Here is the daily technology #threadcast for 11/3/24. The goal is to make this a technology "reddit".
Drop all question, comments, and articles relating to #technology and the future. The goal is make it a technology center.
Here is the daily technology #threadcast for 11/3/24. The goal is to make this a technology "reddit".
Drop all question, comments, and articles relating to #technology and the future. The goal is make it a technology center.
A wind turbine collapsed in Härjedalen, Sweden, between Högvålen and Tännäs, due to strong winds. No injuries were reported, but emergency services were dispatched over concerns about potential oil leakage from the turbine, which contains hydraulic oil. The risk was mitigated after finding a smaller amount of oil than expected. Tännäs Vind, the operator, stated that they will investigate why the structure failed to withstand the wind.
https://www.svt.se/nyheter/lokalt/jamtland/vindkraftverk-har-blast-ner-i-harjedalen
The Guardian: The chatbot optimisation game: can we trust AI web searches?
https://www.theguardian.com/technology/2024/nov/03/the-chatbot-optimisation-game-can-we-trust-ai-web-searches
The Guardian: Microsoft workers fired over Gaza vigil say company ‘crumbled under pressure’
https://www.theguardian.com/us-news/2024/nov/02/microsoft-workers-fired-gaza-vigil
The Guardian: Addicted to love: how dating apps ‘exploit’ their users
https://www.theguardian.com/lifeandstyle/2024/nov/03/addicted-to-love-how-dating-apps-exploit-their-users
The Guardian: From Northern trains to Dolly Parton, the institutions that can’t quit the fax
https://www.theguardian.com/technology/2024/nov/01/northern-rail-dolly-parton-fax-machines-trains-technology
The Guardian: ‘What a privilege … ’ trend catches on as gratitude makes social media comeback
https://www.theguardian.com/media/2024/nov/02/what-a-privilege-trend-catches-on-as-gratitude-makes-social-media-comeback
India is the first country to approach AI with a national strategy and is taking a different view of how AI should be used. The intention they have is to involve all of society, making it work for the many and not just the few
The Union for swedish Engineers, Sveriges Ingenjörer, is dissapointed with the decision made at Ericsson forcing people back to the office.
"Wrong focus and in-the-box thinink are leading them to the wrong conclusions."
https://ingenjoren.se/2024/10/22/ericsson-kraver-tre-kontorsdagar-i-veckan-akademikerna-fyrkantigt-och-fel-fokus/
Question for the tech audience
Could there be a usecase for LeoAI to run around the threads daily and engage with us?
My thinking here is that LeoAI needs more data, and by engaging with the people here we might increase the data big time. Or is this not productive data that would be added?
Just think, it could go through here and add some simple questions one could be curious about on the different links added.
That is certainly productive.
What happens there is two fold:
Maybe @khaleelkazi already thought about this or should I send a feedback notice to him?
Was just thinking about this because I did just this as a test on a article and that worked well.
I am not following.
Back to the point, the key is, as always, data. The more that is generated, the better LeoAI will function. Obviously, human content is required. However, synthetic days is important.
When Leo's chatbot rolls out, we have to be generating at least 10 million tokens per day.
Well Im looking into if I could build something that could use like chatgpt or something to generate a answer until we have leoai.
@mightpossibly could be the one to ask that to. He seems to know a great deal about it.
But there are ways to do things like that.
So basically, @anderssinho, a bot that automatically
That's a cool idea! And shouldn't be neither too complicated to create or costly to run. The llm cost would actually probably be so cheap that even if the bot posted a thousand questions a day, it probably wouldn't spend more than a few cents worth of tokens (a small model would suffice plenty for this, like gpt-4o-mini or claude-3.5-haiku)
I'd be happy to help you get started if you want to try and build it yourself.
Actually, I might look into if I can build something to trial this :)
Machines might not match humans just yet, but sometimes “good enough” is all that’s really needed.
#lovingAIrevolurion
That is so true. I find it funny (and tragic) how so many people are hellbent on pointing out the flaws and limitations, while being completely oblivious to the possibilities
I think who try to do that are just intimidated by the potential it holds. So in a bit to preserve their own relevance do try to highlight the flaws and weaknesses.
The mindset should be working with it instead of fighting it.
As a matter of fact the revolution of AI cannot be stopped, I believe it's either they adopt or get sidetracked (in a polite way).
I think you're definitely right about that. I'm by no means an AI "evangelist", even if most people in my workplace probably see me like one. My point tends to be, like you also touch on, that it's here whether they like it or not. And whether they think it's moral that it's being developed or not, the progress and power of it will keep accelerating.
Get on the train or get left behind at the station. That's the two options you have, really.
I'm glad our thoughts resonate friend. Good morning to you. I hope your day started off and a good note and so going on well?
gm to you too and likewise. Yep, it's going pretty well! I managed to code a working bot that plays Terracore for me (a Hive-based idle-game) yesterday, from scratch, using AI (since I don't know how to code). Super excited about getting that to work.
let me show this to my brother who is into coding and let him think twice about his career. 😂
Hi, @mightpossibly,
This post has been voted on by @darkcloaks because you are an active member of the Darkcloaks gaming community.
Get started with Darkcloaks today, and follow us on Inleo for the latest updates.
Reuters: China's sanction-hit tech industry puzzles over impact of Trump, Harris presidencies
https://www.reuters.com/technology/chinas-sanction-hit-tech-industry-puzzles-over-impact-trump-harris-presidencies-2024-11-01/
With meme coins, I see no utility just hype, invest and dump. Do you think this concept will make it die off in the future or will it evolve into something else?
