancing information density.

First author and DPhil student June Sang Lee, Department of Materials, University of Oxford said: 'We all know that the advantage of photonics over electronics is that light is faster and more functional over large bandwidths. So, our aim was to fully harness such advantages of photonics combining with tunable material to realise faster and denser information processing.'

In collaboration with Professor C David Wright, University of Exeter, the research team developed a HAD (hybridized-active-dielectric) nanowire, using a hybrid glassy material which shows switchable material properties upon the illumination of optical pulses. Each nanowire shows selective responses to a specific polarisation direction, so information can be simultaneously processed using multiple polarisations in different directions.

Using this concept, researchers have developed the first photonic computing processor to utilise polarisations of light.

Photonic computing is carried out through multiple polarisation channels, leading to an enhancement in computing density by several orders compared to that of conventional electronic chips. The computing speeds are faster because these nanowires are modulated by nanosecond optical pulses. The new chip promises to be more than 300 times faster and denser than current electronic chips.

Since the invention of the first integrated circuit in 1958, packing more transistors into a given size of an electronic chip has been the go-to means of maximising computing density – the so-called 'Moore's Law'. However, with Artificial Intelligence and Machine Learning requiring specialised hardware that is beginning to push the boundaries of established computing, the dominant question in this area of electronic engineering has been 'How do we pack more functionalities into a single transistor?'

For over a decade, researchers in Professor Harish Bhaskaran’s lab in the Department of Materials, University of Oxford have been looking into using light as a means to compute.

Professor Bhaskaran, who led the work, said: 'This is just the beginning of what we would like to see in future, which is the exploitation of all degrees of freedoms that light offers, including polarisation to dramatically parallelise information processing. Definitely early-stage work – our speed estimates still need research to verify them experimentally – but super exciting ideas that combine electronics, non-linear materials and computing. Lots of exciting prospects to work on which is always a great place to be in!'

The full paper, Polarisation-selective reconfigurability in hybridized-active-dielectric nanowires, is published in Science Advances.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
British startup Opteran, a spin–out of the University of Sheffield, has a completely different view of neuromorphic engineering compared to most of the industry. The company has reverse–engineered insect brains to derive new algorithms for collision avoidance and navigation that can be used in robotics.

Opteran calls its new approach to AI "natural intelligence," taking direct biological inspiration for the algorithm portion of the system. This approach is separate to existing computer vision approaches, which mainly use either mainstream AI/deep learning or photogrammetry, a technique that uses 2D photographs to infer information about 3D objects, such as dimensions.

Opteran's natural intelligence requires no training data, and no training, more like how a biological brain works. Deep learning today is capable of narrow AI — it can execute carefully defined tasks within a limited environment such as a computer game — but huge amounts of training data is required, as are computation and power consumption. Opteran wants to get around the limitations of deep learning by closely mimicking what brains really do, in order to build autonomous robots that can interact with the real world while on a tight computation and energy budget.

"Our purpose is to reverse– or re–engineer nature's algorithms to create a software brain that enables machines to perceive, behave, and adapt more like natural creatures," said professor James Marshall, chief scientific officer at Opteran, in a recent presentation at the Embedded Vision Summit.

"Imitating the brain to develop AI is an old idea, going back to Alan Turing," he said. "Deep learning, on the other hand, is based on a cartoon of a tiny part of the primate brain visual cortex that ignores the vast complexity of a real brain… modern neuroscience techniques are increasingly being applied to give the information we need to faithfully reverse engineer how real brains solve the problem of autonomy."

Reverse engineering brains requires studying animal behavior, neuroscience, and anatomy together. Opteran has been working with honeybee brains as they are both sufficiently simple and capable of orchestrating complex behavior. Honeybees are able to navigate over distances of 7 miles, and communicate their mental maps accurately to other bees. It does all this with fewer than a million neurons, in an energy–efficient brain the size of a pinhead.

Opteran has successfully reverse–engineered the algorithm honeybees use for optical flow estimation (the apparent motion of objects in a scene caused by relative motion of the observer). This algorithm can do optical flow processing at 10 kHz for under a Watt, running on a small FPGA.

"This performance exceeds the deep learning state of the art by orders of magnitude in all dimensions, including robustness, power, and speed," Marshall said.

BIOLOGICAL ALGORITHMS
Biological motion detection was mathematically modeled in the 1960s based on experiments with insect brains. The model is called the Hassenstein–Reichardt Detector and it has been verified many times over via different experimental methods. In this model, the brain receives signals from two neighboring receptors in the eye. The input from one receptor is delayed. If the brain receives both signals at the same time, the neuron fires, because it means the object you're looking at is moving. Doing this again with the other signal delayed means it works if the object is moving in either direction (hence the symmetry in the model).

Marshall explained in his presentation that the Hassenstein–Reichardt Detector, while sufficient to model motion detection in fruit flies, is highly sensitive to spatial frequency (the distribution pattern of dark and light in an image) and contrast, and therefore not a great fit for generalized visual navigation.

"Honeybees do something cleverer, which is a novel arrangement of these elementary units," Marshall said. "Honeybee flying behavior shows great robustness to spatial frequency and contrast, so there must be something else going on."

Opteran used behavioral and neuroscientific data from honeybees to come up with its own visual inertial odometry estimator and collision avoidance algorithm (on the right in the diagram above). This algorithm was benchmarked and found to be superior to FlowNet2s (a state–of–the–art deep learning algorithm at the time), in terms of theoretical accuracy and noise robustness. Marshall points out that the deep learning implementation would also require GPU acceleration, with the associated power penalty.

REAL–WORLD ROBOTICS
It's a nice theory, but does it work in the real world? Opteran has indeed been applying its algorithms in real–world robotics. The company has developed a robot dog demo, Hopper, in a similar form factor to Boston Dynamics' Spot. Hopper uses an edge–based vision–only solution based on Opteran's collision prediction and avoidance algorithm; when a potential collision is identified, a simple controller makes it turn away.

Opteran is also working on a 3D navigation algorithm, again based on honeybees. This solution will be equivalent to today's SLAM (simultaneous location and mapping) algorithms, but it will also handle path planning, routing, and semantics. Marshall said it will run on a fraction of a Watt on the same hardware.

"Another big saving is in terms of the map size generated by this approach," he said. "Whereas classical photogrammetry–based SLAM generates map sizes in the order of hundreds of megabytes to gigabytes per meter squared, causing significant problems for mapping large areas, we have maps consuming only kilobytes of memory."

A demo of this algorithm powering a small drone in flight uses a single low–resolution camera (less than 10,000 pixels) to perform autonomous vision–based navigation.

HARDWARE AND SOFTWARE
Opteran's development kit uses a small Xilinx Zynqberry FPGA module which weighs less than 30g and consumes under 3W. It requires two cameras. The development kit uses cheap ($20) Raspberry Pi cameras, but Opteran will work with OEMs to calibrate algorithms for other camera types during product development.

The current FPGA can run Opteran's omnidirectional optical flow processing and collision prediction algorithms simultaneously. Future hardware may migrate to larger FPGAs or GPUs as required, Marshall said.

The company is building a software stack for robotics applications. On top of an electronically stabilized panoramic vision system, there is collision avoidance, then navigation. Work is underway on a decision engine to allow a robot to decide where it should go and under what circumstances (due in 2023). Future elements include social, causal, and abstract engines, which will allow robots to interact with each other, to infer causal structures in real world environments, and to abstract general principles from experienced situations. All these engines will be based on biological systems — no deep learning or rule–based systems.

Opteran completed a funding round of $12 million last month, which will fund the commercialization of its natural intelligence approach and the development of the remaining algorithms in its stack. Customer pilots so far have used stabilized vision, collision avoidance, and navigation capabilities in cobot arms, drones, and mining robots.

Future research directions could also include studying other animals with more complex brains, Marshall said.

"We started with insects, but the approach scales," he said. "We'll be looking at vertebrates in due course, that's absolutely on our roadmap."
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
2022-07-29 14:55 HKT

At present, the most advanced process for mass production by TSMC and Samsung is 5nm, and it will enter the 3nm process node in 2022, and the next two have planned for the 2nm process, but the mass production time is still uncertain.

A few days ago, Dr. Luc Van den hove, CEO of IMEC European Microelectronics Center in Belgium, announced the roadmap of the chip process, believing that Moore's Law will continue.

