You are viewing a single comment's thread from:

RE: LeoThread 2025-04-17 10:43

in LeoFinance6 months ago

Here is the daily technology #threadcast for 4/17/24. 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.

Sort:  

Adobe Bets Big on AI Video as Synthesia Hits $100M Milestone

synthesia just landed funding from adobe ventures after crossing $100M in yearly revenue—without turning a profit yet. the ai video startup is now powering internal videos for 70% of the fortune 100, turning bland corporate updates into polished productions. it's neck-and-neck with rival heygen as both sprint to dominate the enterprise comms space. think of it like the netflix of corporate training videos, but still figuring out its business model.

#ai #startup #enterprise #futureofwork #technology

> S👁️URCE <

Lab-Grown Chicken Gets Real: A Big Leap for Cultured Meat

Researchers just grew nugget-sized chicken using a fiber system that mimics blood vessels—basically, a fake circulatory system feeding cells in a gel. Why’s that huge? It brings us closer to lab-grown steaks, fish fillets, even pork chops—without the farm. Think of it like growing a tree from the roots up, not just sprinkling leaves. The same tech might even lead to lab-grown organs one day.

#labgrownmeat #foodtech #sustainableeating #futureofmeat #technology

> S👁️URCE <

Elon Musk Bets Big on Robotaxi, Ignores Tesla's Own Warnings

tesla’s own execs warned the robotaxi project would burn cash, not print it. they urged a pivot to a $25k “model 2” EV to compete with cheaper rivals. instead, musk pulled the plug on that plan and doubled down on the flashy cybercab. now, with no budget-friendly option in sight, tesla’s stuck while competitors speed past. imagine skipping a reliable car to chase a concept limo—it’s bold, but risky.

#elonmusk #tesla #robotaxi #evs #technology

> S👁️URCE <

OpenAI’s New Models Can Think, Browse, and Code at Once

openai just dropped two new models, o3 and o4-mini, and they’re the first that can reason and use every chatgpt tool—like browsing the web or writing code—at the same time. think of it like giving your smartest friend internet access, a calculator, and a notepad all at once. they’re already live for Plus, Pro, and Team users. edu and enterprise folks get access next week, with o3-pro hitting Pro plans soon.

#openai #chatgpt #aiupdate #futureofwork #technology

> S👁️URCE <

Answer Engine Optimization Is Changing the Game

forget chasing single keywords—AEO is about answering whole clusters of related questions, like solving a puzzle instead of hunting one piece. it’s smarter, faster, and built for how we actually search now, especially with ai chat interfaces serving up answers (and traffic). the winners? brands who zero in on product questions, fill content gaps, and get cited by large language models. think less busywork, more results—like trimming the fat and keeping only what feeds growth.

#seo #aiwriting #contentstrategy #digitalmarketing #technology

> S👁️URCE <

THE FUTURE OF CHIPS: HEAT IS THE NEW CHALLENGE

Engineers have doubled the number of transistors in chips every two years, but with more power comes more heat. Excessive heat affects performance, power usage, and energy efficiency, potentially damaging chips long-term. As circuits get more compact and powerful, heat dissipation becomes crucial to keeping chips running smoothly. Managing this growing challenge will be key to the future of tech.

#chips #technology #innovation #hardware

> S👁️URCE <

NVIDIA's New Tech Will Change Video Editing Forever

NVIDIA's EquivDM is revolutionizing video editing. It makes video diffusion smoother by using consistent noise, which improves motion tracking and gives 3D outputs that stay consistent. With fewer sampling steps, you get better results faster. Think of it as the difference between a shaky, blurry video and a seamless, professional one. This tech could save hours in editing while improving quality!

#nvidia #videodiffusion #motiontracking #3dtechnology #technology

> S👁️URCE <

M1: The Next Step in Test-Time Computing

M1 is a new Mamba reasoning model, trained to handle extended test-time computation. While it doesn’t quite match the cutting-edge models, it’s making a strong case for itself, especially in areas like long context processing and throughput. Imagine it as a solid, reliable performer, even if it’s not quite at the top tier yet. It’s a big step forward for scalable test-time compute.

#machinelearning #AI #technology #innovation

> S👁️URCE <

Hugging Face Expands HELMET Benchmark for Better LLM Insights

Hugging Face just leveled up its HELMET benchmark, adding more models and insights for testing long-context LLMs like Phi-4 and Jamba 1.6. This upgrade helps researchers dive deeper into how these models handle extended information—crucial for everything from advanced AI tasks to real-time applications. Think of it as giving researchers better tools to understand and optimize LLMs for complex challenges. Big move for AI research!