I dont see much value in them as portrayed now. Social coins do have value but it has to be around something.
oh this is actually the first time I'm hearing that social coins have value. Maybe I'll start looking into it and see what I can dig up thanks Taskmaster
What I mean by social coins is the social aspect of fandom.
Take for example Taylor Swift....how much value would a Taylor coin have. Or a disney coins. Or star Wars.
Tremendous value, I get it now Taskmaster 💯💯💯💯💯💯💯💯. Then yes social coins really do have value
That also means meme coins may never die, ever
A lot of nonsense. Some might end up having a following that sticks. Most will end up worthless like most NFTs in the past.
yes that's what I'm already seeing man, it's a lot of nonsense
Hi, @taskmaster4450le,
This post has been voted on by @darkcloaks because you are an active member of the Darkcloaks gaming community.
Get started with Darkcloaks today, and follow us on Inleo for the latest updates.
Reuters: U.S. regulators raise questions about siting data centers at power plants
https://www.reuters.com/world/us/us-regulators-mull-issues-around-siting-data-centers-power-plants-2024-11-01/
Reuters: US regulators reject amended interconnect agreement for Amazon data center
https://www.reuters.com/business/energy/us-regulators-reject-amended-interconnect-agreement-amazon-data-center-2024-11-02/
Reuters: France steps up efforts at examining how to protect struggling IT company Atos
https://www.reuters.com/technology/france-steps-up-efforts-examining-how-protect-struggling-it-company-atos-2024-11-01/
BBC: Snapchat most-used app for grooming, says NSPCC
https://www.bbc.com/news/articles/cze3p1j710ko
Reuters: Siemens can make more acquisitions after Altair deal, exec tells paper
https://www.reuters.com/markets/deals/siemens-can-make-more-acquisitions-after-altair-deal-exec-tells-paper-2024-11-01/
BBC: An eco-friendly plant-based alternative to plastic
https://www.bbc.com/reel/video/p0k13dxh/an-eco-friendly-plant-based-alternative-to-plastic
I hope one of these will take of one day. Have seen several companies try to attack this area.
Reuters: Exclusive: Walt Disney forms business unit to coordinate use of AI, augmented reality
https://www.reuters.com/technology/artificial-intelligence/walt-disney-forms-business-unit-coordinate-use-ai-augmented-reality-2024-11-01/
Reuters: A year on, Intel's touted AI-chip deals have fallen short
https://www.reuters.com/technology/artificial-intelligence/year-intels-touted-ai-chip-deals-have-fallen-short-2024-11-01/
And to not to be to serious:
10 recomendações dos neuropediatras para o uso de tecnologia em crianças
Como exemplo de outras sociedades científicas que estabeleceram orientações e limites para o uso de ecrãs - como a Academia Americana de Pediatria (AAP), a Organização Mundial de Saúde (OMS), a Sociedade Canadiana de Pediatria (CPS) e o Departamento de Saúde Australiana (AHD) - a Sociedade Portuguesa de Neuropediatria estabeleceu 10 recomendações "após uma extensa revisão bibliográfica e adaptando os resultados ao contexto social e cultural nacional".
#technology #science #portugal
Evitar ecrãs até aos 3 anos, exceto para videochamadas; neste contexto, exclui-se a televisão que pode ser usada até 30 minutos diários, desde que na presença de um adulto e com conteúdo adequado;
Entre os 4 e 6 anos, limitar o uso de ecrãs a 30 minutos por dia de programação de alta qualidade, assistida na presença de adultos que ajudem a contextualizar o conteúdo; o controlo de canais/mudança de vídeos não deve ser acessível à criança de forma autónoma;
Manter limites consistentes de utilização, dos 7 aos 11 anos até 1 hora por dia e, dos 12 aos 15 anos até 2 horas por dia, garantindo que o seu uso não interfira no sono, na interação social, na atividade física, no estudo e em outras atividades essenciais para a saúde e bem-estar;
Como se prevê na legislação portuguesa, as redes sociais só devem permitidas a partir dos 16 anos;
Em todas as idades, os ecrãs não devem ser utilizados como forma de facilitar as refeições, ultrapassar momentos de espera nem para controlar birras;
Deve promover-se o uso de conteúdo educativo e de alta qualidade, utilizando estes recursos em conjunto com as crianças para ajudar a entender o que estão a ver e a aplicar o que aprenderam;
Devem ser estabelecidos horários e áreas da casa onde o uso de dispositivos móveis não é permitido, como a mesa de refeições e os quartos;
Os pais devem modelar comportamentos saudáveis em relação ao uso de tecnologia, demonstrando equilíbrio entre o uso de dispositivos e outras atividades;
Incentivar atividades físicas diárias e oportunidades para interações sociais presenciais;
Devem ser ponderadas restrições à utilização de ecrãs nos intervalos escolares, promovendo o convívio e atividade física que substituam esta utilização extensiva; igualmente, o uso de manuais digitais e de plataformas online para trabalhos de casa devem ser repensadas até ao final do terceiro ciclo, dado não existir qualquer evidência de que a sua utilização seja vantajosa para os processos de aprendizagem, sendo muitas vezes fonte de distração e dispersão para outros conteúdos não letivos.
De acordo com o documento, evidência científica que já temos disponível hoje aponta para diferentes e preocupantes danos do uso indevido de ecrãs e tecnologia digital ao longo das várias etapas da idade pediátrica.