According to him, the industry will mass produce the 2nm process around 2025, the 1nm process will be mastered around 2027, and the 0.7nm process will go straight to the 0.7nm process in 2029. At this time, it has actually entered the Emmy era.

However, the roadmap given by Luc Van den hove is only preliminary, and there is no detailed technical details. The process after 3nm requires the upgrade of transistor materials and manufacturing equipment, such as upgrading GAA transistors, and the lithography machine must also upgrade the next generation of high NA. The (numerical aperture) standard has been increased from the current 0.33 NA to 0.55 NA. Higher NA means higher resolution, which is a necessary condition for the process after 3nm.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
July 25, 2022, 6:25 PM by Alan Friedman

For the time being, Moore's Law, the observation made by semiconductor legend Gordon Moore, remains alive. Moore, the co-founder of Fairchild Semiconductor and Intel, originally noted in 1965 that the number of transistors inside an integrated circuit (IC) would double every year. He later revised that in the 1970s by stating that the transistor count would double every other year.

The process node used by top foundries like TSMC and Samsung for building cutting-edge chips gives us an idea of how Moore's Law is doing. The smaller the process node, the larger the number of transistors that can fit inside an IC. And that is important because the higher the transistor count, the more powerful and energy-efficient a chip is.

Samsung started shipping its 10nm SoCs in 2016. In 2018 it started mass production of its 7nm chipsets. By 2020, Samsung started mass production of 5nm chipsets, and now Samsung (via wccftech) has become the first to start shipping 3nm GAA chipsets beating out rival TSMC. Not only is Samsung the first to deliver 3nm chips, but it is also the first to ship these chips equipped with GAA or gate-all-around transistors.

With GAA, there is more control over current flow resulting in greater power efficiency. TSMC is still using the previous generation FinFET transistor design for its 3nm SoCs which it will start shipping during the second half of this year. The world's leading independent foundry will start using GAA with its 2nm process node which it hopes to start delivering to customers in 2026.

Samsung was so stoked to beat out TSMC with its 3nm deliveries that the company held a ceremony at its Hwaseong Campus, in Gyeonggi-do, which was attended by several Samsung executives and Korean politicians. The first batch of 3nm SoCs are not being shipped to smartphone manufacturers. Instead, they will be used in equipment employed by cryptocurrency miners, with the new 3nm GAA allowing for a big reduction in power consumption.

Eventually, the 3nm GAA process node will be used to produce smartphone chips including Samsung's own Exynos 2300 and possibly the Qualcomm Snapdragon 8 Gen 2 SoCs. The 3nm GAA process node will reduce power consumption by up to 45% and increase performance by as much as 23% when compared to the 5nm node. A second generation variant of the 3nm GAA chips is expected to reduce energy consumption by up to 50%, and increase performance by up to 30%.

What this will mean to you is the availability of more powerful handsets with improved battery life.

In a release, Samsung said, "On the 25th, Samsung Electronics held a 3nm foundry product shipment ceremony using next-generation transistor GAA (Gate All Around) technology at the V1 line (EUV only) at Hwaseong Campus, Gyeonggi-do. The event was attended by about 100 people including Minister of Trade, Industry and Energy Changyang Lee, suppliers, fabless, Samsung Electronics DS division head, Kyeong-hyeon Kye (President), and executives and employees, and encouraged executives and employees who participated in 3nm GAA R&D and mass production."
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
Ali Yas
Berichten: 7662
Lid geworden op: zo apr 15, 2012 3:24 pm
Contacteer:

Re: Futuristische ontwikkelingen

Bericht door Ali Yas »

Het blijft fascinerend, die Wet van Moore. Ik kan me nog herinneren dat geopperd werd dat de technologie met elektrische stroom aan z'n top zat en dat zou moeten worden overgeschakeld op componenten die met licht werken, maar dat valt dus nogal mee. :victory:
Truth sounds like hate to those who hate truth.
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
July 27, 2022 by David Edwards

Rapid Robotics says it has developed a robot with the "fastest setup time in the industry".

The Rapid Machine Operator (RMO) from Rapid Robotics is "the first industrial robot that can get up and running in a matter of hours", according to the company.

It's also the first manufacturing robot that can move between jobs "in less time than it takes to make a cup of coffee".

Rapid Machine says the secret is Smart Setup, a "revolutionary" new capability that enables anyone to move Rapid Machine Operators onto a new job in as little as 60 seconds via built-in computer vision – no programming or robotics expertise required.

In addition to easy task-switching, Smart Setup puts an end to work stoppages caused by changes or disruptions to the work environment. Other robots need to be reset by trained specialists. With Rapid's Smart Setup, anyone can get an RMO back in action in minutes.

The Smart Setup process is push-button simple. With one touch in Rapid Robotics' convenient tablet-based app, the RMO's cameras analyze the workspace. With another, the RMO calculates distances and orientations of parts and surfaces, then updates motion paths to deliver the fastest and most precise way to execute on the given task. With just one more tap, the RMO starts the new job.

"That's all there is to it," says Rapid. "No support tickets. No systems integrators required. Smart Setup handles everything automatically using the Rapid Machine Operator's built-in AI, running on Rapid's proprietary edge-computing platform."

Tammy Barras, president of Westec Plastics, says: "Smart Setup is shaping up to be an absolute game-changer for busy high-mix environments like ours.

"We can't pay systems integrators to come in every time we want a robot working on a different task. We'd lose more money than we'd save. With Smart Setup, we'll be able to do everything ourselves, with practically no downtime. That means we'll spend less, get more work done and bring in more revenue overall."

Launched in 2020, the Rapid Machine Operator is the first robot designed to solve America's chronic machine operator shortage and help manufacturing businesses regain their competitive advantage.

The RMO arrives ready to work ‘out of the box,' pre-trained to perform hundreds of common machine-tending and warehouse tasks such as pick-and-place, parts inspection, gate cutting and many more.

The RMO comes with all the hardware and software it needs to get started, including a six-axis robotic arm, computer vision system and custom grippers. Everything is pre-integrated so the RMO can get to work the day it arrives. Each RMO can be "hired" for an OpEx-friendly subscription of $2,100 a month.

Fast Company magazine recognized the Rapid Machine Operator as one of its "Next Big Things in Technology" for 2021.

Jordan Kretchmer, Rapid Robotics CEO, says: "Smart Setup gives the Rapid Machine Operator human-like instincts about its surroundings and its assigned task.

"It's so flexible, you can almost think of it as a human colleague. You can drop it into any job it's trained for, at a moment's notice. There's never been anything like it on a factory floor.

"The Rapid Machine Operator with Smart Setup removes the barriers to automation, dramatically reduces cost of ownership and makes robotics available to manufacturers of all sizes, including hundreds of thousands of SMB facilities in the United States alone. It's how we're making manufacturers more profitable, more agile and more competitive in the global market."

Existing customers will receive Smart Setup along with Rapid's regular free updates to the RMO's software and AI.
 
Het ziet er dus naar uit dat er voor de functie Machineoperator binnenkort geen mensen meer zullen worden ingezet.
 
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
Ali Yas
Berichten: 7662
Lid geworden op: zo apr 15, 2012 3:24 pm
Contacteer:

Re: Futuristische ontwikkelingen

Bericht door Ali Yas »

xplosive schreef:
za jul 30, 2022 7:27 pm
Het ziet er dus naar uit dat er voor de functie Machineoperator binnenkort geen mensen meer zullen worden ingezet.
Dat zal voor steeds meer beroepen het geval zijn, tenminste voor zover het massa- en serieproductie betreft. En eigenlijk is dat ook wel goed, mensen kunnen zich dan bezighouden met leukere dingen. Het is alleen zo jammer dat van dat laatste niets terecht komt omdat alle winst die met productiviteitsverhoging wordt geboekt verdwijnt in de steeds diepere zakken van overheden, die die waarde vervolgens besteden aan andere doelen dan jij en ik wel zouden willen.
Truth sounds like hate to those who hate truth.
Gebruikersavatar
King George
Berichten: 24957
Lid geworden op: zo sep 11, 2011 1:22 pm

Re: Futuristische ontwikkelingen

Bericht door King George »

In Japan zijn ze al ver gevorderd met robots in de samenleving in te zetten.
Het morele gelijk ligt bij het volk




Citaten van Mustafa Kemal Atatürk over de Islam
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Chris Young Jul 26, 2022 5:14 PM EDT

Nuclear fusion promises practically limitless energy and an unshackling from the harmful impact of fossil fuel consumption.