#huggingface #llm #machinelearning #technology

> S👁️URCE <

Google Fights Scam Ads with AI Power

In its 2024 Ads Safety Report, Google revealed its AI tech blocked billions of scam ads, taking down 700K+ malicious accounts. With major upgrades to its language models, Google has drastically reduced impersonation scams, protecting users and businesses alike. Think of it as a virtual bouncer, catching the bad guys before they can cause damage. This move is a huge step in making the internet safer.

#google #scams #adsafety #technology

> S👁️URCE <

Stable Diffusion Just Got a Speed Boost on AMD GPUs

Stability AI and AMD teamed up to turbocharge Stable Diffusion models for Radeon GPUs and Ryzen AI. The optimization means faster image generation, making AMD users’ workflows smoother than ever. If you’ve been using AMD hardware, you’ll notice quicker rendering times—ideal for creative projects where every second counts. Imagine cutting your wait time in half while working on a new design or artwork.

#amd #stablediffusion #ai #technology

> S👁️URCE <

A IA da rede social X será capaz de se lembrar de conversas anteriores

Por enquanto, a funcionalidade está em fase beta e disponível apenas para dispositivos Android e iOS

#technology #ai #x #hivebr

A ferramenta de inteligência artificial (IA) da rede social X, de Elon Musk, conhecida como Grok, ganhou uma nova funcionalidade.

O anúncio foi feito nesta quinta-feira (17) pela conta oficial da plataforma, que destacou: “Agora o Grok é capaz de lembrar suas conversas”. A publicação também incluiu um vídeo demonstrando como o recurso funciona.

Segundo a rede social, a atualização permitirá que a IA ofereça respostas mais personalizadas quando o usuário pedir conselhos ou sugestões, com base em interações anteriores.

Por enquanto, a funcionalidade está em fase beta e disponível apenas para dispositivos Android e iOS.

No entanto, não pode ser acessada por usuários da União Europeia e do Reino Unido, possivelmente por questões regulatórias relacionadas à proteção de dados.

Especialistas cobram monitoramento contra conteúdo nocivo a crianças nas redes

As tragédias levantam discussão sobre regulação das plataformas, o uso de redes sociais pelos mais novos e a responsabilidade da família.

#TECHNOLOGY

Para especialistas ouvidos pela reportagem, uma das principais dificuldades é limitar a presença de menores de 13 anos nas redes sociais

Na maioria das plataformas, usuários precisam apenas autodeclarar suas idades, o que torna mais fácil o acesso a esse tipo de conteúdo nocivo.

O Ministério da Justiça e Segurança Pública afirma que desenvolve uma ferramenta de verificação etária para limitar o acesso de crianças e adolescentes a conteúdos que não condizem com a indicação da sua faixa etária.

Ainda sem modelo definido, a proposta prevê o uso de um token anônimo, gerado pelo governo e vinculado aos dados oficiais do cidadão.

O código seria inserido em aplicativos de celular para comprovar a faixa etária do usuário, sem violar a privacidade.

Big techs compraram uma empresa a cada 11 dias, entre 2019 e 2024, mostra relatório da entidade holandesa Somo (Centro de Pesquisa sobre Corporações Multinacionais) divulgado nesta terça-feira (15).

O levantamento considerou as operações de Amazon, Alphabet (dona do Google), Apple, Meta e Microsoft.

De acordo com o grupo, a prática visa expandir a dominância dos gigantes da tecnologia sobre áreas estratégicas, como inteligência artificial, robótica e computação na nuvem.

Paragraph

Instagram, WhatsApp at legal crossroads
Facebook acquired Instagram when the photo-sharing app had just 13 employees. At the time, Zuckerberg was preparing to take Facebook public in what would become the largest tech IPO on record.

At the time of their acquisitions, Instagram had about 30 million users and was bought for $1 billion, while WhatsApp, with around 450 million users, was acquired for a whopping $19 billion. Both have since outpaced Facebook’s growth, according to The New York Times.

Facebook’s acquisitions of both platforms are central to the antitrust trial that began Monday and could last for weeks.

The federal regulatory agency has accused the tech company of holding a monopoly over the social networking market, arguing that it should never have been allowed to acquire Instagram and WhatsApp.

Political pressure surfaces again
Zuckerberg’s legal clash with the FTC was initiated during Donald Trump’s First term as US president.

As reported by The Wall Street Journal, the Meta boss had directly approached Trump, asking him to press the FTC to drop the antitrust case against the tech giant.

Previously tense, the relationship between Trump and Zuckerberg seems to have improved since the fallout from Trump’s ban on Meta platforms following the January 6 Capitol riots.

The case, overseen by U.S. District Judge James Boasberg, could last up to two months.

“This research paves the way for user-friendly visual assistance systems, offering alternative avenues to enhance the quality of life for people with visual impairment,” the team wrote.