Os autores da pesquisa ressaltam a importância da interação com o meio para a maturação do sistema nervoso nos bebés, e como a utilização precoce de ecrãs condiciona obstáculos como mais tempo de atividades sedentárias, dificuldade de focar a atenção e de gerir adversidades e enfrentar momentos de tédio, o que interfere com o desenvolvimento da criatividade.
A redução do tempo de interação com adultos e outras crianças, por sua vez, amplia o risco de comportamentos sociais evitáveis e atraso no desenvolvimento de linguagem e empobrecimento lexical. A qualidade do sono também pode ser fortemente afetada.
BBC: Swapped at birth: How two women discovered they weren’t who they thought they were
https://www.bbc.com/news/articles/cp3njqd9nl9o
MIT robot combines vision and touch to learn the game of Jenga
In the basement of MIT’s Building 3, a robot is carefully contemplating its next move. It gently pokes at a tower of blocks, looking for the best block to extract without toppling the tower, in a solitary, slow-moving, yet surprisingly agile game of Jenga.
#technology #robot #mit
The robot, developed by MIT engineers, is equipped with a soft-pronged gripper, a force-sensing wrist cuff, and an external camera, all of which it uses to see and feel the tower and its individual blocks.
As the robot carefully pushes against a block, a computer takes in visual and tactile feedback from its camera and cuff, and compares these measurements to moves that the robot previously made. It also considers the outcomes of those moves — specifically, whether a block, in a certain configuration and pushed with a certain amount of force, was successfully extracted or not. In real-time, the robot then “learns” whether to keep pushing or move to a new block, in order to keep the tower from falling.
Details of the Jenga-playing robot are published today in the journal Science Robotics. Alberto Rodriguez, the Walter Henry Gale Career Development Assistant Professor in the Department of Mechanical Engineering at MIT, says the robot demonstrates something that’s been tricky to attain in previous systems: the ability to quickly learn the best way to carry out a task, not just from visual cues, as it is commonly studied today, but also from tactile, physical interactions.
“Unlike in more purely cognitive tasks or games such as chess or Go, playing the game of Jenga also requires mastery of physical skills such as probing, pushing, pulling, placing, and aligning pieces. It requires interactive perception and manipulation, where you have to go and touch the tower to learn how and when to move blocks,” Rodriguez says. “This is very difficult to simulate, so the robot has to learn in the real world, by interacting with the real Jenga tower. The key challenge is to learn from a relatively small number of experiments by exploiting common sense about objects and physics.”
He says the tactile learning system the researchers have developed can be used in applications beyond Jenga, especially in tasks that need careful physical interaction, including separating recyclable objects from landfill trash and assembling consumer products.
“In a cellphone assembly line, in almost every single step, the feeling of a snap-fit, or a threaded screw, is coming from force and touch rather than vision,” Rodriguez says. “Learning models for those actions is prime real-estate for this kind of technology.”
The paper’s lead author is MIT graduate student Nima Fazeli. The team also includes Miquel Oller, Jiajun Wu, Zheng Wu, and Joshua Tenenbaum, professor of brain and cognitive sciences at MIT.
In the game of Jenga — Swahili for “build” — 54 rectangular blocks are stacked in 18 layers of three blocks each, with the blocks in each layer oriented perpendicular to the blocks below. The aim of the game is to carefully extract a block and place it at the top of the tower, thus building a new level, without toppling the entire structure.
To program a robot to play Jenga, traditional machine-learning schemes might require capturing everything that could possibly happen between a block, the robot, and the tower — an expensive computational task requiring data from thousands if not tens of thousands of block-extraction attempts.
Instead, Rodriguez and his colleagues looked for a more data-efficient way for a robot to learn to play Jenga, inspired by human cognition and the way we ourselves might approach the game.
The team customized an industry-standard ABB IRB 120 robotic arm, then set up a Jenga tower within the robot’s reach, and began a training period in which the robot first chose a random block and a location on the block against which to push. It then exerted a small amount of force in an attempt to push the block out of the tower.
For each block attempt, a computer recorded the associated visual and force measurements, and labeled whether each attempt was a success.
Rather than carry out tens of thousands of such attempts (which would involve reconstructing the tower almost as many times), the robot trained on just about 300, with attempts of similar measurements and outcomes grouped in clusters representing certain block behaviors. For instance, one cluster of data might represent attempts on a block that was hard to move, versus one that was easier to move, or that toppled the tower when moved. For each data cluster, the robot developed a simple model to predict a block’s behavior given its current visual and tactile measurements.
Fazeli says this clustering technique dramatically increases the efficiency with which the robot can learn to play the game, and is inspired by the natural way in which humans cluster similar behavior: “The robot builds clusters and then learns models for each of these clusters, instead of learning a model that captures absolutely everything that could happen.”
The researchers tested their approach against other state-of-the-art machine learning algorithms, in a computer simulation of the game using the simulator MuJoCo. The lessons learned in the simulator informed the researchers of the way the robot would learn in the real world.
“We provide to these algorithms the same information our system gets, to see how they learn to play Jenga at a similar level,” Oller says. “Compared with our approach, these algorithms need to explore orders of magnitude more towers to learn the game.”
Curious as to how their machine-learning approach stacks up against actual human players, the team carried out a few informal trials with several volunteers.
“We saw how many blocks a human was able to extract before the tower fell, and the difference was not that much,” Oller says.