Now, researchers from the U.S. Department of Energy's (DOE) Princeton Plasma Physics Laboratory (PPPL) announced they found a way to build powerful magnets much smaller than ever before, a press statement reveals.

The new innovation could help in the development of tokamak reactors, unlocking the potential of nuclear fusion.

We're on the verge of viable nuclear fusion
The scientists found a new method for building high-temperature superconducting magnets that are made of material that conducts electricity with practically no resistance at temperatures warmer than before. The smaller magnets will more easily fit inside spherical tokamaks, which are being investigated as a potential alternative to the more conventional doughnut-shaped tokamaks.

Fusion scientists and engineers use these incredibly powerful magnets to control and maintain the hot plasma required for the nuclear fusion reaction to take place. Crucially, the new magnets could be placed separately from other machinery in the spherical tokamak's central cavity. This means scientists would be able to repair them without having to dismantle any other parts of the tokamak.

"To do this, you need a magnet with a stronger magnetic field and a smaller size than current magnets," explained Yuhu Zhai, a principal engineer at PPPL and lead author of a paper on the new magnets published in IEEE Transactions on Applied Superconductivity. "The only way you do that is with superconducting wires, and that's what we've done."

The magnets could also potentially allow scientists to develop smaller tokamaks, which could improve performance as well as reduce the cost of construction and operation. "Tokamaks are sensitive to the conditions in their central regions, including the size of the central magnet, or solenoid, the shielding, and the vacuum vessel," said Jon Menard, PPPL's deputy director for research. "A lot depends on the center. So if you can shrink things in the middle, you can shrink the whole machine and reduce cost while, in theory, improving performance."

New technique makes magnets cheaper, smaller, and more powerful
The new magnets were designed using a technique developed by Zhai and colleagues at Advanced Conductor Technologies, the University of Colorado, Boulder, and the National High Magnetic Field Laboratory, in Tallahassee, Florida. They devised a technique that does not require traditional epoxy and glass fiber insulation for their magnet's wires, allowing them to reduce the size.

By removing epoxy from the equation, the researchers also lower the cost of magnet production, which will also result in cheaper tokamaks. The costs to wind the coils are much lower because we don't have to go through the expensive and error-prone epoxy vacuum-impregnation process," Zhai said. "Instead, you're directly winding the conductor into the coil form."

The smaller magnets will, in theory, allow for more design iterations of tokamaks, as they can be more easily placed in different locations, allowing for more configurations. We may still be a long way from seeing the first fully operational fusion reactor, but this new development brings us one step closer to commercially viable nuclear fusion.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Ed Browne on 7/25/2022 AT 5:03 AM EDT

Scientists at a U.S. government plasma lab have discovered a missing component in nuclear fusion equations that could speed up development of a working reactor.

Specifically, the discovery could improve the design of the donut-shaped fusion reactors known as tokamaks.

Nuclear fusion is the process of joining two atomic nuclei together in order to form a single, larger nucleus whilst releasing energy in the process. It's the same process that powers our sun, where hydrogen atoms are fused together to form helium.

Scientists have been working on nuclear fusion reactors for decades since fusion promises to be a clean, safe, and virtually limitless energy source. However, scientists have yet to achieve a stable reaction that gives out more energy than it consumes.

Tokamaks work by creating a material known as plasma, in which an element—usually hydrogen—is heated so much that it becomes an electrically-charged soup of electrons and atomic nuclei. Powerful magnets then contain this plasma into a safe, stable flow, creating conditions where fusion should be possible.

In order to perfect tokamak designs, scientists use computer models to predict how the plasma will act under certain conditions. Now, scientists at the Princeton Plasma Physics Laboratory (PPPL), a U.S. Department of Energy lab managed by Princeton University, have found that the equations used to create these computer models have been missing an important detail — resistivity.

Resistivity refers to the ability of any material or substance to prevent the flow of electricity. Just like how a rock will move more easily through air than through water, electricity moves more easily through some things than others.

In a study, PPPL scientists have found that resistivity is an important property of plasma since it can cause instabilities known as edge-localized modes (ELMs), which are essentially small eruptions of plasma. If left unchecked, these eruptions could cause damage to fusion reactors which would mean they'd need to be taken offline more often for repairs.

"We need to have confidence that the plasma in these future facilities will be stable without having to build full-scale prototypes, which is prohibitively expensive and time-consuming," said Nathaniel Ferraro, a PPPL researcher, in a press release. "In the case of edge-localized modes and some other phenomena, failing to stabilize the plasma could lead to damage or reduced component lifetimes in these facilities, so it's very important to get it right."

Having accurate computer models is important, since it means scientists can use time and money as efficiently as possible to build a reactor they know will probably work well, rather than wasting resources on a trial-and-error approach.

"You want a model that is simple enough to calculate but complete enough to capture the phenomenon you are interested in," said Ferraro. "Andreas found that resistivity is one of the physical effects that we should include in our models."

The PPPL study was published in the journal Nuclear Fusion in May.

Future research will investigate which specific tokamak properties cause these resistive plasma eruptions to occur, which could result in improved designs.

Earlier this month, a report found private investment in fusion companies had skyrocketed.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Charles Q. Choi | 06 August 2022

New artificial versions of the neurons and synapses in the human brain may be as small as one-thousandth the size of neurons and at least 10,000 times as fast as biological synapses, a study now finds.

These new devices may help improve the speed at which the increasingly common and powerful artificial-intelligence systems known as deep neural networks learn, researchers say.

In artificial neural networks, electrical components dubbed "neurons" are fed data and cooperate to solve a problem, such as recognizing images. The neural net repeatedly adjusts the links between its ersatz neurons and sees if the resulting patterns of behavior are better at finding a solution. Over time, the network discovers which patterns are best at computing results. It then adopts these as defaults, mimicking the process of learning in the human brain.

A neural network is dubbed "deep" if it possesses multiple layers of neurons. Deep neural networks are increasingly finding use in applications such as analyzing medical scans,designing microchips, predicting how proteins fold, and empowering autonomous vehicles.

The amount of time, energy, and money needed to train deep neural networks is skyrocketing. One approach that researchers are pursuing to help overcome this challenge involves training brain-imitating deep neural networks on brain-mimicking hardware instead of conventional computers, a strategy called analog deep learning.

Just as transistors are the core elements of digital computers, so too are neuron- and synapselike components the key building blocks in analog deep learning. In the new study, researchers experimented with artificial synapses called programmable resistors.

The new programmable resistors are similar to memristors, or memory resistors. Both kinds of devices are essentially electric switches that can remember which state they were toggled to after their power is turned off. As such, they resemble synapses, whose electrical conductivity strengthens or weakens depending on how much electrical charge has passed through them in the past. Memristors are two-terminal devices, whereas the new programmable resistors are three-terminal devices, says study lead author Murat Onen, an electrical engineer at MIT.

The research team's programmable resistors increased or decreased their electrical conductance by moving protons around. To increase conductance, electric fields helped insert protons into the devices. To decrease conductance, protons were taken out.

These protonic programmable resistors used an electrolyte similar to those found in batteries to let protons pass while blocking electrons. Their electrolyte was phosphosilicate glass, which the researchers suspected would possess high proton conductivity at room temperature. This glass accommodated many nanometer-size pores for proton transport and could also withstand very strong pulsed electric fields to help protons move quickly.

Unlike the organic Nafion electrolyte used in an earlier version of the team's device, phosphosilicate glass is compatible with silicon fabrication techniques. This helped scale the devices "all the way down to 10-nanometer scale," Onen says. In contrast, biological neurons are roughly 1,000 times as long.

The speed at which biological neurons and synapses can process and transfer data is limited by the weak voltages and watery medium in which these signals are shuffled. Anything more than 1.23 volts causes liquid water to split into hydrogen and oxygen gas. As such, the speed of thought in animals is typically limited to millisecond timescales. In contrast, artificial solid-state neurons and synapses are not hemmed in by these constraints. However, it was unclear how fast they were compared with their biological counterparts.

In experiments, the scientists found their protonic programmable resistors could perform at least 10,000 times as fast as biological synapses at room temperature. "The most surprising part was to see how fast we could move protons within solid media," Onen says. "Previously the operation timescales were around milliseconds, whereas in this work we achieved nanoseconds."