The role of simple audio and haptic cues
Lead researcher Gu Leilei, an associate professor at Shanghai Jiao Tong University, emphasised the team’s commitment to making the system as practical and easy to use as possible.

“This system can partially replace the eyes,” he said.

“Lengthy audio descriptions of the environment can overwhelm and tire users, making them reluctant to use such systems,” Gu told The South China Morning Post.

The extent of object recognition
Researchers are planning to expand these capabilities further. Gu said the team’s next focus is to refine the system for outdoor environments, where navigational challenges are far more complex. Enhancements could include improved object detection, dynamic route adaptation, and integration with real-world GPS systems.

This AI-powered wearable system may offer a new level of autonomy with further developments, and with that, confidence to millions of visually impaired people to be independent worldwide.

The robot’s facial recognition capabilities enable it to recognize people who have been marked as high-risk or wanted. Its technology can identify and locate people in busy event locations by analyzing a variety of visual clues, such as gender, body shape, clothing, and face features. Additionally, the robot is taught to recognize offensive activities, including fighting, physical altercations, and other disturbances.

Together with behavioral analysis, the robot can recognize possible weapons like knives and blunt objects like wooden sticks while deftly ruling out innocuous devices like water cannons, which are frequently used during Songkran celebrations.

RTP claims that the cutting-edge patrol system is a major advancement in Thailand’s use of AI in public safety operations.

The deployment of these humanoid robots is part of China’s broader strategy to integrate robotics into public safety, emergency services, and law enforcement. This initiative also includes the testing of other advanced robotic systems. In the same month, Chinese authorities trialed the RT-G robot, an autonomous spherical unit created by Logon Technology.

Built for tough conditions, the RT-G can traverse both land and water at speeds of up to 22 mph (35 km/h) and endure impacts of up to 4 tons (8,818 pounds), making it ideal for use in extreme environments.

Together, these developments signal countries’ growing investment in robotic assistance for public safety and policing.

The successful submission and acceptance of the construction permit application (CPA) to the NRC is also an important milestone, and so is the over a years’ worth of review with the NRC on the company’s CPA and Topical Report Submittals, with the NRC recently announcing they are ahead of schedule on the review.

The company had also recently been awarded a state-level construction permit from the State of Wyoming; where the first Natrium project is being built.

The bid by Terrapower can be seen as a challenge to Rolls-Royce which is hoping to get its mini-nuclear reactors accepted for UK’s nuclear energy resurgence.

Alongside these two, GE Hitachi, Holtec and Westinghouse have also been participating in the tender approved by the UK government, as per a report by Telegraph. The winner of the tender process is slated to be announced by mid-2025.

According to the UK Ministry of Defence (MOD), the successful trial comes as drone swarms are increasingly seen in use in frontline combat in Ukraine.

UK Defence Intelligence estimates that last year, Ukraine had to defend against attacks from more than 18,000 drones.

Frying 100 drones
In this test, the British Army successfully killed two swarms of drones in a single engagement.

The project saw more than 100 drones being tracked, engaged, and defeated using the weapon across all trials.

Team Hersa, a collaboration between Defence Equipment & Support and the Defence Science and Technology Laboratory, has delivered the project.

The RFDEW demonstrator has been developed by an industry consortium led by Thales UK.

Zuckerberg weighed spinning off Instagram in 2018 amid antitrust pressure, email shows

On the second day of his testimony, Meta CEO was shown an email, in which he weighed the option of spinning off Instagram.

Meta CEO Mark Zuckerberg weighed spinning off Instagram in 2018 amid growing antitrust concerns against Facebook, according to an internal email revealed Tuesday in a Washington, D.C. courtroom.

On the second day of his testimony in Meta’s antitrust trial with the U.S. Federal Trade Commission (FTC), Zuckerberg was shown an email, in which he considered the option of spinning off Instagram, the app his company bought for $1 billion in 2012.

“And I’m beginning to wonder whether spinning Instagram out is the only structure that will accomplish a number of important goals,” the 40-year-old had written in the email from May 2018.

“As calls to break up the big tech companies grow, there is a non-trivial chance that we will be forced to spin out Instagram and perhaps WhatsApp in the next 5-10 years anyway. This is one more factor we should consider,” he further stated.

Breakthrough: China unveils AI eyes for the blind to move independently and safely

Visually-impaired people can move more smoothly and freely with this AI system.

Imagine a world where people who cannot see can still move around confidently without any self-doubt or fear. That world isn’t far away with the arrival of AI.

According to a study published in Nature Machine Intelligence, researchers from China unveiled a new wearable AI system that empowers blind and visually impaired individuals to navigate easily and independently. The system provides real-time guidance to users with a combination of video, vibrations, and audio prompts.