But there is still a way to go if the researchers want to competitively pit their robot against a human player. In addition to physical interactions, Jenga requires strategy, such as extracting just the right block that will make it difficult for an opponent to pull out the next block without toppling the tower.
For now, the team is less interested in developing a robotic Jenga champion, and more focused on applying the robot’s new skills to other application domains.
“There are many tasks that we do with our hands where the feeling of doing it ‘the right way’ comes in the language of forces and tactile cues,” Rodriguez says. “For tasks like these, a similar approach to ours could figure it out.”
This research was supported, in part, by the National Science Foundation through the National Robotics Initiative.
Stirling engine used for residential energy generation.
While the technology is not financially efficient compared to modern heating solutions, it significantly enhances quality of life by supporting self-sufficiency and sustainability. The homeowner values the system for its innovation and environmental benefits rather than cost savings. The piece highlights the niche appeal of the Stirling engine, appealing to those prioritizing eco-conscious living over economic gain.
https://www.nyteknik.se/energi/stirlingmotor-i-pannrummet-ingen-ekonomi-men-livskvalitet/4302621
AI can design so maybe can tell the engineers exactly what to do step by step to create this, but how long till we get to this level of engineering
A new model offers robots precise pick-and-place solutions
Pick-and-place machines are a type of automated equipment used to place objects into structured, organized locations. These machines are used for a variety of applications — from electronics assembly to packaging, bin picking, and even inspection — but many current pick-and-place solutions are limited. Current solutions lack “precise generalization,” or the ability to solve many tasks without compromising on accuracy.
#technology #robot
“In industry, you often see that [manufacturers] end up with very tailored solutions to the particular problem that they have, so a lot of engineering and not so much flexibility in terms of the solution,” Maria Bauza Villalonga PhD ’22, a senior research scientist at Google DeepMind where she works on robotics and robotic manipulation. “SimPLE solves this problem and provides a solution to pick-and-place that is flexible and still provides the needed precision.”
A new paper by MechE researchers published in the journal Science Robotics explores pick-and-place solutions with more precision. In precise pick-and-place, also known as kitting, the robot transforms an unstructured arrangement of objects into an organized arrangement. The approach, dubbed SimPLE (Simulation to Pick Localize and placE), learns to pick, regrasp and place objects using the object’s computer-aided design (CAD) model, and all without any prior experience or encounters with the specific objects.
“The promise of SimPLE is that we can solve many different tasks with the same hardware and software using simulation to learn models that adapt to each specific task,” says Alberto Rodriguez, an MIT visiting scientist who is a former member of the MechE faculty and now associate director of manipulation research for Boston Dynamics. SimPLE was developed by members of the Manipulation and Mechanisms Lab at MIT (MCube) under Rodriguez’ direction.
“In this work we show that it is possible to achieve the levels of positional accuracy that are required for many industrial pick and place tasks without any other specialization,” Rodriguez says.
Using a dual-arm robot equipped with visuotactile sensing, the SimPLE solution employs three main components: task-aware grasping, perception by sight and touch (visuotactile perception), and regrasp planning. Real observations are matched against a set of simulated observations through supervised learning so that a distribution of likely object poses can be estimated, and placement accomplished.
“There’s an intuitive understanding in the robotics community that vision and touch are both useful, but [until now] there haven’t been many systematic demonstrations of how it can be useful for complex robotics tasks,” says mechanical engineering doctoral student Antonia Delores Bronars SM ’22. Bronars, who is now working with Pulkit Agrawal, assistant professor in the department of Electrical Engineering and Computer Science (EECS), is continuing her PhD work investigating the incorporation of tactile capabilities into robotic systems.
“Most work on grasping ignores the downstream tasks,” says Matt Mason, chief scientist at Berkshire Grey and professor emeritus at Carnegie Mellon University who was not involved in the work. “This paper goes beyond the desire to mimic humans, and shows from a strictly functional viewpoint the utility of combining tactile sensing, vision, with two hands.”
Ken Goldberg, the William S. Floyd Jr. Distinguished Chair in Engineering at the University of California at Berkeley, who was also not involved in the study, says the robot manipulation methodology described in the paper offers a valuable alternative to the trend toward AI and machine learning methods.
“The authors combine well-founded geometric algorithms that can reliably achieve high-precision for a specific set of object shapes and demonstrate that this combination can significantly improve performance over AI methods,” says Goldberg, who is also co-founder and chief scientist for Ambi Robotics and Jacobi Robotics. “This can be immediately useful in industry and is an excellent example of what I call 'good old fashioned engineering' (GOFE).”
Bauza and Bronars say this work was informed by several generations of collaboration.
“In order to really demonstrate how vision and touch can be useful together, it’s necessary to build a full robotic system, which is something that’s very difficult to do as one person over a short horizon of time,” says Bronars. “Collaboration, with each other and with Nikhil [Chavan-Dafle PhD ‘20] and Yifan [Hou PhD ’21 CMU], and across many generations and labs really allowed us to build an end-to-end system.”
What happens when a lot of jobs get automated? Most jobs, especially those that pay well, exist for a reason. Companies pay for labor because they need it, and the more special or necessary your skills are, the more you can earn.
But if AI can do your job better, companies might decide they don’t need you anymore. That raises a serious question: if you’re no longer needed, how do you make a living? Who’s going to pay you then?
That is the big question that gets ignored because people think there will always be more jobs created.