The devices could run for millions of cycles without breaking down. Furthermore, the amount of heat they generated during computation compared well with that of human synapses. The insulating properties of the phosphosilicate glass meant that almost no electric current passed through the material as protons moved, making the gadgets highly energy efficient.

"The primary technological implication is that we can now have protonic programmable devices for analog deep learning applications," Onen says. "Predecessors of such devices already had many promising qualities compared to competing technologies but were very slow, which meant they were not appropriate to be used in processors."

In addition, Onen says, "the discovery of ultrafast ion transport in solids could have broader implications beyond analog deep learning, whenever fast ion motion is required, such as in microbatteries, fuel cells, artificial photosynthesis, and electrochromism."

The scientists detailed their findings in the 29 July issue of the journal Science.
 
Dit is verreweg het meest shockerend griezelig nieuws dat ik sinds tijden heb gezien. Dit zou kunstmatige breinen mogelijk kunnen maken die én veel sneller én veel kleiner én veel energiezuiniger zijn dan menselijke breinen.
 
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Kevin Roose, Aug. 24, 2022

We're in a golden age of progress in artificial intelligence. It's time to start taking its potential and risks seriously.

For the past few days, I've been playing around with DALL-E 2, an app developed by the San Francisco company OpenAI that turns text descriptions into hyper-realistic images.

OpenAI invited me to test DALL-E 2 (the name is a play on Pixar's WALL-E and the artist Salvador Dalí) during its beta period, and I quickly got obsessed. I spent hours thinking up weird, funny and abstract prompts to feed the A.I. — "a 3-D rendering of a suburban home shaped like a croissant," "an 1850s daguerreotype portrait of Kermit the Frog," "a charcoal sketch of two penguins drinking wine in a Parisian bistro." Within seconds, DALL-E 2 would spit out a handful of images depicting my request — often with jaw-dropping realism.

Here, for example, is one of the images DALL-E 2 produced when I typed in "black-and-white vintage photograph of a 1920s mobster taking a selfie." And how it rendered my request for a high-quality photograph of "a sailboat knitted out of blue yarn."

DALL-E 2 can also go more abstract. The illustration at the top of this article, for example, is what it generated when I asked for a rendering of "infinite joy." (I liked this one so much I'm going to have it printed and framed for my wall.)

What's impressive about DALL-E 2 isn't just the art it generates. It's how it generates art. These aren't composites made out of existing internet images — they're wholly new creations made through a complex A.I. process known as "diffusion," which starts with a random series of pixels and refines it repeatedly until it matches a given text description. And it's improving quickly — DALL-E 2's images are four times as detailed as the images generated by the original DALL-E, which was introduced only last year.

DALL-E 2 got a lot of attention when it was announced this year, and rightfully so. It's an impressive piece of technology with big implications for anyone who makes a living working with images — illustrators, graphic designers, photographers and so on. It also raises important questions about what all of this A.I.-generated art will be used for, and whether we need to worry about a surge in synthetic propaganda, hyper-realistic deepfakes or even nonconsensual pornography.

But art is not the only area where artificial intelligence has been making major strides.

Over the past 10 years — a period some A.I. researchers have begun referring to as a "golden decade" — there's been a wave of progress in many areas of A.I. research, fueled by the rise of techniques like deep learning and the advent of specialized hardware for running huge, computationally intensive A.I. models.

Some of that progress has been slow and steady — bigger models with more data and processing power behind them yielding slightly better results.

But other times, it feels more like the flick of a switch — impossible acts of magic suddenly becoming possible.

Just five years ago, for example, the biggest story in the A.I. world was AlphaGo, a deep learning model built by Google's DeepMind that could beat the best humans in the world at the board game Go. Training an A.I. to win Go tournaments was a fun party trick, but it wasn't exactly the kind of progress most people care about.

But last year, DeepMind's AlphaFold — an A.I. system descended from the Go-playing one — did something truly profound. Using a deep neural network trained to predict the three-dimensional structures of proteins from their one-dimensional amino acid sequences, it essentially solved what's known as the "protein-folding problem," which had vexed molecular biologists for decades.

This summer, DeepMind announced that AlphaFold had made predictions for nearly all of the 200 million proteins known to exist — producing a treasure trove of data that will help medical researchers develop new drugs and vaccines for years to come. Last year, the journal Science recognized AlphaFold's importance, naming it the biggest scientific breakthrough of the year.

Or look at what's happening with A.I.-generated text.

Only a few years ago, A.I. chatbots struggled even with rudimentary conversations — to say nothing of more difficult language-based tasks.

But now, large language models like OpenAI's GPT-3 are being used to write screenplays, compose marketing emails and develop video games. (I even used GPT-3 to write a book review for this paper last year — and, had I not clued in my editors beforehand, I doubt they would have suspected anything.)

A.I. is writing code, too — more than a million people have signed up to use GitHub's Copilot, a tool released last year that helps programmers work faster by automatically finishing their code snippets.

Then there's Google's LaMDA, an A.I. model that made headlines a couple of months ago when Blake Lemoine, a senior Google engineer, was fired after claiming that it had become sentient.

Google disputed Mr. Lemoine's claims, and lots of A.I. researchers have quibbled with his conclusions. But take out the sentience part, and a weaker version of his argument — that LaMDA and other state-of-the-art language models are becoming eerily good at having humanlike text conversations — would not have raised nearly as many eyebrows.

In fact, many experts will tell you that A.I. is getting better at lots of things these days — even in areas, such as language and reasoning, where it once seemed that humans had the upper hand.

"It feels like we're going from spring to summer," said Jack Clark, a co-chair of Stanford University's annual A.I. Index Report. "In spring, you have these vague suggestions of progress, and little green shoots everywhere. Now, everything's in bloom."

In the past, A.I. progress was mostly obvious only to insiders who kept up with the latest research papers and conference presentations. But recently, Mr. Clark said, even laypeople can sense the difference.

"You used to look at A.I.-generated language and say, ‘Wow, it kind of wrote a sentence,'" Mr. Clark said. "And now you're looking at stuff that's A.I.-generated and saying, ‘This is really funny, I'm enjoying reading this,' or ‘I had no idea this was even generated by A.I.'"

There is still plenty of bad, broken A.I. out there, from racist chatbots to faulty automated driving systems that result in crashes and injury. And even when A.I. improves quickly, it often takes a while to filter down into products and services that people actually use. An A.I. breakthrough at Google or OpenAI today doesn't mean that your Roomba will be able to write novels tomorrow.

But the best A.I. systems are now so capable — and improving at such fast rates — that the conversation in Silicon Valley is starting to shift. Fewer experts are confidently predicting that we have years or even decades to prepare for a wave of world-changing A.I.; many now believe that major changes are right around the corner, for better or worse.

Ajeya Cotra, a senior analyst with Open Philanthropy who studies A.I. risk, estimated two years ago that there was a 15 percent chance of "transformational A.I." — which she and others have defined as A.I. that is good enough to usher in large-scale economic and societal changes, such as eliminating most white-collar knowledge jobs — emerging by 2036.

But in a recent post, Ms. Cotra raised that to a 35 percent chance, citing the rapid improvement of systems like GPT-3.

"A.I. systems can go from adorable and useless toys to very powerful products in a surprisingly short period of time," Ms. Cotra told me. "People should take more seriously that A.I. could change things soon, and that could be really scary."

There are, to be fair, plenty of skeptics who say claims of A.I. progress are overblown. They'll tell you that A.I. is still nowhere close to becoming sentient, or replacing humans in a wide variety of jobs. They'll say that models like GPT-3 and LaMDA are just glorified parrots, blindly regurgitating their training data, and that we're still decades away from creating true A.G.I. — artificial general intelligence — that is capable of "thinking" for itself.

There are also tech optimists who believe that A.I. progress is accelerating, and who want it to accelerate faster. Speeding A.I.'s rate of improvement, they believe, will give us new tools to cure diseases, colonize space and avert ecological disaster.

I'm not asking you to take a side in this debate. All I'm saying is: You should be paying closer attention to the real, tangible developments that are fueling it.

After all, A.I. that works doesn't stay in a lab. It gets built into the social media apps we use every day, in the form of Facebook feed-ranking algorithms, YouTube recommendations and TikTok "For You" pages. It makes its way into weapons used by the military and software used by children in their classrooms. Banks use A.I. to determine who's eligible for loans, and police departments use it to investigate crimes.