How can AI help the blind see
The AI system includes a camera, an AI processor, and bone conduction headphones. The system is mounted between the user’s eyebrows. The camera captures the live footage instantly for the AI system to analyze. Then, short audio cues are delivered directly through the headphones, without blocking ambient sounds.

A team from Shanghai Jiao Tong University developed the system. Shanghai Artificial Intelligence Laboratory, East China Normal University, Hong Kong University of Science and Technology, and the State Key Laboratory of Medical Neurobiology at Fudan University.

Cyborg 1.0: Thai Robocop patrols streets with 360° eyes, live face-tracking power

AI Cyborg 1.0 analyzes CCTV and drone data in real time with 360° cameras linked to a command center for rapid response.

Thailand has unveiled its first AI-powered police robot, deployed during the Songkran festival in Nakhon Pathom province to boost public safety.

The Royal Thai Police (RTP) introduced the robot on April 16 via a Facebook post.

Named “Pol Col Nakhonpathom Plod Phai,” meaning “Nakhon Pathom is safe,” the robot was stationed at the Tonson Road festival venue in Muang district.

In March, China introduced humanoid robots in Shenzhen to patrol with police, interact with pedestrians, respond to voice commands, and wear high-visibility vests, drawing public attention.

Surveillance gets smarter
The robot, named “AI Police Cyborg 1.0,” made its debut during the Songkran festival in Nakhon Pathom province. Developed collaboratively by Provincial Police Region 7, Nakhon Pathom Provincial Police, and Nakhon Pathom Municipality, this Robocop-style unit is equipped with advanced surveillance and threat detection technologies.

AI Police Cyborg 1.0 uses onboard AI to immediately process and analyze data by integrating real-time data from aerial drone footage and local CCTV networks. Rapid reaction coordination is made possible by the robot’s in-built 360-degree smart cameras, which are immediately connected to the province’s Command and Control Center and backed by video analytics software, according to The Nation.

Bill Gates vs Rolls-Royce? 345MW Terrapower nuclear reactor to shake up UK energy race

Terrapower has submitted a letter to the UK’s Department for Energy Security and Net Zero (DESNZ).

Bill Gates-backed Terrapower has decided to give a challenge to Rolls-Royce in the United Kingdom market for the deployment of nuclear reactors.

Terrapower has submitted a letter to the UK’s Department for Energy Security and Net Zero (DESNZ) which establishes its intention of entering the Generic Design Assessment (GDA) process.

It should be noted that the company’s Natrium reactor is currently being developed in the United States.

This is the first step taken by Terrapower with the intent to deploy its Natrium nuclear reactor in the international market.

“I am incredibly excited to begin of the process of licensing the Natrium technology in the UK,” said Chris Levesque, TerraPower President and CEO.

“TerraPower is committed to deploying Natrium units globally and has been in active discussions in the UK for years. There is immense interest and opportunity for the United States and United Kingdom to cooperate on deploying advanced nuclear plants over the coming decade.”

Terrapower’s bid for UK’s GDA process
Terrapower says that the UK’s GDA process will build on its successful regulatory endeavors and allow for the company to establish deployment timelines for Natrium sites in the country.

The regulatory milestones with its first Natrium plant, currently being developed in the United States, will be used as the basis of its GDA application. These milestones include the pre-application meetings it held with the US Nuclear Regulatory Commission (NRC).

UK Army’s invisible blast melts 100+ strong drone swarm with just a 13-cent shot

The weapon system can defeat airborne targets at ranges of up to 1km.

The British Army has tested an invisible radio wave weapon, known as the Radiofrequency Directed Energy Weapon (RFDEW), to knock out drone swarms for the first time.

British soldiers have successfully tracked, targeted, and defeated swarms of drones in the latest trial of a new directed energy weapon developed in the UK.

The trial was held at a weapons range in West Wales and was the largest counter-drone swarm exercise the British Army has conducted to date.

13-cent-per-shot radio waves to fry drones
According to a press release by the British Army, the weapon system demonstrator is a type of RFDEW. It has proven capable of neutralizing multiple targets simultaneously with near-instant effects.

The UK Government has invested more than £40 million in RFDEW research and development to date, the British Army said.

The weapon uses high-frequency radio waves to disrupt or damage critical electronic components inside drones, causing them to crash or malfunction.

At an estimated cost of GBP 0.10 (USD 0.13 or 13 cents) per shot fired, if developed into operational service, the weapon could provide a cost-effective complement to traditional missile-based air defense systems, the British Army added.

The weapon system can defeat airborne targets at ranges of up to 1km and is effective against threats that cannot be jammed using electronic warfare.