It isnt the case. I just posted that a Polish radio station eliminated the on air hosts and replaced them with AI. This is only going to keep growing.
Oh boy, the revolution has began
Do you trust Current AI models with Facts?
#askleo
Depends upon the model.
It is why we have to get data like that into LeoAI.
So posted things such as dates, stats, prices of stocks, etc... is crucial for the output when people query for facts.
oh I get that now.
What we input here on inleo will determine how dependable the LeoAi would be.
How will it handle people putting subjective opinions here, so it's basically a platform for that
#askleo #inleo
Well opinions vary. There are other opinions put in that counter that. There is also the training of the underlying LLM model.
So if someone says that Howard Stern was president of the United States (a fact that is incorrect) the numeric correlation will overwhelm is as number of occurrences will reveal something else.
Oh by this you mean the AI would lean more to the number of times a fact was stated to be something.
In other words if the masses get it right the AI would and vice versa?
#askleo #leoai
Yes. When a vector database is constructed, closeness and frequency are important variables to what the model gives weight to.
That is why it is important to get a lot of the same topics covered in different ways. All of that starts to establish connections throughout the database.
oooooh makes a lot of sense now.
The US Government created a vacum when they failed to embrace the new technology called the digital ledger and tether moved into the vacum.
https://inleo.io/@shortsegments/tether-has-become-the-worlds-digital-dollar-bq5?referral=shortsegments
AI has evolved into my go-to knowledge source, outperforming Reddit & YouTube for most (90%) of my queries. Only turn to traditional platforms when I need visual demonstrations or personal stories/experiences. The accuracy has improved dramatically since 2023, with far fewer AI hallucinations.
#AI #ArtificialIntelligence #LLM
I am going to surmise we will see a big leap with the next generation of models.
The amount of compute that XAI and Meta are using is enormous. This is going create some very powerful systems.
2025 would be a year like no other. All these systems would probably be in place before it ends.
Do you suppose XAi would be more powerful than OpenAi GPT5. Or you believe they'd have their various areas of strength. Knowing the background story.
XAI is going to surpass OpenAi. The level of LLMs will mirror each other. The only question is how quickly they bring them out.
By training using more compute, they can get things to market quicker. This is where XAI and Meta have an advantage against some of the others. The wildcard in that is Google since they use their own chips.
makes sense. the faster you are , the faster you'd be noticed.
With Googles resources, funding would also not be a problem and therefore they may try to take advantage as much as they can to remain the top dogs.
How to Make AI Work for You, and Why It Won't Replace Software Engineering
At Gartner's annual expo, analysts offer a deeper dive into how businesses should approach AI, from when to avoid gen AI and how to scale for a future dominated by the technology.
Not surprisingly, AI was a major theme at Gartner's annual Symposium/IT Expo in Orlando last week, with the keynote explaining why companies should focus on value and move to AI at their own pace. But I was more interested in some of the smaller sessions where they focused on more concrete examples, from when not to use generative AI to how to scale and govern the technology to the future of AI. Here are some of the things I found most interesting.
#ai #software #technology #engineering
"AI does not revolve around gen AI, although it might feel like it right now," Gartner Fellow Rita Sallam said in a presentation entitled "When Not to Use Generative AI." She noted that while boards may now be asking technology leaders to use generative AI, in reality many organizations have used AI of different kinds for many years, in things such as supply chain optimization, sales forecasting, and fraud detection.
Sallam shared data from a recent survey that showed that gen AI is already the most popular technique that organizations are using in adopting AI solutions, followed by machine learning with things like regression techniques.
She stressed that generative AI is very useful for the right use cases, but not for everything. She said it was very good at content generation, knowledge discovery, and conversational user interfaces; but has weaknesses with reliability, hallucinations, and a lack of reasoning. Generative AI is probabilistic, not deterministic, she noted, and said it was at the "peak of inflated expectations" in Gartner's hype cycle.
She warned that organizations that solely focus on gen AI increase the risk of failure in their AI projects and may miss out on many opportunities.
Gen AI is not a good fit for planning and optimization, prediction and forecasting, decision intelligence, and autonomous systems, Sallam said. In each of these categories, she listed examples, explained why gen AI fails in those areas, and suggested alternative techniques.
Article
Can you build a startup without sacrificing your mental health? Bonobos founder Andy Dunn thinks so
Bonobos founder Andy Dunn is back in the builder's seat, working on an in-person social media platform called Pie.
Bonobos founder Andy Dunn is back in the builder’s seat, working on an in-person social media platform called Pie. But the biggest lessons he learned from his $310 million Bonobos exit don’t have as much to do with entrepreneurship as they do with staying sane.
#startup #bonobos #founder #business #andydunn
When Dunn was in college, he was diagnosed with bipolar disorder, but he didn’t get adequate treatment until 2016, when he was hospitalized during a manic episode for the second time.
“The manic state is just a disaster — that’s like being in psychosis, you know, messianic delusions. … You can’t accomplish anything in that state,” Dunn said onstage at TechCrunch Disrupt 2024. The incident was enough of a wakeup call that 16 years after his initial diagnosis, he finally took his condition seriously and started going to therapy, taking medication, and monitoring his sleep.
Dunn wrote a book called “Burn Rate: Launching a Startup and Losing My Mind,” documenting the parallel processes of building Bonobos and figuring out how to accept and then manage his bipolar disorder. But the lessons from the book are applicable for entrepreneurs beyond those with Dunn’s diagnosis.