Even if the skeptics are right, and A.I. doesn't achieve human-level sentience for many years, it's easy to see how systems like GPT-3, LaMDA and DALL-E 2 could become a powerful force in society. In a few years, the vast majority of the photos, videos and text we encounter on the internet could be A.I.-generated. Our online interactions could become stranger and more fraught, as we struggle to figure out which of our conversational partners are human and which are convincing bots. And tech-savvy propagandists could use the technology to churn out targeted misinformation on a vast scale, distorting the political process in ways we won't see coming.

It's a cliché, in the A.I. world, to say things like "we need to have a societal conversation about A.I. risk." There are already plenty of Davos panels, TED talks, think tanks and A.I. ethics committees out there, sketching out contingency plans for a dystopian future.

What's missing is a shared, value-neutral way of talking about what today's A.I. systems are actually capable of doing, and what specific risks and opportunities those capabilities present.

I think three things could help here.

First, regulators and politicians need to get up to speed.

Because of how new many of these A.I. systems are, few public officials have any firsthand experience with tools like GPT-3 or DALL-E 2, nor do they grasp how quickly progress is happening at the A.I. frontier.

We've seen a few efforts to close the gap — Stanford's Institute for Human-Centered Artificial Intelligence recently held a three-day "A.I. boot camp" for congressional staff members, for example — but we need more politicians and regulators to take an interest in the technology. (And I don't mean that they need to start stoking fears of an A.I. apocalypse, Andrew Yang-style. Even reading a book like Brian Christian's "The Alignment Problem" or understanding a few basic details about how a model like GPT-3 works would represent enormous progress.)

Otherwise, we could end up with a repeat of what happened with social media companies after the 2016 election — a collision of Silicon Valley power and Washington ignorance, which resulted in nothing but gridlock and testy hearings.

Second, big tech companies investing billions in A.I. development — the Googles, Metas and OpenAIs of the world — need to do a better job of explaining what they're working on, without sugarcoating or soft-pedaling the risks. Right now, many of the biggest A.I. models are developed behind closed doors, using private data sets and tested only by internal teams. When information about them is made public, it's often either watered down by corporate P.R. or buried in inscrutable scientific papers.

Downplaying A.I. risks to avoid backlash may be a smart short-term strategy, but tech companies won't survive long term if they're seen as having a hidden A.I. agenda that's at odds with the public interest. And if these companies won't open up voluntarily, A.I. engineers should go around their bosses and talk directly to policymakers and journalists themselves.

Third, the news media needs to do a better job of explaining A.I. progress to nonexperts. Too often, journalists — and I admit I've been a guilty party here — rely on outdated sci-fi shorthand to translate what's happening in A.I. to a general audience. We sometimes compare large language models to Skynet and HAL 9000, and flatten promising machine learning breakthroughs to panicky "The robots are coming!" headlines that we think will resonate with readers. Occasionally, we betray our ignorance by illustrating articles about software-based A.I. models with photos of hardware-based factory robots — an error that is as inexplicable as slapping a photo of a BMW on a story about bicycles.

In a broad sense, most people think about A.I. narrowly as it relates to us — Will it take my job? Is it better or worse than me at Skill X or Task Y? — rather than trying to understand all of the ways A.I. is evolving, and what that might mean for our future.

I'll do my part, by writing about A.I. in all its complexity and weirdness without resorting to hyperbole or Hollywood tropes. But we all need to start adjusting our mental models to make space for the new, incredible machines in our midst.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

 
Dit is hoe LLM's (large language models) ongeveer een half jaar geleden scoorden in vergelijking met hoe een mens gemiddeld scoort in diverse intelligentie testen :
 
Afbeelding
 
Vooral PaLM (Pathways Language Model) 540B scoorde al aardig dicht in de buurt van hoe een mens gemiddeld scoort. Het IQ gemiddeld over deze testen zou voor PaLM 540B dan ongeveer 95 bedragen. De intelligentie van het systeem begint die van een gemiddeld mens dus al aardig te naderen.
 
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
Gieljan de Vries 04-08-2022 09:00:00

Natuurkunde studeren vond ik fantastisch, maar de wiskunde die erbij kwam kijken? Afzien. Computerprogramma's hebben daar geen last van, blijkt uit een nieuwe publicatie in het wetenschapsblad PNAS. Computerwetenschapper Iddo Drori (MIT) en collega’s presenteren daarin OpenAI Codex, een kunstmatig intelligent systeem dat wiskundetentamenvragen op universitair niveau kan lezen en oplossen, en zelfs nieuwe vragen kan bedenken.

Zelfstandig programmeren
Zoals zo veel AI draait de kunstmatige wiskundestudent van Drodi op patroonherkenning. Zo gaven de onderzoekers OpenAI Codex bergen vraagstukken uit wiskundevakken bij MIT te lezen. Ook kreeg het programma de opdrachten in de programmeertaal Python te zien waarmee je vergelijkbare vragen oplost. Zo leert het welke tekst er bij een vraag hoort.

OpenAI Codex herkent niet alleen wat er gevraagd wordt in een tentamenvraag, in 81 procent van de gevallen kan het programma daarna ook zonder hulp de computercode schrijven om de vraag ook echt op te lossen. Het programma levert zelfs een keurige stap-voor-stap uitleg van zijn oplossing. Dat kan helpen de uitkomst te controleren, én laat studenten zien hoe ze een probleem zelf aan kunnen pakken.

Souffleren
Kunstmatige intelligentie-expert Aleksa Gordić van Google Deepmind vindt het niet enorm bijzonder dat OpenAI Codex zelf computerprogramma's schrijft om wiskundevragen te beantwoorden. De onderliggende taalherkenner GPT-3 kan namelijk al realistische antwoorden geven op vragen, of teksten aanvullen in dezelfde stijl.

"Uiteindelijk zijn programmeerinstructies gewoon taal. Je ziet hier wel hoe goed de computer die taal in de vingers heeft. Deze AI bouwt niet alleen computercode, die werkt en geeft ook antwoorden die echt kloppen." Knap werk, want in computercode telt elk detail – en wie weleens een chatbot tegenover zich heeft, weet dat die dingen weleens fouten maken.

Helemaal perfect is OpenAI Codex nog niet. In de overgebleven 19 procent van de gevallen moeten de begeleiders de computer namelijk nog iets helpen. Bijvoorbeeld als er verwijzingen naar films of wereldnieuws in een vraag staan die de computer niet snapt. Serieuzere hulp is ook nodig, bijvoorbeeld door te souffleren welke rekenmiddelen nodig zijn, of door te helpen een grafiek te tekenen in een ander programma.

Automatische tentamenmaker
OpenAI Codex lost niet alleen tentamenvragen op, het programma kan ze ook zelf in elkaar zetten. Door het een aantal vragen te voeren en die aan te laten vullen, rollen er vanzelf nieuwe vraagstukken uit de computer.

Die computergegenereerde vragen passen volgens een panel studenten inhoudelijk en qua moeilijkheidsgraad goed bij het desbetreffende wiskundevak. Ook konden ze de kunstmatige tentamenvragen maar lastig onderscheiden van vraagstukken die hun eigen docenten hadden opgesteld. Of OpenAI Codex zijn uitwerkingen al zo kan schrijven, dat docenten denken dat die van echte studenten komen, is niet onderzocht.

Bronnen: PNAS, YouTube
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Antonio Regalado, October 25, 2022

A little over 15 years ago, scientists at Kyoto University in Japan made a remarkable discovery. When they added just four proteins to a skin cell and waited about two weeks, some of the cells underwent an unexpected and astounding transformation: they became young again. They turned into stem cells almost identical to the kind found in a days-old embryo, just beginning life's journey.

At least in a petri dish, researchers using the procedure can take withered skin cells from a 101-year-old and rewind them so they act as if they'd never aged at all.

Now, after more than a decade of studying and tweaking so-called cellular reprogramming, a number of biotech companies and research labs say they have tantalizing hints the process could be the gateway to an unprecedented new technology for age reversal. By applying limited, controlled doses of the reprogramming proteins to lab animals, the scientists say, they are seeing evidence that the procedure makes the animals—or at least some of their organs—more youthful.

One of the key promoters of this idea, Richard Klausner, took the stage in June at a glitzy, $4,000-per-ticket retreat in San Diego, where he flashed data from unpublished experiments in which sick mice bounced back to health after undergoing the experimental treatment.

Klausner was pitching nothing less than "medical rejuvenation"—a means of taking old animals and making them "young." He is the organizer and chief scientist of Altos Labs, a new research company seeded with more than $3 billion from ultra-wealthy figures in Silicon Valley and oil money from the Persian Gulf. Klausner and his financiers had swept up dozens of top scientists—offering salaries of $1 million and more—and set them to work on a technology the company now calls "rejuvenation programming."

It seems to work at least in part by resetting what's called the epigenome—chemical marks on DNA that control which genes are turned on, or off, in a cell. In aging, some of these markers get flipped to the wrong positions. Reprogramming is a technology that can flip them back. But it can also change cells in dangerous ways, even causing cancer.

The objective of Altos is to tame this phenomenon, understand it, and eventually apply it as a treatment to reverse a wide range of diseases. This may be possible, Klausner says, because youthful cells have more resilience and can bounce back from biological stress in ways old ones don't. And Klausner has data to suggest it might already be working. During his talk, he showed slides marked "Confidential" claiming that fat mice had recovered from diabetes after treatment, and that others were able to survive normally lethal doses of painkillers—all thanks to a healthy dose of the medical rejuvenation.

"We think we can turn back the clock," he told the audience.

Klausner is the former head of the National Cancer Institute and onetime leader for global health at the Gates Foundation. He is a heavy hitter who has also been behind some of today's most high-profile biotech ventures, like the cancer blood-test company Grail. Yet even for him, rejuvenation is wildly ambitious. That is because if you can make cells act younger, healthier, and more resilient, you might have a general-purpose means of forestalling many diseases all at once. "This is the opposite of precision medicine," Klausner said.

Fountain of rejuvenation
To be sure, the word "rejuvenation" sounds suspicious, like a conquistador's quest or a promise made on a bottle of high-priced face cream. Yet rejuvenation is all around us, if you look. Millions of babies are born every year from the aging sperm and egg cells of their parents. Cloning of animals is another example. When Barbra Streisand had her 14-year-old dog cloned, cells from its mouth and stomach were returned to her as two frolicking puppies. These are all examples of cells being reprogrammed from age to youth—exactly the phenomenon companies like Altos want to capture, bottle, and one day sell.

For now, no one has a firm idea what these future treatments could look like. Some say they will be genetic therapies added to people's DNA; others expect it's possible to discover chemical pills that do the job. One proponent of the technology, David Sinclair, who runs an aging-research lab at Harvard University, says it could allow people to live much longer than they do today. "I predict one day it will be normal to go to a doctor and get a prescription for a medicine that will take you back a decade," Sinclair said at the same California event. "There is no reason we couldn't live 200 years."

It's this type of claim that raises so much skepticism. Critics see ballooning hype, runaway egos, and science that's on uncertain ground. But the doubters this year were drowned out by the sound of stampeding investors. In addition to Altos, whose $3 billion ranked as possibly the single largest startup fundraising drive in biotech history, the cryptocurrency billionaire Brian Armstrong, the cofounder of Coinbase, helped bring $105 million into his own reprogramming company, NewLimit, whose mission he says is "radical extension of human health span." Retro Biosciences, which says it wants to "increase healthy human lifespan by 10 years," raised $180 million.

These huge expenditures are being made despite the fact that scientists still disagree on the causes of aging. Indeed, there's no real consensus on when in life aging even begins. Some say it starts at conception, while others think it's at birth or after puberty.

But all the unknowns are part of what makes the reprogramming phenomenon so attractive. Klausner admits that the details of why reprogramming works remain a "complete mystery," but that too helps explain the sudden rush to invest in the idea. If there is a fountain of youth in the genome, the first to locate it could reinvent medicine and revolutionize how we treat the myriad of diseases that plague our old age.

Alchemy project
To get a reality check on Klausner's lecture, I asked an embryologist and stem-cell specialist, Alfonso Martinez Arias, to watch a recording. Martinez, whose lab is at the Pompeu Fabra University, in Barcelona, wrote back that he had to hold his stomach while he watched, so grandiose were the claims. "He was evangelical about something which, at the moment, is interesting but very preliminary and [on] shaky ground," says Martinez. Klausner was speaking "as if he had drunk some Kool-Aid."

Martinez says that to him, Altos is an alchemy project, the kind that medieval rulers once financed in the search for the philosopher's stone—a substance they believed could turn lead into gold, not to mention cure all disease. Martinez wasn't entirely negative, though. "There are people at Altos who know how to do science," he says. And, he notes, even alchemists ended up making valuable discoveries.

The basic technique Altos is exploring is the procedure discovered in 2006 by the Japanese scientist Shinya Yamanaka, who is now a scientific advisor to the company. The four proteins (now called "Yamanaka factors") that he and his students identified could cause ordinary cells to turn into potent stem cells, just like those found in embryos. This discovery earned him a Nobel Prize in medicine in 2012.

Initially, Yamanaka's discovery was employed to reprogram cells from patients to make stem cells, which could then be used to try to manufacture transplantable tissues, retina cells, or neurons. Other scientists wondered what would happen if they introduced Yamanaka's factors into living animals. In 2013, a Spanish team did exactly that, with gruesome results. The mice sprouted tumors called teratomas, blobs of renegade embryonic tissue.

The problem for these reprogrammed mice was that the process doesn't just make cells young; it also erases their identity and turns them into embryonic stem cells, which don't belong in an adult. Joe Betts-Lacroix, the CEO and founder of Retro, says researchers were soon asking a new question: "Is there some way that those two phenomena can be uncoupled so that you can have some of the age wiped away, but not have all your identity wiped away so that you become a pile of stem-cell protoplasm and die?"

In 2016, researchers at the Salk Institute in California, headed by Juan Carlos Izpisua Belmonte, reported that the answer might be yes. They genetically engineered mice afflicted with progeria, a condition that causes extremely rapid aging, so that all their cells would make the Yamanaka factors, but only when they were fed a special supplement in their food. That allowed the scientists to turn on the factors for a limited period—just a few hours at a time. Leave the genes on for too long, and the mice got cancer. But with shorter pulses—a tactic now known as partial reprogramming—they didn't. What's more, the mice seemed to become healthier and live a bit longer.

  • How it works
    1. Four proteins can "reprogram" a skin cell into a youthful stem cell.
    2. Apply these proteins to mice … but only in limited doses.
    3. Try to make the mice young, without causing cancer.
    4. Test the mice to see if they are healthier or live longer.

"You rejuvenate cells, but you didn't lose the identity," says Klausner, who calls it an "Aha!" moment. "That could be safe. And this has [now] been done with many animals. They don't get cancer as long as you don't go past this point."
Exactly how this partial-reprogramming phenomenon works is now a major focus of Altos and other research organizations. During a meeting held in June at a Maine ski resort, reprogramming scientists described studying individual cells by the tens of thousands—tracking in detail what changes they undergo after they're exposed to more limited pulses of the Yamanaka factors, or to subsets of them. Researchers from the United Kingdom with connections to Altos reported that they'd made skin cells from a 53-year-old person as youthful as those of someone just out of college. They claimed the "rejuvenation point" was reached after 13 days of exposure to Yamanaka's factors, but no more.

One way the British team concluded that the cells had become younger was by using an "aging clock." These are measurements that detect epigenetic modifications to DNA, the chemical marks that determine whether a given gene is on or shut off. (Epigenetic controls are part of what gives every cell its specialized identity; an olfactory neuron in your nose doesn't need the same genes activated as a liver cell that oozes bile.) Because these markers undergo telltale changes over a lifetime, it's possible to estimate a person's age, or that of any animal, within a couple of years by checking just two or three hundred of them.

In part because the clocks are eerily accurate, some researchers now believe aging may be caused primarily by the gradual degradation of the epigenetic code, a little like a compact disc that's been scratched and skips tracks. It's an attractive theory, and not least because one thing that reprogramming does reliably is reset these marks; after a little treatment with Yamanaka factors, a cell from a 90-year-old will have the epigenetic profile of one from a teenager.

To Klausner, the fact that cells can regain a youthful epigenetic state is remarkable and likely a gateway to important new biology. "Understanding how cells remember how to be an unscratched CD" could lead to the discovery of "missing codes" regulating the whole process of aging, he thinks.