“We all have mental health, right? It doesn’t take a diagnosis to suffer or struggle,” he said.
Still, entrepreneurs tend to report a higher incidence of mental health issues throughout their lives than the average person.
Article
MIT debuts a large language model-inspired method for teaching robots new skills
MIT this week showcased a new model for training robots. Rather than the standard set of focused data used to teach robots new tasks, the method goes big, mimicking the massive troves of information used to train large language models (LLMs).
#mit #llm #robots #technology #ai
The researchers note that imitation learning — in which the agent learns by following an individual performing a task — can fail when small challenges are introduced. These could be things like lighting, a different setting, or new obstacles. In those scenarios, the robots simply don’t have enough data to draw upon in order to adapt.
The team looked to models like GPT-4 for a kind of brute force data approach to problem solving.
“In the language domain, the data are all just sentences,” says Lirui Wang, the new paper’s lead author. “In robotics, given all the heterogeneity in the data, if you want to pretrain in a similar manner, we need a different architecture.”
The team introduced a new architecture called Heterogeneous Pretrained Transformers (HPT), which pulls together information from different sensors and different environments. A transformer was then used to pull together the data into training models. The larger the transformer, the better the output.
Users then input the robot design, configuration, and the job they want done.
“Our dream is to have a universal robot brain that you could download and use for your robot without any training at all,” CMU associate professor David Held said of the research. “While we are just in the early stages, we are going to keep pushing hard and hope scaling leads to a breakthrough in robotic policies, like it did with large language models.”
Article
Hacked TP-Link routers at center of massive botnet used to attack Azure customers
The network is probably still operational and likely acquiring new infrastructure
In context: CovertNetwork-1658 is a stark reminder of the ongoing cat-and-mouse game between cybersecurity professionals and threat actors. Using compromised devices as part of a botnet also underscores the importance of securing IoT devices and regularly updating firmware on routers and other network equipment.
#hacked #azure #microsoft #botnet #tplink #technology
Microsoft has exposed a complex network of compromised devices that Chinese hackers are using to launch highly evasive password spray attacks against Microsoft Azure customers. This network, dubbed CovertNetwork-1658 by Microsoft, has been actively stealing credentials from multiple Microsoft customers since August 2023.
The attacks use a botnet of thousands of small office and home office (SOHO) routers, cameras, and other Internet-connected devices. At its peak, there were more than 16,000 devices in the botnet, most of which were TP-Link routers.
CovertNetwork-1658, also known as xlogin and Quad7 (7777), is believed to be established and maintained by a threat actor located in China. The network's name stems from Microsoft's classification system, where "CovertNetwork" refers to a collection of egress IPs consisting of compromised or leased devices that may be used by one or more threat actors.
The hackers exploit a vulnerability in the routers to gain remote code execution capability, although the specific exploit method is still under investigation. Once access is achieved, the threat actors take several steps to prepare the router for password spray operations. These steps include downloading Telnet and xlogin backdoor binaries from a remote File Transfer Protocol (FTP) server, starting an access-controlled command shell on TCP port 7777, and setting up a SOCKS5 server on TCP port 11288.
Article
TP-Link launches $600 Archer GE800, its first Wi-Fi 7 gaming router
The Wi-Fi 7 router to go for if visible antennas are not your thing
Highly anticipated: Wi-Fi 7 routers don't come cheap right now, but the advantages could make them worth it for power users. The obvious key benefits include increased speed and bandwidth, thanks to wider 320MHz channels and the ability to combine multiple bands simultaneously using multi-link operation. They also maintain low-latency performance, making them perfect for cloud gaming and AR/VR applications.
#wifi #tplink #technology #gaming #router #technology
Announced way back in 2022 and teased at CES over the past two years, TP-Link's Archer GE800 tri-band gaming router is now available for $599.99, with a $100 discount for early adopters. TP-Link claims the Archer GE800 can deliver 11,520 Mbps on the 6 GHz band, 5,760 Mbps on the 5 GHz band, and 1,376 Mbps on the 2.4 GHz band.
Of course, you'll need Wi-Fi 7-enabled devices to take advantage of the router's advanced features. Only newer models like the Lenovo Legion 7i or Razer Blade 16 laptops offer support for it at the moment. None of the current crop of consoles do.
Starlink shines in FL after Hurricane Milton: ‘A game changer’
“So Starlink has been a game changer,” the CFO of Florida Jimmy Patronis, told Teslarati in a recent interview. Patronis was in a vehicle during our conversation visiting the areas in Florida impacted by Hurricane Milton. He had just left a part of Florida that was ravaged by at least 30 tornadoes during the hurricane.
“I think it had a lot of use. I bought the Starlink. I built a harness on the back of our car and we were traveling down the road with the Starlink. And I was doing TV interviews, going down the highway at 60 miles an hour and the communication was really impressive,” the CFO of Florida added.
#starlink #technology #hurricane #milton #florida #telecommunications
Preparing for Hurricane Milton with Starlink
My interview with Patronis was ironic and interesting, to say the least. While he was in the car traveling to people affected by Milton, I was on the other side of the world, preparing for Typhoon Kristine, the third storm that would hit my home over the last 3-4 months.
I know how important communication lines are during a hurricane. In some cases, it is your only lifeline, and Patronis was well aware of that.
“My own personal experience, I went ahead and decided to buy a Starlink the week before Hurricane Milton made landfall,” Patronis told me.