Other scientists say it's an open question whether aging clocks measure true rejuvenation, a term they say is already being used too loosely. To Charles Brenner, a senior researcher at the City of Hope National Medical Center, people may even be falling victim to circular reasoning when they celebrate those epigenetic changes. "There isn't a difference between saying they applied the Yamanaka factors and that they have changed the epigenetic profile, since that is what the factors do," he says. "They then score their study as a rejuvenation success, but there is no scientific basis for doing that. They still don't know what the intervention does. People should not be assuming more youthful scores on an epigenetic clock equate to better health or longer life expectancy."

To answer that question, more researchers are applying bursts of the reprogramming factors to mice in bids to reverse specific diseases, or just to see what happens. In 2020, researchers at Harvard led by Sinclair reported that mice exposed to three reprogramming factors could regenerate their optic nerve and regain sight after it was crushed, something usually only a newborn rodent can do. That result earned them the cover of the journal Nature and the headline "Turning Back Time." Others have claimed that after partial reprogramming, mice perform better on a grip test (they're hung from tiny bars) and show signs of renewed muscle growth or even improved memory.

So far, many of these individual rejuvenation claims for live mice haven't been widely replicated by other labs, and some people are skeptical they ever will be. Measuring the relative health of animals or their tissues isn't necessarily a precise science. And in unblinded studies (where the researchers know which animals were treated), wishful thinking can play a role, perhaps especially if billions in venture capital dollars ride on the result. "Frankly, I doubt the reproducibility of these papers," says Hiro Nakauchi, a professor of genetics at Stanford University. Nakauchi says he also created mice with Yamanaka factors, but he never saw any sign they got younger. He suspects that some of the most dramatic claims are "timely and catchy" but that the science that went into them is "not very accurate."

One rejuvenation claim Brenner found troubling this year came from the Salk Institute, in La Jolla, California, which issued a press release saying a group of scientists there (who have since joined Altos) had been able to "safely and effectively reverse the aging process" in mice. It sounded as if they were describing a drug ready for market, not an exploratory form of genetic engineering. Izpisua Belmonte, the chief researcher involved, who now directs a San Diego research center for Altos, separately claimed he could "slow down aging" in the animals.

In reality, the results were less definitive than advertised. The researchers had not seen tumors, but they had significantly changed the epigenetic age of cells in just two organs: kidneys and skin. And something else about the result jumped out as puzzling to researchers like Brenner, as well as others who reviewed the paper. Despite saying they'd slowed aging, the Salk team didn't comment on how long the partially reprogrammed mice lived. Some data in their publication suggests that the rodents' life spans were unremarkable.

Indeed, so far no research group or company has reported normal mice living longer after being exposed to partial reprogramming. And that's something you might expect them to do, if the alchemy is real. To João Pedro de Magalhães, at the University of Birmingham, the gap in the data is puzzling, since he believes that whether the technology affects life span "is the billion-dollar question, so to speak." George Daley, a prominent stem-cell biologist who is dean of Harvard Medical School, wrote in response to the Salk paper that "rigorous demonstration of such an effect" was necessary to call reprogramming a true anti-aging intervention.

"Let's not pretend that the most important thing has happened if it hasn't," says Martin Borch Jensen, chief scientist at Gordian Biotechnology and founder of a grant-making organization. "I mean, is there any evidence for your $3 billion project?"

Disease reversal
When Altos officially launched, in January of 2022, Klausner and other executives strove to distance the company from the concept of life-span extension, even telling reporters that Altos "is not an aging or longevity company." They'd been stung by suggestions that the project existed to help billionaires cheat death. Instead, in its debut, Altos sought to align itself with a concept called "health span," which means extending the number of years people spend in good health.

Klausner says reprogramming promises an approach to "disease reversal" that might be applied regardless of how old someone is. If any extension in longevity resulted, it would be only "an accidental consequence" of making people healthier, according to comments made by Hans Bishop, the president of Altos.

Altos seeks to align itself with a concept called health span, which means extending the number of years that people spend in good health as they age.

In an email, Klausner even said that the company will not try to determine whether reprogramming generally extends life. "We have no intention of ever doing life-span extension studies," he wrote. He noted that an experiment would be impractical—such a test in humans could take too long. Instead, Altos hopes to carry out "very specific" attempts to reverse certain diseases or disabilities, using familiar frameworks for clinical trials that are accepted by regulators and attractive to large drug companies.

To some observers, like Magalhães, Altos is just trying to position anti-aging technology in a guise that's credible, even though some of the company's own scientists, like Izpisua Belmonte, have predicted that people will live to 130. "It is curious psychology," Magalhães says. "We say we are not trying to cure aging, just make people healthy longer. But I don't think we should be ashamed about what we are trying to do, which is to slow down aging. And rejuvenation, if we achieve it, would be the best way of doing that."

Klausner told me he thinks the longevity–versus–health span debate is "a distraction." The average American lives for around 77 years, which is still decades short of the longest lives (the oldest person on record died at 122). That means there are plenty of healthy years to be gained before anyone reaches an unnatural birthday. Nor are gains in average life expectancy unusual—that figure has roughly doubled since 1850, thanks mostly to vaccines, antibiotics, and public health advances.

"There is a lot of room for average life span to increase," Klausner says, "and that is essentially the goal of all medicine, whether curing cancer or heart disease."
 
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Mahalingam
Berichten: 52154
Lid geworden op: za feb 24, 2007 8:39 pm

Re: Futuristische ontwikkelingen

Bericht door Mahalingam »


De 3 roboticawetten van Asimov


Artificial Intelligence en robots die opereren op AI zijn in opmars en de komende jaren zullen we ze steeds meer in het dagelijks leven gaan tegenkomen.

Dit roept natuurlijk ook ethische vragen op. Als robots zelfstandig in het openbare leven opereren, hoe zorgen we dan dat ze mensen geen kwaad doen? Wie is er verantwoordelijk als dat wel gebeurt? Etc.

Een veel geopperde set met regels om ellende te voorkomen is om robots te voorzien van de 3 wetten van Isaac Asimov (1920-1992). Isaac Asimov is een schrijver van science fiction boeken (zijn bekendste boeken zijn I, Robot en Do androids dream of electric sheep?). In zijn boeken spelen robots een grote rol en daarbij heeft de mensheid vastgelegd dat robots zich aan 3 regels moeten houden:

Eerste Wet:
Een robot mag een mens geen letsel toebrengen of door niet te handelen toestaan dat een mens letsel oploopt.

Tweede Wet:

Een robot moet de bevelen uitvoeren die hem door mensen gegeven worden, behalve als die opdrachten in strijd zijn met de Eerste Wet.

Derde Wet:

Een robot moet zijn eigen bestaan beschermen, voor zover die bescherming niet in strijd is met de Eerste of Tweede Wet.

Wat later is daar nog de "nulde wet aan toegevoegd die nog voor de eerste gaat:

Nulde wet:
Een robot mag geen schade toebrengen aan de mensheid, of toelaten dat de mensheid schade toegebracht wordt door zijn nalatigheid.
https://informatica.sgdb.nl/index.php/n ... van-asimov
Wie in de Islam zijn hersens gebruikt, zal zijn hoofd moeten missen.
Mahalingam
Berichten: 52154
Lid geworden op: za feb 24, 2007 8:39 pm

Re: Futuristische ontwikkelingen

Bericht door Mahalingam »

En zo iets als bovenstaand is natuurlijk veel te simpel voor de Brusselse bureaucraten.
Dan krijg je dus dit:
EU-ministers zijn akkoord met wet voor reguleren van kunstmatige intelligentie

De telecomministers van de EU-lidstaten hebben een akkoord bereikt over de Artificial Intelligence Act, een wetsvoorstel met als doel om te verzekeren dat AI-systemen in de EU veilig zijn en de fundamentele vrijheden en Unie-waarden eerbiedigen.


De Raad schrijft dat het voorstel uitgaat van een risicogebaseerde aanpak. Dat betekent dat het niet zozeer ertoe doet of het gaat om een complex machinelearningsysteem of een simpel filter in Excel, maar dat er gekeken wordt waar er risico's voor mensen ontstaan. Er wordt dus niet geregeld hoe iets moet werken, als de werking maar veilig is. Het voorstel creëert een uniform raamwerk voor AI waarmee er juridische duidelijkheid komt, wat ook de innovatie en investeringen moet bespoedigen. Daarnaast stelt de Raad dat er ook duidelijk aandacht is voor het toezicht en de handhaving van bestaande regels over fundamentele rechten en veiligheid.

Het voorstel bevat onder meer een definitie van een AI-systeem en gaat uit van een drietal niveaus. Eerst is er het hoogste niveau van AI met een onaanvaardbaar risico. Dat zijn systemen die inbreuk maken op de fundamentele rechten, waarbij gedacht kan worden aan socialcreditscoring of systemen die kwetsbaarheden van bepaalde groepen of personen uitbuiten. Eerder noemde de Commissie ook al het voorbeeld van speelgoed dat via spraakassistentie gevaarlijk gedrag van minderjarigen kan aanmoedigen. Het tweede niveau betreft AI met een hoog risico als er sprake zou zijn van ongecontroleerde inzet. Voorbeelden zijn biometrie in de openbare ruimte, beheer van de infrastructuur, selectie en werving van personeel, rechtshandhaving of grenscontroles. Systemen die daarvoor bedoeld zijn, zijn op grond van het voorstel alleen toegestaan onder strenge voorwaarden. De derde categorie behelst AI met een beperkt risico.

Ook het Nederlandse ministerie van Economische Zaken stelt dat de Europese verordening mogelijkheden en economische kansen biedt. De AI Act zorgt bijvoorbeeld voor duidelijkheid voor ontwikkelaars als ze hun AI-producten op de markt willen brengen, is de gedachte. Ook benadrukt het ministerie dat de nieuwe regels een gelijk speelveld creëren, omdat ook niet-Europese bedrijven en aanbieders die hun met AI gerelateerde diensten en producten in Europa willen uitbrengen, aan dezelfde regels moeten voldoen. Verder wordt genoemd dat als organisaties en kleinere bedrijven kunstmatige intelligentie willen gebruiken, zij erop kunnen vertrouwen dat het goed werkt en dat er daarbij ondersteuning van de ontwikkelaar aanwezig is; voor het mkb is daartoe een specifieke lijst met bepalingen opgesteld.

Het wetsvoorstel werd in april vorig jaar ingediend door de Europese Commissie. Nu de Raad er akkoord mee is, begint de fase van onderhandelingen tussen het Europees Parlement en de Commissie. Het streven is dat in de herfst van volgend jaar een definitieve overeenstemming is bereikt, waarna de verordening in de lidstaten moet worden doorgevoerd. In principe werkt een verordening rechtstreeks door in de nationale rechtsorders van lidstaten, maar er zullen waarschijnlijk nationale wetten moeten worden aangepast ten gunste van de AI Act.
https://tweakers.net/nieuws/204244/eu-m ... entie.html
Wie in de Islam zijn hersens gebruikt, zal zijn hoofd moeten missen.
Mahalingam
Berichten: 52154
Lid geworden op: za feb 24, 2007 8:39 pm

Re: Futuristische ontwikkelingen

Bericht door Mahalingam »

En als we die AI-wet eerder hadden gehad zou dit dan niet zijn voorgekomen?
Parkeren met autopilot gaat flink mis in Arnhem

ARNHEM – Op de Annastraat en de Graaf Lodewijkstraat in Arnhem is woensdagmiddag een ongeluk gebeurd. Hierbij raakte meerdere auto’s beschadigd.

Afbeelding

De hulpdiensten werden rond 15:00 uur opgeroepen. Het ongeluk gebeurde toen de bestuurder zijn voertuig wilde parkeren op de Annastraat met de ‘automatische piloot’ van de auto. Dit keer ging het mis en sloeg volgens hem de auto op hol. Hierbij raakte in totaal vijf andere geparkeerde auto’s beschadigd. De auto kwam uiteindelijk tegen een boom tot stilstand op de Graaf Lodewijkstraat, waarna de boom in twee is gebroken. Één deel van de boom is tegen een voordeur van een huis terecht gekomen.

De bestuurder en zijn vrouw kwamen met de schrik vrij. Hij noemt het een horror droom die uit is gekomen, maar is opgelucht dat hij samen met zijn vrouw er heelhuids is uitgekomen.
https://www.gelrenieuws.nl/2022/12/park ... rnhem.html
Wie in de Islam zijn hersens gebruikt, zal zijn hoofd moeten missen.
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

 
Mensen slaan ook wel eens op hol (denk bijvoorbeeld aan "feestende" Marokkanen), dus waarom zouden machines niet op hol kunnen slaan?
 
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Gebruikersavatar
xplosive
Berichten: 8906
Lid geworden op: do jun 30, 2011 11:18 pm

Re: Futuristische ontwikkelingen

Bericht door xplosive »

   
By Calum Chace, Oct 19, 2022,02:30pm EDT

There is a paradox in artificial intelligence (AI). The technology is already very powerful, and most people agree that it will transform every industry and every aspect of our lives. But deployment of AI in industry seems to be proceeding slower than expected. One explanation for this is that CEOs and CTOs are understandably nervous about deploying systems that are unpredictable. They are even more nervous about systems which make mistakes, but unless they make mistakes they cannot learn.

Fast progress in AI research
Whatever the hold-ups in industry, AI is making great strides in the lab. Researchers are surprised, and some of them are becoming concerned. Simon Thorpe is one researcher who believes that AI is progressing much faster than almost anyone recognises, and that much more attention should be paid to it. He has spent 40 years researching how the human brain works, and what we can learn about that from computers. For much of that time he has been a director at France’s CNRS, the Centre Nationale pour Recherche Scientifique (National Centre for Scientific Research). He was the guest on the latest episode of the London Futurist Podcast.

Simon has some most eyebrow-raising predictions about how fast AI is advancing, but before we can get to that, we need to look at his ideas about why human brains are so much more energy-efficient than machines.

The human visual system
The visual perception system in the human brain is basically a feed-forward system, meaning that it can make high level decisions without the need to use feedback. This enables it to be fast and efficient, so we can recognise important images – like the faces of family members – very fast, and even when there is noise in the signal, like an out-of-focus photo. This feature of our visual perception system has drawbacks: when we are looking out for something specific we are very likely to miss something else which is even more important. Simon observes that security guards looking for guns are surprisingly prone to failing to notice the presence of a hand grenade. This also explains how conjuring tricks work.

It is a form of bias known as "inattentional blindness", and a well-known demonstration can be found in this video of a game of basketball. If you haven’t seen this before, enjoy. I can almost guarantee it will be the most surprising thing you see today.

Energy-efficient brains
Human brains are currently far more energy-efficient than AI systems. The human brain consumes around 20 watts, about the same as a light bulb. GPT-3 and other large natural language processing models use thousands of times more.

Indeed, if you wanted to simulate the 86 billion neurons in the human brain using the sort of model used in the current range of Deep Learning trained neural networks you would need an enormous amount of computation - around 500 petaflops, or 500,000,000,000,000,000 floating-point operations per second. This is half an exaflop, and the largest supercomputers in the world have only just reached that scale. Those supercomputers use around 30 Megawatts of power – over 1 million time more energy than the brain.

Sparse and spiking
Simon is convinced that to make machines as efficient as brains they need to employ "sparse networks". The basic idea here is that when humans recognise an image, only the neurons which are trained to expect the components of that image need to fire. In machines, by contrast, the state of every neuron needs to be taken into account for every computation.

Furthermore, machines also need to adopt a spiking model. Artificial neurons have continuously varying activation values, encoded using floating point numbers that are computationally very expensive. By contrast, neurons send information using electrical pulses or spikes that can be very sparse. When you have lots of neurons, the order in which they fire can convey information very efficiently, and with very few spikes. Simon argues that the great majority of AI researchers are simply ignoring this crucial fact, and that as a result, they are going to be very surprised by the speed of some imminent developments.

Terabrain
Simon is impressed with the power of Apple's latest proprietary chips, found in its most recent laptop computers. He thinks that by employing the ideas described above, he can design AI systems to run on these computers that have billions of neurons and hundreds of billions of connections. He calls this his Terabrain project, and he plans to open source these designs, making them freely available to researchers everywhere. Remarkably, he believes that using such designs, it may be possible to create something similar to artificial general intelligence (AGI) before the end of 2023. Superintelligence, he says, may not be far behind. If he's right, the world is about to change completely.
Gun jezelf wat je een ander toewenst     islam = racisme   & de hel op aarde voor mens en dier
                                   koran = racistisch & handboek voor criminelen
      Moslimlanden bewijzen dagelijks:    meer islam = meer verkrachte mensenrechten
Plaats reactie