“When a storm hits, being able to facilitate phone calls, content sharing, interviews, reports, it’s critical. People want to know what is happening. But unless you have dependable communication, it’s very difficult to do that. And we found Starlink to be invaluable when it came to providing those up-to-date communications for our first responders, for our state of operations…” he elaborated.
The internet was our only connection to the rest of the world when the first hurricane hit our house and flooded the streets of our neighborhood and our car. The second time, the water reached about 7 feet high and flooded our car and house. My family in New Jersey were on the phone talking to me as I hurried up the stairs with stuff, trying to beat the rising water rushing into our yard—and eventually into our house.
From my experience, floods are the worst part of hurricanes—and the scariest. I knew we were in trouble when we saw our neighbors asking for rescue through the HOA Facebook group. In the Philippines, people often call for help through social media posts during hurricanes, so the internet is critical.
First responders often use the internet or cellular lines to see if anyone needs help. It was no different during Hurricane Milton in Florida.
“So my office also coordinates all urban search and rescue where our first responders are on the field literally minutes after the storm has made landfall. They depend on Starlink. They will have Starlink out in the field with them. This is how they’re communicating, which homes they have visited, if there’s any need for any other assistance, equipment, help, you name it. If they just need more urban rescue first responders to show up,” Patronis told me.
Article
Why did Wiz turn down $23 billion?
Wiz co-founder and CEO Assaf Rappaport said at TechCrunch Disrupt that turning down Alphabet’s offer was “the toughest decision ever,” but they did so because they believed the cloud security space is a $100 billion opportunity for the company. At the time Wiz turned down the offer, the startup had a private valuation of $12 billion. The CEO admitted that they also had other offers, though none as big as Google’s.
Dropbox is laying off 20% of its workforce.
In a letter to staff, CEO Drew Houston said the cloud company is undergoing a “transitional period” and that the goal was to make cuts in areas where Dropbox has “over-invested” while designing a “flatter, more efficient” team structure. The reduction in headcount will affect 528 employees.
#dropbox #layoffs #technology #cloud #jobs
Big Tech doubles down on AI, $200 billion gamble raises concerns on Wall Street
ROI? Who said anything about ROI?
Cutting corners: Tech executives say there are long-term benefits to their AI investments, drawing parallels to the early days of cloud technology. However, Silicon Valley's spend-first, profit-later attitude tests many investors' patience. Amazon, Microsoft, Meta, and Alphabet have invested significant sums in AI infrastructure – money that has not yielded justifiable returns at this point – to the displeasure of Wall Street. And yet Big Tech is not to be deterred.
#bigtech #technology #ai #investment #wallstreet
Tech giants are set to spend a staggering $200 billion on AI-related capital expenditures this year, according to Bloomberg's calculations, marking an all-time high for these companies. It is an unprecedented level of investment, ranging from securing scarce high-end chips and constructing expansive data centers to forging deals with energy providers and even reviving a controversial nuclear plant for power.
Amazon is leading the charge with a projected record $75 billion in spending for 2024, as CEO Andy Jassy describes AI as a "once-in-a-lifetime opportunity." Analysts at MoffettNathanson called the sum "truly staggering."
Meta is not far behind, with capital spending potentially reaching up to $40 billion in 2024, while CEO Mark Zuckerberg commits to increased investment in AI language models and futuristic projects.
Alphabet has reported higher-than-expected capital expenditures and is projecting "substantial" increases in spending for 2025.
Microsoft's AI-related expenses are also soaring, having spent $14.9 billion in a single quarter, a 50 percent increase from the previous year. Meanwhile, it faces challenges in meeting demand due to data center capacity constraints.
Apple, while not as aggressive in its AI spending, has introduced "Apple Intelligence," a suite of AI-enhanced services, though these new AI products have not significantly impacted its financial results.
Article
CoffeeSpace is a Hinge-like app that wants to help you find your co-founder
The startup has launched a social networking app that matches people exploring startup ideas and looking for co-founders.
CoffeeSpace is on a mission to help people find partners for their startup ideas online. The startup, which exhibited as part of the Startup Battlefield 200 at TechCrunch Disrupt 2024, has launched a social networking app that matches people exploring startup ideas and looking for co-founders.
#coffeespace #application #technology #cofounder
The startup’s algorithm only matches candidates who meet each other’s requirements. You can filter through potential candidates based on several different filters, including expertise, location, industry, and more.
CoffeeSpace CEO Hazim Mohamad told TechCrunch he believes that when you look for a co-founder for your business idea, you want to go beyond a traditional résumé, which is why the app gives users a peek into others’ personalities and working styles in order to help people get an idea of whether a match would be an ideal candidate.
“We’re helping people realize their dreams of exploring their startup ideas,” Mohamed said. “We believe CoffeeSpace will change the nature of how people can find business partners, just like what Tinder did for online dating 10 years ago.”
Article
Maybe this is the place Hive/Inleo should go to attract the right people.
Is ot only US or?
Intel says integrating RAM into Lunar Lake SoC was a mistake, might abandon desktop GPUs again
Panther Lake and Nova Lake laptops will return to traditional RAM sticks
In brief: Intel Lunar Lake mobile CPUs controversially integrated system memory onto the SoC, preventing users from installing additional RAM. The company now partially blames the decision for its latest disastrous earnings report. Furthermore, Intel's restructuring plans might involve downsizing or ceasing development Arc desktop GPUs.
#cpu #intel #desktop #lunarlake #technology #gpu #semiconductor
During the company's third-quarter 2024 earnings call, Intel confirmed that its future laptop chips will return to the traditional use of RAM sticks, reversing Lunar Lake's radical shift to baked-in system memory. The company is also reassessing GPU development over the coming years.
The recently introduced Lunar Lake Core Ultra 200V light mobile CPUs come with either 16GB or 32GB of memory built into the package. Intel claims that forgoing SODIMM sticks reduces power consumption during memory transfers by 40 percent, but users are stuck with the RAM that ships with the device.
When asked whether the company would continue using this design, CEO Pat Gelsinger confirmed that future architectures such as Panther Lake and Nova Lake won't use on-package memory. Moreover, Gelsinger described Lunar Lake as a one-off project initially intended for a niche market before it grew in response to the emerging AI PC market.
Intel CFO David Zinsner admitted that integrating system memory significantly impacted Lunar Lake's gross margins. The comments were made during an earnings call following the company's Q3 2024 financial report, which showed a $16.6 billion net loss – 10 times the previous quarter's loss of $1.6 billion. Analysts expect Intel to post billions in losses for 2024 overall, Chipzilla's first annual net loss since 1986.
Article
Tesla Cybertrucks have started arriving in Canada ahead of deliveries
Tesla Cybertrucks have started arriving in Canada just a few weeks after the company officially launched sales in the country, and as the company prepares to begin initial deliveries of the unique electric vehicle (EV).
This week, multiple sources have reported spotting some of the first Tesla Cybertruck units being prepped for delivery in Canada, and one user in the Tesla Motors subreddit has shared photos of the EVs parked at a store in Kitchener, Ontario. Others also said they spotted the vehicle on the road in British Columbia and Quebec, and Tesla has officially launched Cybertruck accessories in the country, just as it said would happen in the coming weeks.
#cybertruck #canada #pickup #tesla #automotive
I am very curious when we will see them in Europe. At this time the European safety restrictions don’t allow them over here…
Is Canada as hostile against Elon as the US or do they have a more balanced view of him?
BBC: DNA-testing site 23andMe fights for survival
https://www.bbc.com/news/articles/c4gm08nlxr3o
BBC: Daylight savings: What happens to baby sleep when the clocks change
https://www.bbc.com/future/article/20241101-daylight-savings-what-happens-to-baby-sleep-when-the-clocks-change
BBC: The 'bias machine': How Google tells you what you want to hear
https://www.bbc.com/future/article/20241031-how-google-tells-you-what-you-want-to-hear
BBC: Monkeys will never type Shakespeare, study finds
https://www.bbc.com/news/articles/c748kmvwyv9o
Reuters: Bain aiming for Fuji Soft tender offer in early Nov, awaits company approval
https://www.reuters.com/markets/deals/bain-aiming-fuji-soft-tender-offer-early-nov-awaits-company-approval-2024-11-01/
Quantum computing is a revolutionary technology that has the potential to solve complex problems that are currently unsolvable with classical computers. In this article, we will explore the basics of quantum computing, its benefits, and examples of companies that are investing heavily in its development.
What is Quantum Computing?
Classical computers use bits to store and process information. A bit can have a value of either 0 or 1, and these values are used to perform calculations and operations. In contrast, quantum computers use quantum bits or qubits. Qubits are unique because they can exist in multiple states simultaneously, which is known as superposition. This means that a qubit can be both 0 and 1 at the same time, allowing quantum computers to process a vast number of possibilities simultaneously.
Another fundamental property of qubits is entanglement. When two or more qubits are entangled, their properties are connected in such a way that the state of one qubit is dependent on the state of the other qubits, even if they are separated by large distances. This property enables quantum computers to perform calculations on vast amounts of data in parallel, which is essential for solving complex problems.
How Does Quantum Computing Work?
Quantum computing is based on the principles of quantum mechanics, which are different from the classical laws of physics that govern the behavior of classical computers. In a classical computer, the qubits are typically cooled to very low temperatures to reduce thermal noise, which can disrupt the fragile quantum states. This is why quantum computers require cryogenic cooling, which is expensive and challenging.
s.
Benefits of Quantum Computing
Quantum computing has the potential to solve a wide range of complex problems that are currently unsolvable with classical computers. Some examples of problems that quantum computers can solve include:
Companies Working on Quantum Computing
Many companies are investing heavily in the development of quantum computing. Some examples of companies that are working on quantum computing include:
Challenges and Limitations
While quantum computing has the potential to solve complex problems, it is still in its early stages of development. Some of the challenges and limitations of quantum computing include:
Conclusion
Quantum computing is a revolutionary technology that has the potential to solve complex problems that are currently unsolvable with classical computers. While it is still in its early stages of development, many companies are investing heavily in its development. Some of the benefits of quantum computing include optimization, cryptography, materials science, and machine learning. However, challenges and limitations such as error correction, scalability, quantum noise, and quantum control need to be addressed before quantum computing can be used in practical applications.
Quantum computers use a variety of quantum algorithms to solve problems. These algorithms are designed to take advantage of the unique properties of qubits, such as superposition and entanglement. Some common quantum algorithms include the Shor algorithm, the Grover algorithm, and the Simons algorithm.
The Shor algorithm is a quantum algorithm that can factor large numbers exponentially faster than any known classical algorithm. This has significant implications for cryptography, as it would allow for the breaking of many encryption algorithms currently in use.
Yeah its a big thing in the security business. Much of the current setup could be obselete with quantum.